The Analytics Power Hour · 2026-06-23 · 58 min
Key moments - from our scoring
Substance score
48 / 100
Five dimensions, 20 points each
In this 300th episode milestone, Michael Helbling, Moe Kiss, and Julie Hoyer challenge the prevailing narrative that semantic layers are essential infrastructure for analytics by examining whether AI-driven approaches offer viable alternatives. Guest Jacob Matson, developer advocate at MotherDuck and former senior data leader at Funko and Verimatrix, shares his hard-won perspective from decades in accounting and data systems. He explores the tension between standardization and flexibility: organizations need enough conformity (roughly 80-90%) so stakeholders communicate consistently about metrics like monthly active users, but over-constraining systems kills the differentiated insights that actually move business forward. Using the Jenga metaphor, Matson illustrates how organizations must discover through trial and error which constraints are immovable (core definitions like MAU) and where flexibility matters - such as product-specific MAU interpretations at Canva. The discussion surfaces why semantic layers, OLAP cubes, and ERP templates often fail: they force businesses into rigid boxes rather than accommodating the bespoke logic that drives competitive advantage. Relevant for data leaders, analytics managers, and organizations evaluating BI tooling who wrestle with metric governance and semantic layer ROI.
Organizations need roughly 80-90% standardization so teams can communicate consistently about metrics, but excessive conformity prevents capturing the unique, differentiating business logic that actually moves revenue - the magic happens in the margins and is hard to define in someone else's model.
Monthly active users by product is messy because different people interpret product definitions differently - some might count anyone using video features, while others only count those using advanced video editor features, creating a gray zone that defies a single standardized definition.
Zoom out to understand the company's competitive engine and how it increases cash flow, then take managed risk through trial and error - test your model in ways that won't get you fired if wrong, using mentors and intuition to distinguish risky bets from safe conformity.
CFOs often abdicated the analytics realm entirely to other departments when they couldn't achieve the accuracy standards used in financial reporting, creating separate non-finance analytics functions rather than solving the underlying tension.
Our reviewer’s read on each dimension, with quotes from the episode.
The episode contains a handful of genuinely useful framings - reframing the semantic layer problem as a search problem, the progressive 'jungle paths to highway' analogy for context building, and the idea that LLMs primarily solve a translation-between-mental-models problem. However, these insights are embedded in heavy biographical preamble, extended Jenga and jungle analogies, two ad reads, and a closing 'last call' section covering squirrel emails and book recommendations, dragging the insight-per-minute ratio down considerably.
my current kind of position on this stuff is that we really have a search problem. And if we can find a way to make our metrics searchable, then the way that we define it may be less important.
The reason you buy a semantic layer in the first place is because some number got somewhere and it was wrong and legal's pissed.
The reframe of semantic layers as a search problem and the progressive context-building spectrum (paths to gravel to pavement to highway) are fresh angles not commonly articulated this way. However, the core critique - that semantic layers are inflexible, expensive to maintain, and nobody updates them - is standard industry discourse, and the accounting-disciplines-analytics-thinking narrative is familiar territory. Moderately original rather than genuinely contrarian.
the biggest challenge that LLMs solve is they're really good at translating between languages
a semantic layer is like a highway. Right? It's like, we are just like building the thing straight through the jungle. I would almost say like, we can build it a little more progressively
Jacob Matson brings a genuinely differentiated cross-domain background (public accounting to data analytics) that produces interesting perspective on metrics governance and risk. However, he is currently a Developer Advocate - a marketing and evangelism role - not a senior practitioner operating at scale, and his prior hands-on experience is at mid-market companies. His ideas are thoughtful and exploratory but not battle-tested at enterprise scale, and he frequently acknowledges open questions he cannot yet answer.
I graduated from college, I worked in accounting, in public accounting. I sat for the exams, did all that stuff
we launched our MCP server in December at MoetherDuck. And we had our own existing set of dashboards for the sales team.
A handful of concrete specifics land - OpenAI's double-querying validation approach, MotherDuck's MCP server launch in December, the Kimball book being ~25-30 years old, and the 1 million vs. 65K Excel row limit. However, there are virtually no hard performance metrics, no case study outcomes with measurable results, no dollar figures, and even the MotherDuck internal MCP example is described only anecdotally without quantified improvements.
open AI has like a analytics agent and they basically say, all right, well, we're just going to ask every question twice
I remember it being a big deal when we got the Excel that could handle 1 million rows and not just 65,000 rows
Moe Kiss and Julie Hoyer ask several genuinely sharp questions - explicitly separating 'search' from 'context' as distinct problems, probing the echo-chamber and overfitting risk of building on historical queries, and asking where the legwork migrates in an AI-mediated workflow. These are substantive follow-ups that push the conversation forward. The episode is held back by frequent 'that's such a good question' affirmations, a reluctance to challenge Jacob's more speculative claims, and a host who mostly facilitates without pushing back on assumptions.
Can we separate search from the context? Because I feel like those are two very different things and where you struggle with both.
how do you fight the echo chamber effect too from some of this? Sorry, when you were talking about like using AI to search for queries, like if it's just always bringing back historical data
Computed from the transcript - who did the talking, and the words that came up most.
If you've ever poured months into building a semantic layer only to watch it become shelfware the moment the business pivoted, Jacob Matson has some thoughts. And a metaphor. Your data is a jungle - and a semantic layer is a highway. Great if you need to get somewhere fast and reliably (monthly active users: highway, please). But the interesting business questions? The slicing, the dicing, the nuanced dimensions that actually differentiate your company from its competitors? There's no highway for that. There never will be. Jacob, a developer advocate at MotherDuck with deep roots in accounting and ERP systems, joined Michael, Moe, and Julie to talk through what comes after the semantic layer - or at least alongside it. The conversation covered why the most important parts of any business are precisely the parts that resist being modeled in someone else's framework, why AI is actually pretty good at writing SQL but not so great at remembering what it figured out yesterday, and whether the real job to be done here is less about modeling and more about search. Oh, and the uncomfortable truth that at episode 300, we still don't have a great answer for metric drift.
Transcribed and scored by The B2B Podcast Index.
[Moe Kiss]: Welcome to the Analytics Power Hour. [Announcer]: Analytics topics covered conversationally and sometimes with explicit language. [Michael Helbling]: Hi everybody, welcome to the Analytics Power Hour. [Michael Helbling]: This is episode 300.
[Michael Helbling]: This is Analytics Power Hour. [Michael Helbling]: Okay, sorry, that was just a dumb joke on the movie. [Michael Helbling]: Okay. [Michael Helbling]: Every time you turn around.
[Michael Helbling]: I think you're hearing about the semantic layer. [Michael Helbling]: We even did a show on the topic recently. [Michael Helbling]: It's what AI needs to be successful. [Michael Helbling]: Well, that's at least what we keep hearing from the vendors.
[Michael Helbling]: And well, honestly, that was right around the time we ran across an article that grabbed [Michael Helbling]: our attention. [Michael Helbling]: What if we didn't need semantic layers? [Michael Helbling]: Then we read an article by OpenAI. [Michael Helbling]: They published about how they're using AI to analyze data and took a closer look at a [Michael Helbling]: couple of vendor websites.
[Michael Helbling]: We started seeing the context for AI isn't only a semantic layer thing. [Michael Helbling]: And well, we wanted to talk about that. [Michael Helbling]: So let me introduce my co-hosts, Moe Kiss of Canva. [Michael Helbling]: How you going?
[Michael Helbling]: I'm going great. [Michael Helbling]: Thanks for asking, Michael. [Michael Helbling]: I see you're down a remote on the wall there. [Michael Helbling]: So that's...
[Michael Helbling]: Oh jeez. [Moe Kiss]: I love when we give a visual in-joke that no one else can follow. [Michael Helbling]: Yeah, for an audio podcast. [Michael Helbling]: It's okay.
[Michael Helbling]: We'll make a clip out of it. [Michael Helbling]: Julie Hoyer of Further. [Michael Helbling]: Welcome. [Moe Kiss]: Glad to see you.
[Julie Hoyer]: Hello. [Julie Hoyer]: Hello. [Michael Helbling]: Glad to be here. [Michael Helbling]: Awesome.
[Michael Helbling]: And go Browns. [Michael Helbling]: And I'm Michael Helbling. [Michael Helbling]: So naturally, we reached out to the author of one of those articles. [Michael Helbling]: And I'm excited that he is our guest.
[Michael Helbling]: Jacob Matson is a developer advocate at MotherDuck, the cloud data warehouse built for answers. [Michael Helbling]: He has also held senior data and accounting roles at firms like Simetrics, Funko, and Verimatrix. [Michael Helbling]: And today he is our guest. [Michael Helbling]: Welcome to the show, Jacob.
[Michael Helbling]: Hey, Michael and Julie and Moe. [Michael Helbling]: It's great to be here. [Michael Helbling]: I'm super pumped. [Michael Helbling]: Awesome.
[Michael Helbling]: Well, we're excited to have you. [Michael Helbling]: So I think, Jacob, maybe to kick off the conversation, I think it would be great for us to understand [Michael Helbling]: a little bit more both about your background and exposure to this and sort of what started [Michael Helbling]: formulating for you that led to you kind of digging in and doing research in this area [Michael Helbling]: around sort of semantic layer or not or other alternatives to semantic layers.
[Jacob Matson]: Yeah. [Jacob Matson]: That's such a good question. [Jacob Matson]: I guess I'll start with a little bit of biography that got us here. [Jacob Matson]: I'll try not to be too self-indulgent.
[Jacob Matson]: So when I graduated from college, I worked in accounting, in public accounting. [Jacob Matson]: I sat for the exams, did all that stuff, and worked in public accounting and then had a [Moe Kiss]: company called Verimatrix that was called out there, doing all of the normal accounting [Jacob Matson]: things, climbing the ladder in a very specific kind of governed way, working for people with [Jacob Matson]: titles like Controller or CFO, and eventually taking some moves on myself.
[Jacob Matson]: I think one of the things that we always talked about was especially on the financial side [Jacob Matson]: was like, how do we know what numbers are right for this definition of this thing? [Jacob Matson]: And at the time, the tools we had were much, much worse than we have now. [Jacob Matson]: I remember it being a big deal when we got the Excel that could handle 1 million rows [Jacob Matson]: and not just 65,000 rows. [Jacob Matson]: That's why I think it was Excel 2003, maybe, which of course, actually, what it all did [Jacob Matson]: was train everyone to just have a horrible experience in Excel all the time and just [Jacob Matson]: be totally fine with it.
[Jacob Matson]: It's just too much information, it can't handle it, and then we all dealt with saving [Jacob Matson]: issues and crashing and all these things all the time. [Michael Helbling]: Yeah. [Michael Helbling]: Don't calculate your metrics until you're really ready. [Jacob Matson]: Yeah, exactly.
[Jacob Matson]: Turn that automatic calculation off. [Jacob Matson]: That's right. [Jacob Matson]: Yeah. [Jacob Matson]: Step one.
[Jacob Matson]: And then you can do that as a binary format Excel file. [Jacob Matson]: So did all those things and had the pleasure of working on things like MDX and DAX along [Jacob Matson]: the way, which are both the modeling languages that are built into the Microsoft stack. [Jacob Matson]: And along the way through that journey, really found my way towards using SQL for a lot of [Jacob Matson]: the work I was doing, and that just naturally came out of the data that I had that was too [Jacob Matson]: big for Excel and it was too complicated.
[Jacob Matson]: And there was lots of really, I was just driven 100% on just the business need for solving [Jacob Matson]: these problems and I needed to get more robust tooling and we had SQL server and it had more, [Jacob Matson]: it was running on a server that had more compute than my laptop and all these things. [Jacob Matson]: And so it was very natural to kind of progress up there and so I've done lots of fun things [Jacob Matson]: kind of in that space. [Jacob Matson]: I kind of like the joke that I always worked in data from the beginning of my career.
[Jacob Matson]: It's just my pipelines ran like once a month, right? [Jacob Matson]: It was a month in close process for those of you at home. [Jacob Matson]: And it was just kind of how I got there. [Jacob Matson]: And so you do a lot of things along the way and you see lots of errors along the way too.
[Jacob Matson]: Some material, some not, right? [Jacob Matson]: And I worked on the IT side at a public company and you see lots of interesting things produced [Jacob Matson]: internally that never make their way into the filing documents, for example, sent to [Jacob Matson]: the SEC. [Jacob Matson]: So I think for me, kind of some of the genesis that led to this notion of like, do we need [Jacob Matson]: semantic layers anymore? [Jacob Matson]: It was like two things, A, working in accounting for a long time and like understanding the [Jacob Matson]: quality that goes into those numbers, which is very high, but also not as high as you'd [Jacob Matson]: like.
[Jacob Matson]: That's what I would say. [Jacob Matson]: And the second part of that is that like we often, at least what I would see on accounting, [Jacob Matson]: the accounting side was like, we would get too precious about the exact accuracy and [Jacob Matson]: precision of a number instead of instead of actually moving the business forward, right? [Jacob Matson]: And so what actually happened in my career, at least, is that like it seemed like I, you [Jacob Matson]: know, I wasn't that close to it early in my career, but like what it felt like is that [Jacob Matson]: like CFOs in particular, completely abdicated the realm of analytics to like some other domain.
[Jacob Matson]: Like, you know what, we can't get accurate enough, you know, it's not good enough for [Jacob Matson]: whatever reporting we're building, we're just going to let some other part of the org like [Jacob Matson]: handle that. [Jacob Matson]: In fact, I remember even seeing like job descriptions that were like, director of analytics, non-finance, [Jacob Matson]: like type of roles. [Jacob Matson]: I experienced that when I was at a company room really fast and was trying to build all [Jacob Matson]: of these things.
[Jacob Matson]: And I built, you know, our first day to warehouse from very much, you know, accounting principles [Jacob Matson]: first, and eventually just got to the point where we had to break that apart because it [Jacob Matson]: was just, we didn't have the right primitives to answer the questions in a way that we were, [Jacob Matson]: we were comfortable with. [Jacob Matson]: And we tried all the things and it was just really hard to manage and to update. [Jacob Matson]: So a little bit of this idea is me manifesting like, what if I just took away all the pain [Jacob Matson]: that I experienced when I was like building these analytics cubes back in the day?
[Jacob Matson]: Like what if we could just like ask AI those questions and it would like reformulate those [Jacob Matson]: on the fly? [Jacob Matson]: And so that was kind of like what led me to exploring the idea and then beginning to [Jacob Matson]: do like research around it. [Moe Kiss]: Can you tell me a little bit about the tension that you just, you kind of touched on, but [Moe Kiss]: we didn't go date the tension. [Moe Kiss]: So I mean, we talked about this with Cindy a little while back about semantic layers [Moe Kiss]: and just like, it's been sold at the moment is like the holy grail as every new, not actually [Moe Kiss]: new idea in analytics is that all will solve all of our problems.
[Moe Kiss]: But that like, that core tension that you touched on on how hard it is to maintain the [Moe Kiss]: like inflexibility perhaps that then makes you not able to answer your business questions [Moe Kiss]: like what were some of your lived experiences? [Jacob Matson]: Yeah, this is such a good question. [Jacob Matson]: I mean, I think like the first one was, I think we, so I was working on an ERP system [Jacob Matson]: and we were like, hey, we want to implement like better reporting analytics on it.
[Jacob Matson]: We're going to buy this software package that will just like automatically build out all [Jacob Matson]: the OLAP cubes for us and then we can like tie them into our BI tool. [Jacob Matson]: I think this was even like pre-Tablo maybe and well, Tableau existed, but it didn't exist [Jacob Matson]: for the company I worked at, I think that's what I would say, it was certainly not something [Jacob Matson]: I was tracking at the time, you know, we bought the software and then I was like, okay, now [Jacob Matson]: let's like implement it.
[Jacob Matson]: And it was like, it had to fit in a very narrow box for us to actually take advantage of it. [Jacob Matson]: And that was a pattern I saw repeated a lot kind of in the ERP space too, which was like, [Jacob Matson]: hey, just like, you know, make your business fit into this template of how we run our systems. [Jacob Matson]: And then you get all of these awesome synergies or whatever, right? [Jacob Matson]: Like now you don't need people to do your buying.
[Jacob Matson]: You just like run this report and it tells you what to buy. [Jacob Matson]: And so what I kind of came to believe, I think like, I definitely wanted to say, all right, [Jacob Matson]: let's just like apply this system, let's apply these SAP primitives that are, you know, very [Jacob Matson]: old and well tested, let's just apply this like blindly to our business. [Jacob Matson]: Like why are we over complicating it? [Jacob Matson]: Like our business is not this hard.
[Moe Kiss]: But then what I kind of discovered is that the interesting parts of your business are [Jacob Matson]: really hard to define in someone else's model, right? [Jacob Matson]: They end up being, unless you're like a pure commodities trader, like the magic happens [Jacob Matson]: kind of in the margins. [Jacob Matson]: And so like defining those systematically is super hard. [Jacob Matson]: I really struggled with that.
[Jacob Matson]: And so like when I kind of realized that that's or like that was the mental model that I was [Jacob Matson]: bringing to these problems, I started being like, hang on, how do I design this ERP system [Jacob Matson]: that we're working on that was my accountability? [Jacob Matson]: How do I make it so that like we can do the thing that we need to do, you know, to make [Jacob Matson]: the system work? [Jacob Matson]: Also, we allow kind of space in the way that we interact with this so that like the magic [Jacob Matson]: of the company would differentiate the organization can still happen too.
[Jacob Matson]: And so like once I started thinking about it that way, that really kind of unlocked for [Jacob Matson]: me kind of a way for us to move forward. [Jacob Matson]: And it was much less difficult to kind of get people on board because it wasn't like, [Jacob Matson]: hey, we're going to change your totally change your job and make it so that like it just [Jacob Matson]: fits into this box. [Jacob Matson]: It was more like, okay, how do we meet in the middle? [Jacob Matson]: And so, you know, I think there's like a, I think there's a little bit of a paradox [Jacob Matson]: in that, right?
[Jacob Matson]: Which is like the paradox is that you need some level of conformity across the organization [Jacob Matson]: for like everyone to be able to communicate well. [Jacob Matson]: But also if you have too much conformity, you have a commodity and you need space for [Jacob Matson]: your, you know, to have some sort of differentiation. [Jacob Matson]: And so I think like that's kind of the perspective I brought there, you know, I think figuring [Jacob Matson]: out which constraints, I kind of like think about it sometimes like, like this game Jenga, [Jacob Matson]: I don't know if you, if you all played that, but you have like a stack of blocks, right?
[Jacob Matson]: Some of them you touch them and you're like, okay, I'm not, I can't pull that one out. [Jacob Matson]: That one has to stay there. [Jacob Matson]: It's like, but like you only figure that out like kind of existentially, right? [Jacob Matson]: Like you don't.
[Jacob Matson]: So like for me, I spent a lot of time just like trying stuff and like, okay, you know [Jacob Matson]: what, that didn't work. [Jacob Matson]: The CFO just got really mad at me, like we won't do that, but like, let's, let's just, [Jacob Matson]: let's just, let's try this other, other path. [Jacob Matson]: And I think a lot of it just became like, and then at the end you have this beautiful [Jacob Matson]: tower, right? [Jacob Matson]: Hopefully you don't knock it over, but you have this beautiful tower and that tower [Jacob Matson]: is unique shape that like hopefully fits what the actual, what, you know, actually represents [Jacob Matson]: what the business is.
[Moe Kiss]: And so that's kind of what I think about it. [Moe Kiss]: Hey, Tim. [Michael Helbling]: Have you ever opened GTM preview mode and immediately thought, well, there goes my afternoon. [Tim Wilson]: Absolutely.
[Tim Wilson]: Nothing says fun like hunting through a giant pile of tags, trying to figure out which one [Michael Helbling]: broke. [Michael Helbling]: Yeah. [Michael Helbling]: That's why state built, state GTM helper, a free Chrome extension for debugging Google [Michael Helbling]: tag manager. [Tim Wilson]: And free means actually, well, free, no sign up, no subscription.
[Tim Wilson]: Just install it from the Chrome web store and start debugging. [Michael Helbling]: Yeah. [Michael Helbling]: And it works with both web and server side GTM. [Michael Helbling]: It helps you focus on what matters by filtering down to the specific tags you're testing.
[Tim Wilson]: Your tags from Google, Meta and Microsoft are color coded, so they're easy to spot. [Tim Wilson]: And it makes JSON payloads readable instead of whatever they normally are. [Michael Helbling]: Yeah. [Michael Helbling]: And for server side GTM, it gives you better visibility into consent status and it can [Michael Helbling]: help with Shopify checkout debugging too.
[Tim Wilson]: There's even a Website Tracking Checker that gives you a web and server side tracking [Tim Wilson]: report with actionable fixes. [Michael Helbling]: The state GTM helper is a must have for anyone deploying our managing tags in GTM, search [Michael Helbling]: for state GTM helper in the Chrome web store, or use the link in the show notes page on [Michael Helbling]: our site. [Michael Helbling]: It's free, installs fast, and might just save your afternoon. [Tim Wilson]: Michael, where does your best AI analysis live right now?
[Michael Helbling]: Oh, I've got this Claude conversation called GA4 help for this meeting I've got coming [Michael Helbling]: up. [Michael Helbling]: I'm buried between a lunch recommendation and me asking it to explain regex to me like [Moe Kiss]: I'm a fifth grader. [Tim Wilson]: Exactly, that's the problem. [Tim Wilson]: Your AI work gets trapped in one chat with one person in one thread.
[Tim Wilson]: Ah, yes, the modern knowledge base. [Tim Wilson]: I swear Claude told me this somewhere. [Tim Wilson]: And that's why I ask why I built Prism with memory and shared context across users. [Tim Wilson]: So the useful stuff doesn't vanish into my private little AI cave?
[Tim Wilson]: Exactly, it's out of the cave into the sun. [Tim Wilson]: Prism keeps the context, your metric definitions, source of truth tables, business rules, prior [Tim Wilson]: analyses and makes it usable across the entire team. [Michael Helbling]: I like this. [Michael Helbling]: So if I teach it that active user means three sessions in 30 days, Julie doesn't have to [Michael Helbling]: teach it again tomorrow.
[Tim Wilson]: Exactly. [Tim Wilson]: And if Val runs a GA4 cohort analysis, that knowledge can live in Prism, organized and [Tim Wilson]: traceable, not locked inside her chat history like a tiny little analytics hostage. [Michael Helbling]: I am starting to like this team memory, not ask Michael because he remembers the cursed [Michael Helbling]: dashboard lore. [Tim Wilson]: Plus, with Claude co-work in Prism, your analysis becomes shareable, auditable and ready to [Tim Wilson]: build on.
[Michael Helbling]: I like this. [Michael Helbling]: So the AI becomes company knowledge, not just some vibes I had with the chatbot at 11.42 [Tim Wilson]: p.m.
[Tim Wilson]: That's right, because that's way after my bedtime. [Tim Wilson]: So go to ask-y.ai and join the wait list. [Michael Helbling]: And you can use the code APH and I'll take you to the top of the list.
[Michael Helbling]: That's ask-y.ai code APH because your team's brain should not be trapped in one person's [Michael Helbling]: chat tab. [Moe Kiss]: Exactly. [Moe Kiss]: Everyone can see my face, obviously not our lovely listeners, but everyone in the podcast.
[Moe Kiss]: I love an analogy. [Moe Kiss]: My company loves an analogy. [Moe Kiss]: And I feel like this Jenga one is going to take away too far because at some point you [Moe Kiss]: do knock it down. [Moe Kiss]: That's the reality of when we build data architecture and systems, at some point you do end up rebuilding.
[Moe Kiss]: But I think the exact tension that I feel right now, and I keep banging on about 80-90%, [Moe Kiss]: we do need standardization. [Moe Kiss]: We do need some conformity because otherwise, if one person over here calculates it this [Moe Kiss]: way and one person don't, we can't ever have a mature conversation. [Moe Kiss]: But I think the really challenging part is how we get that 10 to 15 or 20% that should [Moe Kiss]: be bespoke or is actually a unique situation. [Moe Kiss]: And I'm thinking about those are the Jenga blocks that you push through and you can move.
[Moe Kiss]: But you just have such a wealth of experience here. [Moe Kiss]: It sounds like for you a bit of that was trial and error. [Moe Kiss]: If I want to learn from all your trialing and the erroring, how do you figure out what [Moe Kiss]: the standardization bit is and where the flexibility needs to be? [Jacob Matson]: Oh, that's such a good question.
[Jacob Matson]: I think some of it is being able to zoom out and understand what the engine is of the [Jacob Matson]: company. [Jacob Matson]: How does it function? [Jacob Matson]: What's differentiating about your competitors? [Jacob Matson]: But then also how does that build the feedback loop that ultimately increases the cash on [Jacob Matson]: the balance sheet, hopefully?
[Jacob Matson]: I always felt like that was an advantage for me as someone coming from an accounting background [Jacob Matson]: where I'm just like, it's very easy for me to visualize, okay, if this business is great [Jacob Matson]: at X, they will increase their cash flow. [Jacob Matson]: And so I think a little bit is developing good intuition around that and I'm very thankful [Jacob Matson]: to have grown up and working for CFOs who are very excellent mentors as it related to [Jacob Matson]: that.
[Jacob Matson]: But I think the second part of that is your model for reality is imperfect. [Jacob Matson]: And so you need to be able to test that model in a way that is sort of safe. [Jacob Matson]: And what I mean by that is you don't get fired if you're wrong, you might get reprimanded. [Jacob Matson]: That's okay.
[Jacob Matson]: That's the threshold. [Jacob Matson]: Yeah, exactly. [Jacob Matson]: You can take some risk, but you want it to be the right risk. [Jacob Matson]: And so I think a lot of the trial and error part was just how do we take some risk here [Jacob Matson]: that is not too drastic, but is opinionated in a way that if we're correct, we win more.
[Julie Hoyer]: Do you have an example of what that risk is? [Julie Hoyer]: I don't know why. [Julie Hoyer]: I'm having a hard time conceptualizing the risky metric. [Julie Hoyer]: In the previous episode that we talked about semantic layers, I think you had thrown out [Julie Hoyer]: the example of monthly average users when we were talking about semantic layers.
[Julie Hoyer]: Is that a risky metric? [Julie Hoyer]: Is that a metric that moves the business forward? [Julie Hoyer]: Can we talk through a metric like that for a business like Canva? [Moe Kiss]: I think monthly active users is one that I wouldn't take a risk on.
[Moe Kiss]: And that's because it's one of our company foundational goals. [Moe Kiss]: So we have long historical reporting, but actually where it does get complicated, right? [Moe Kiss]: And I'm going to give you a specific example is we often will look at monthly active users [Moe Kiss]: by different products. [Moe Kiss]: And sometimes different people have different interpretations of what that product monthly [Moe Kiss]: active user is.
[Moe Kiss]: And that's why there's so much devil in the detail, right? [Moe Kiss]: Like it's such a gray zone because someone might be like, oh, anyone that used any kind [Moe Kiss]: of like video or social media and some people might be like, well, it's only video if you [Moe Kiss]: did X, Y and Z, like you were in our video editor and you used advanced video features. [Moe Kiss]: And that's why it's like, it's just messy, our jobs are messy. [Moe Kiss]: Totally agree.
[Jacob Matson]: When I think about risk, I guess like I would put it in a slightly different, I would not [Jacob Matson]: necessarily frame it as like analytics first, but I would just say that like I left a job [Jacob Matson]: and then like a month later, someone on the team who was still there sent me a message [Jacob Matson]: and was like, man, it has been rough since you've been gone. [Jacob Matson]: And I was like, why? [Jacob Matson]: And I'm like, everyone knows all the things that was nothing interesting happening.
[Jacob Matson]: He's like, well, no one's making any decisions. [Jacob Matson]: And I was just like, oh, yeah, okay, I could see that. [Jacob Matson]: And so I think like some of it, like when I talk about risk, I just honestly, I'm just [Jacob Matson]: like, make a decision, right? [Jacob Matson]: Like be opinionated on what it means to have a monthly active user by product X, Y, X, [Jacob Matson]: Y and Z, not said, sorry, you know, and like, you know, maybe that there's lots of interesting [Jacob Matson]: things that happen when you start breaking those things apart, right?
[Jacob Matson]: And you know, one thing that I think I was well trained on because I was in accounting [Jacob Matson]: is you get really good at like delivering bad news, like, you know, and so you're always [Jacob Matson]: in the, you're always, you know, one of the first objectives in accounting is like, you [Jacob Matson]: want this to be true. [Jacob Matson]: You want it to be the numbers you're showing are a reflection of reality as you understand [Jacob Matson]: it, right?
[Jacob Matson]: In a way that is defensible. [Jacob Matson]: And like sometimes when you're dealing with metrics, especially with product teams, like, [Jacob Matson]: you know, they want to show that their thing is working, right? [Jacob Matson]: And like what you're, you know, it's a tension, right? [Jacob Matson]: Between the domain team and like a central team, which is, okay, like what is truth to [Jacob Matson]: that, you know, you know, to the company and like what moves the business forward.
[Jacob Matson]: And I would almost always say that like, we want to be measuring things in a way that [Jacob Matson]: when they're tested against reality, they, they're proven to be right, right? [Jacob Matson]: And so when people are bringing agendas into things like, hey, like I want to define something [Jacob Matson]: in a way that says, you know, I get more monthly active users, well, if that's not moving [Jacob Matson]: the company forward, that metric when it's tested is going to fail, right?
[Jacob Matson]: And so how do we, how do we do that? [Jacob Matson]: How do we test them more closely against reality is a really interesting question. [Jacob Matson]: And the reason the way we do that is by like, you know, potentially taking, taking bets [Jacob Matson]: and like making decisions on them, right? [Moe Kiss]: Can I, can I, I'm taking us completely down off topic as per usual.
[Moe Kiss]: And I just want to push on this a little bit because I do see this happen and I'm curious [Moe Kiss]: if your experience in accounting has perhaps given you confidence or like you've built [Moe Kiss]: the confidence to sometimes have an opinionated decision. [Moe Kiss]: Whereas I feel like often in data lands, sometimes there is this like desire to debate [Moe Kiss]: every which way something can be cut and like maybe like make a proposal, but like not often [Moe Kiss]: enough be like, you know what, I'm going to have an opinion here of like, we're going [Moe Kiss]: to calculate it this way.
[Moe Kiss]: Let's do this. [Moe Kiss]: Let's move the business forward. [Moe Kiss]: Like, do you think sometimes like that, is that your accounting background? [Moe Kiss]: Do you think that helps you have that perspective because you're more willing to have a position [Moe Kiss]: knowing it's the best, the best of where you can get to, or do you think that's like you [Moe Kiss]: personally, like what do you think's driven that willingness to take a gamble and have [Moe Kiss]: an opinion?
[Jacob Matson]: I mean, I think it's a little bit of self selection. [Jacob Matson]: Like part of why I liked accounting was because it let me do things like that or like gave [Jacob Matson]: me a framework to reason about those things, right? [Jacob Matson]: I think I'm probably a little contrarian by nature. [Jacob Matson]: And so, you know, when I see people getting like too precious about metrics, I'm definitely [Jacob Matson]: just like, let's make a decision, let's, let's figure it out and we will test it.
[Jacob Matson]: And if it's wrong, we can fix it, right? [Jacob Matson]: One of the things that's great about, you know, analytics in general, compared to, I [Jacob Matson]: don't know, financial numbers that you're publishing to your board or whatever is that [Jacob Matson]: you have a lot more degrees of freedom in terms of what it looks like to go back and [Jacob Matson]: make something better. [Jacob Matson]: You know, one of the biggest challenges, right? [Jacob Matson]: And analytics is like, hey, that number got put into, you know, our regulatory filings.
[Jacob Matson]: So that's how we do that now. [Jacob Matson]: You cannot change that anymore. [Jacob Matson]: So like, obviously, like, you know, you don't want to take, you could, I don't know if this [Jacob Matson]: happened to me specifically, but I'm sure, well, actually, yes, it has. [Jacob Matson]: Or we made up some, some way to bin some set of data and then suddenly it was in, you [Jacob Matson]: know, annual reports.
[Jacob Matson]: And now it's like, okay, now we're always presenting that. [Jacob Matson]: And if we had known, if I'd known, I think from experience would have been like, hey, [Jacob Matson]: let's be a little more precious about this. [Jacob Matson]: And I think like, it's definitely a tough balance, but we don't need to be perfect. [Jacob Matson]: Right.
[Jacob Matson]: We can be, we, as long as we kind of know, you know, and it's justifiable and we can [Jacob Matson]: defend it, I think we can go pretty far. [Jacob Matson]: But like, you know, there's risk, right? [Jacob Matson]: There's risk that you could be wrong. [Julie Hoyer]: Do you feel like of all the context in a business, what percentage of it that people [Julie Hoyer]: use day to day, like when creating metrics, defining metrics, like doing their analysis, [Julie Hoyer]: making decisions, what percentage of it is actually captured in a formal, like static [Moe Kiss]: semantic layer for like broad knowledge compared to they're just doing it, like adding [Julie Hoyer]: in their own context and their own SQL queries and their own, you know, the way [Julie Hoyer]: they're pulling the data.
[Jacob Matson]: I mean, now that I work in marketing, it's a lot less than it was when I worked in [Jacob Matson]: finance. [Jacob Matson]: So it's contextual, I think. [Julie Hoyer]: Yeah. [Michael Helbling]: There are no generally, there are no generally accepted analytics principles, if you [Michael Helbling]: will.
[Michael Helbling]: Yeah. [Michael Helbling]: Yeah, sure. [Julie Hoyer]: Yeah. [Julie Hoyer]: I feel like it's a small percentage, smaller than maybe people want it to be.
[Julie Hoyer]: And do you feel like people are always fighting to like make it as close to a hundred [Julie Hoyer]: as possible? [Jacob Matson]: Or I think like the tension is that like, everyone wants the risk to be low. [Jacob Matson]: Like, hey, if I'm going to use data to make this decision, well, then it better be [Jacob Matson]: right. The data better be right.
[Jacob Matson]: And therefore, I'm not going to use the data because I don't want to take someone [Jacob Matson]: accountability for someone else, you know, something produced by someone else. [Jacob Matson]: I want to take my own accountability. [Jacob Matson]: You know, I think that's a core, definitely a challenge. [Jacob Matson]: Do I see it moving towards a hundred percent?
[Moe Kiss]: I mean, I think like, if you're moving toward a hundred percent, like you just [Jacob Matson]: automating the entire function, right? [Jacob Matson]: I mean, I guess that's like Google AdWords bidding, right? [Jacob Matson]: Like, okay, the whole thing's automated. [Jacob Matson]: The price is the price.
[Jacob Matson]: Um, in some ways, that's like the fully actualized form of analytics, right? [Jacob Matson]: Like auction pricing. [Jacob Matson]: Um, do I think that's the right way to do it? [Jacob Matson]: Like, I don't think, you know, most, most jobs don't are not that, not that [Jacob Matson]: straightforward.
[Jacob Matson]: I think that it's probably a smaller percentage than, than most analytics [Jacob Matson]: people would, would want it to be and probably, you know, roughly around the [Jacob Matson]: right number for where, where things are today. [Michael Helbling]: All right. [Michael Helbling]: I want to start to pivot into what we actually need to talk about, which was [Michael Helbling]: sure, let's say you've been struggling with the semantic layer and you're [Michael Helbling]: running into all the problems that semantic layers kind of introduce, you [Moe Kiss]: know, they're inflexible, uh, not easy to pull together.
[Michael Helbling]: Don't work well across different departments and teams. [Michael Helbling]: Like there's lots of reasons why a semantic layer is a challenge and, and [Michael Helbling]: lots of people spend a lot of time on it. [Michael Helbling]: But like, what are the alternatives? [Michael Helbling]: What are people doing to lower their dependency on the so-called semantic [Moe Kiss]: layer that is sort of like the, uh, favorite of the AI world right now in [Michael Helbling]: data?
[Jacob Matson]: I mean, I think like, you know, ultimately what we're seeing a lot of right [Jacob Matson]: now is that everything is kind of going into the notion of like skills, right? [Jacob Matson]: Which is just marked down, marked down on your laptop or in a GitHub repo or [Jacob Matson]: somewhere. [Jacob Matson]: And I think people are capturing a lot of context that way, that they are [Jacob Matson]: ultimately using either personally or sharing inside their company.
[Jacob Matson]: I think there's a lot of new service area here for products. [Jacob Matson]: Um, I know like, for example, inside of, uh, Claude, they have some of this [Jacob Matson]: notion called projects and projects that you kind of put marked down in [Jacob Matson]: there and then link, link it to other things. [Jacob Matson]: Um, and then whenever you ask a question, you can select a project and then [Jacob Matson]: you'll, you bring context along, you know, with it. [Jacob Matson]: So I think we're seeing that.
[Jacob Matson]: I think we're, we're definitely seeing like vendors coming along in the space. [Jacob Matson]: You know, we're seeing, we're seeing a lot of like, we're also seeing [Jacob Matson]: like the perspective from the labs, right? [Jacob Matson]: The labs who haven't limited tokens are just like, Oh yeah, we just [Jacob Matson]: like, you know, use AI to do, do everything. [Jacob Matson]: Like we're just like, you know, maximizing our tokens, spend, solve these [Jacob Matson]: problems.
[Jacob Matson]: I think the one that I saw that was, was killing me was like, open AI has [Jacob Matson]: like a analytics agent and they basically say, all right, well, we're [Jacob Matson]: just going to ask every question twice. [Jacob Matson]: So basically we'll ask, you know, the user will ask and then we're [Jacob Matson]: going to reformulate it and then have our, you know, our background agent, [Jacob Matson]: just see if it gets the same answer. [Jacob Matson]: And if they're too far apart, we're going to, we're going to escalate it.
[Jacob Matson]: And I think like, you know, certainly that's one approach. [Jacob Matson]: Uh, I, I can't imagine it's cost effective for anyone at reasonable [Jacob Matson]: scale who's not a lab at the moment. [Jacob Matson]: We're seeing lots of ways that people do it. [Jacob Matson]: I mean, from, from what we have been working on, you know, at mother duck, [Jacob Matson]: we've started to do is just, um, put context in a database because of [Jacob Matson]: course we're a database vendor.
[Jacob Matson]: So like put it in database database or AI is really good at writing SQL. [Jacob Matson]: It's really good at retrieving the right thing. [Jacob Matson]: Um, and so when you do that, then you can start, you know, um, treating it [Jacob Matson]: in a more structured way, right? [Jacob Matson]: Whether that's, uh, you know, more of like a graph that has like nodes and [Jacob Matson]: edges that you can use to like navigate across, or if it's just like, you [Jacob Matson]: know, straight up comments on columns or whatever, which is a, you know, [Jacob Matson]: old, old part of the SQL spec.
[Moe Kiss]: I am very much feeling the semantic world bubbling at the moment and the [Moe Kiss]: pressure of it and the perception that it's going to solve a lot of our [Moe Kiss]: problems or the perception that it's required to do AI and data well. [Moe Kiss]: I think the thing, one of the points that you made that really resonated [Moe Kiss]: with me in the article was about semantic layers being static and how challenging [Moe Kiss]: that is, but I think the real like thing that's keeping me up at night, right?
[Moe Kiss]: Is if, if we don't go down that semantic layer path, it's the [Moe Kiss]: validation, right? [Moe Kiss]: Which, which you just touched on, right? [Moe Kiss]: So the bit that's challenging that I, I see pop up constantly is if we don't [Moe Kiss]: have somewhere for people to self validate, I find that's, that's a hard [Moe Kiss]: thing that we, I want to solve for because I don't, my job to be, to QA other [Moe Kiss]: people's shitty outputs and AI hallucinations with data, which it feels [Moe Kiss]: like I spend some time doing now.
[Moe Kiss]: And so like, and I guess, I guess what I'm trying to say is like at the [Moe Kiss]: moment, I feel like it's binary that you either have some type of semantic layer [Moe Kiss]: thing where people can validate or you go down the evaluation framework. [Moe Kiss]: Am I, am I totally off here? [Moe Kiss]: Like, is it one of those binary options? [Moe Kiss]: Or is there just like a range of options?
[Moe Kiss]: And I haven't thought deeply enough about it yet. [Jacob Matson]: I mean, I think the first question is about the interface, right? [Jacob Matson]: That you let users interact with, right? [Jacob Matson]: If they're interacting with a spreadsheet, there's different set of [Jacob Matson]: constraints than if they're interacting with a BI tool, than if they're [Jacob Matson]: inner interfacing with like a chat app, right?
[Jacob Matson]: There's more engineering freedom, I think on, on obviously a chat application, [Jacob Matson]: which is what, you know, people have proven to love, you know, just asking [Jacob Matson]: questions to a chat bot, than there is on like a BI tool. [Jacob Matson]: I think that one of the biggest challenges on using, you know, even the best [Jacob Matson]: BI tool in the world at this moment is that it's very difficult to interface [Jacob Matson]: with something else that someone else built.
[Jacob Matson]: Like you're just, I think one of the things that I've really found kind of [Jacob Matson]: when using AI generally is that like the closer the framework you're using to [Jacob Matson]: answer questions is like a snap fit to your own brain, the easier it is to [Jacob Matson]: like use it and use it well. [Jacob Matson]: When you're using someone else's model, like even a really well-defined model [Jacob Matson]: by a really good engineer or really good BI analyst or whatever, it's like, [Jacob Matson]: that's their model.
[Jacob Matson]: That's not your model. [Jacob Matson]: That's not necessarily how you're thinking about the problem space. [Jacob Matson]: And like the biggest challenge that LLMs solve is they're really good at [Jacob Matson]: translating between languages, right? [Jacob Matson]: And that really means they're really good at for me as a, let's say, someone [Jacob Matson]: working at marketing to ask a question and then have the LLM reframe it to be [Jacob Matson]: like, oh, I see what this really means is, you know, it means this and, you know, [Jacob Matson]: that translates to this language in your current model or whatever.
[Jacob Matson]: And so I know I'm not really that kind of answering your question. [Jacob Matson]: I don't know, like, I think there is a spectrum. [Jacob Matson]: I don't know if we know if, like, I think the products are not super mature [Jacob Matson]: as or outside of the semantic layer because the semantic layer buyer, I think [Jacob Matson]: traditionally has been very risk averse. [Jacob Matson]: That's why they're buy it.
[Jacob Matson]: The reason you buy a semantic layer in the first place is because some [Jacob Matson]: number got somewhere and it was wrong and legal's pissed. [Jacob Matson]: You know, like that's how you buy a semantic layer. [Jacob Matson]: I'm being maybe a little too cynical, but like only maybe. [Jacob Matson]: Um, so I think like figuring out how to, like, how do we, again, I keep, I keep [Jacob Matson]: saying this word risk, but like, it all comes down to like, how much risk do you [Jacob Matson]: accept?
[Jacob Matson]: And like that determines the spectrum of tools that you can implement. [Jacob Matson]: Right. [Jacob Matson]: Um, if you can, if you can take on a lot of risk, like in marketing analytics, [Jacob Matson]: you can probably take on more, more risk than you can in financial analytics. [Jacob Matson]: That's just true.
[Jacob Matson]: Right. [Jacob Matson]: The cost of being low of wrong is way lower in marketing than it is in [Jacob Matson]: finance. [Jacob Matson]: That's just true. [Jacob Matson]: Like that's, that's a physics problem.
[Jacob Matson]: Um, so I think like you may not have the same solution across the entire [Jacob Matson]: company. [Jacob Matson]: It just depends, you know, horses for courses, as they say, I suppose. [Moe Kiss]: Just to be clear, I will keep asking Jacob 50,000 questions. [Michael Helbling]: I'm trying to make space for other people by not speaking.
[Michael Helbling]: That's okay. [Michael Helbling]: And we, we added out dead air. [Michael Helbling]: So don't, don't worry about that. [Michael Helbling]: That's fine.
[Julie Hoyer]: I wanted to ask actually, Jacob, if you could talk about your proposed [Julie Hoyer]: solution in your article that you talked about, like using AI, because [Julie Hoyer]: everybody's obsessed with having a semantic layer to help with AI, but [Julie Hoyer]: you kind of flipped in said also AI could help with the semantic layer problem. [Jacob Matson]: So when I wrote the original article, what I was really thinking about at the [Jacob Matson]: time was, uh, the notion of skills and using skills kind of locally on your [Jacob Matson]: machine to, to kind of codify the, the way to go here.
[Jacob Matson]: I think I've gotten a little more nuanced recently, but like also I think that [Jacob Matson]: we've seen a lot of mature maturity and development, especially like in the [Jacob Matson]: anthropic ecosystem with projects inside of, inside of cloud, for example. [Jacob Matson]: I think it's a very natural way to think about it is like, how do I, how do I [Jacob Matson]: make it easier to maintain and, or actually create and maintain and skills [Jacob Matson]: are incredibly, incredibly easy to create and maintain, maybe too easy, right?
[Jacob Matson]: They may not be the right abstraction. [Jacob Matson]: But certainly I think skills is the way that I, I thought about it or, you know, [Jacob Matson]: at the time, and then I just, you know, and I've been doing kind of evals in [Moe Kiss]: that front against, you know, benchmarking data sets. [Jacob Matson]: And, you know, we can debate the, the efficacy of benchmarks versus, you know, [Jacob Matson]: real, real life data and all these things. [Jacob Matson]: But I think what we do know is true is that if we can find the right context, we [Jacob Matson]: can very reliably return the right answer.
[Jacob Matson]: And so I think like where my, where I think my, where my thinking has gone [Jacob Matson]: recently is, is how do we make our data easy to search, right? [Jacob Matson]: Which is a different way than maybe we've designed it before, as like a [Jacob Matson]: Kimball model, which is like, how do we make it easy to retrieve? [Jacob Matson]: You know, there's a whole bunch of really good research that those guys [Jacob Matson]: all wrote around how to make a data warehouse and how to make it work [Jacob Matson]: and easy to retrieve.
[Jacob Matson]: And those were written in the constraints of the time, right, which I [Jacob Matson]: think probably the original Kimball book is probably going on 25 or 30 years now. [Jacob Matson]: We have better technology now, and maybe we can revisit some of those assumptions. [Jacob Matson]: And so my, I think my current kind of position on this stuff is that we really [Jacob Matson]: have a search problem. [Jacob Matson]: And if we can find a way to make our metrics searchable, then the way that [Jacob Matson]: we define it may be less important.
[Jacob Matson]: It might be a semantic layer that's in YAML. [Jacob Matson]: It might be something that's more, that's more well defined than that. [Jacob Matson]: It might be, you know, something more programmatic than even YAML that is [Jacob Matson]: like kind of like a SQL alternative. [Jacob Matson]: But like what I, what I have also found is that like LLMs in particular are [Jacob Matson]: so good at writing SQL and understanding it, but like adding another language [Jacob Matson]: that is new and bespoke is really hard to get good results out of compared to [Jacob Matson]: just using the trusted good old thing that is very verbose and like has weird [Jacob Matson]: syntax and like has a whole bunch of downsides, but also like there's 50 years [Jacob Matson]: of training data in the, in the training set, right?
[Jacob Matson]: So if we can figure out how to say, how do we make it search to find the right [Jacob Matson]: thing and then write the right SQL, we can get really good results. [Jacob Matson]: And I think that's what I'm seeing is most promising, you know, at this moment. [Jacob Matson]: Can we separate search from the context? [Moe Kiss]: Because I feel like those are two very different things and where you struggle [Moe Kiss]: with both.
[Moe Kiss]: So search is more about like, how do I point it in the right place? [Moe Kiss]: How do I help find the right table or whatever it is, right? [Moe Kiss]: Whereas the context is more, how do I understand this table? [Moe Kiss]: And is, okay.
[Moe Kiss]: And do you think that what you're proposing, the way you're thinking about [Moe Kiss]: this solves for both equally? [Moe Kiss]: Or do you think perhaps like, I feel like we probably need a different approach [Moe Kiss]: for each or different thinking? [Moe Kiss]: The answer is the devil's in the details, I think. [Jacob Matson]: So let me, I think where I'm struggling with this question is like, if you assume [Jacob Matson]: that there is a natural language interface, right?
[Jacob Matson]: And then maybe some sort of tooling like an MCP or something that has a search [Jacob Matson]: tool. [Jacob Matson]: Well, that's, you know, someone should solve the search tool problem, right? [Jacob Matson]: Which will say, hey, let me search and find this context and return it to you. [Jacob Matson]: And the next thing is, you know, writing the SQL based on that understanding.
[Jacob Matson]: I think the second part is basically solved, which is if you give good context [Jacob Matson]: to an agent and say, like, describe a set of tables and then ask a question, [Jacob Matson]: you get really good results. [Jacob Matson]: And, you know, the challenge we have, of course, is that like, they don't have [Jacob Matson]: good memory. [Jacob Matson]: And so like every day, you have to remind them. [Jacob Matson]: And so how do we make it so they have good memory?
[Jacob Matson]: And there's a whole bunch of people, I'm sure, working on that problem. [Jacob Matson]: I've read quite a few papers, you know, in the space of like, how do we make it? [Jacob Matson]: So when we ask a question, you know, the next time, the next time we ask it, [Jacob Matson]: it's faster to get the answer. [Jacob Matson]: Or we already know, you know, we already have some kind of, you know, [Jacob Matson]: pathway built out.
[Jacob Matson]: The metaphor I kind of think about is like, your data is like a jungle, right? [Jacob Matson]: Unless you really know where the things are. [Jacob Matson]: It's really hard to find stuff. [Jacob Matson]: We can build a semantic layer, but a semantic layer is like a highway.
[Jacob Matson]: Right? [Jacob Matson]: It's like, we are just like building the thing straight through the jungle. [Jacob Matson]: I would almost say like, we can build it a little more progressively, [Jacob Matson]: which is, hey, we need like these little paths, right? [Jacob Matson]: Maybe these paths are just wide enough for a human to walk on.
[Jacob Matson]: And maybe this one we can put gravel on. [Jacob Matson]: And maybe this one we can put gravel on. [Jacob Matson]: And maybe this one we can pave. [Jacob Matson]: And then maybe this one we can build a highway.
[Jacob Matson]: But like, so I kind of think like, I think this actually goes back to the spectrum [Jacob Matson]: question you asked earlier, like there's probably a spectrum of solutions here [Jacob Matson]: where a certain set of questions need to be on the highway, right? [Jacob Matson]: Like if you're one of your key metrics as monthly active users, [Jacob Matson]: that needs to be right all the time. [Jacob Matson]: And there's a highway for that, right? [Jacob Matson]: And so the search problem is like, can my agent find the highway?
[Jacob Matson]: Okay. [Jacob Matson]: And if you can find the highway, then you get the answer. [Jacob Matson]: And then, you know, there's a bunch of, but the interesting stuff, [Jacob Matson]: the reality is the interesting stuff is when you start slicing and dicing [Jacob Matson]: by all these arbitrary dimensions. [Jacob Matson]: And then like, there's no highway for that.
[Jacob Matson]: I can't afford to build that even in the age of AI today. [Jacob Matson]: And so, you know, how do you make it easy to do a little bit of off-roading [Jacob Matson]: and then still get the right thing, right? [Jacob Matson]: And I think like this is where, you know, I think I'm, [Jacob Matson]: I'm in particular probably well served by like, you know, accounting principles [Jacob Matson]: as in building this stuff out. [Jacob Matson]: And I'm just like, well, what if we just made this a ledger [Jacob Matson]: and we can just like, you know, walk forward and backwards through time [Jacob Matson]: or whatever.
[Jacob Matson]: Now it becomes very easy to trace it back to the highway. [Jacob Matson]: You know, there's a whole bunch of abstractions like that we can, [Jacob Matson]: we can talk about. [Jacob Matson]: But I think like the answer is probably, you know, and not, not or in the long term, [Jacob Matson]: I do think that there's probably some set of questions that are so important [Jacob Matson]: that you need to always get the right answer. [Jacob Matson]: And that might mean maybe a different interface.
[Jacob Matson]: Like, hey, you know, for our financial reporting interface, we use X [Jacob Matson]: for marketing, we use Y. I don't know. [Moe Kiss]: Do you think it depends on stakeholders too? [Moe Kiss]: Like I'm just thinking.
[Moe Kiss]: So for example, if you're a data person, the context is something that exists [Moe Kiss]: within you. [Moe Kiss]: Like you have such good context. [Moe Kiss]: So you know how to prompt, you know how to give the right context. [Moe Kiss]: And I'm talking as you're like in the build stage and this is evolving, [Moe Kiss]: right?
[Moe Kiss]: Because if we have essentially, you know, you're kind of suggesting like an [Moe Kiss]: evolving semantic layer or a knowledge-based graph base, [Moe Kiss]: at the earlier stage, right, when that context isn't fully built, [Moe Kiss]: what do you think the like the risk is for the business user to be exposed [Moe Kiss]: who potentially doesn't have that same content? [Moe Kiss]: Like, do you know what I mean? [Moe Kiss]: Like you haven't built the enough of the highways yet.
[Jacob Matson]: Yeah, such a good question. [Jacob Matson]: So when we first were, so we launched our MCP server in December at MoetherDuck. [Jacob Matson]: And we had our own existing set of dashboards for the sales team. [Jacob Matson]: And immediately what started happening is the sales team started like [Jacob Matson]: building their own kind of dashboards using our tool, which is really [Jacob Matson]: interesting to see, right?
[Jacob Matson]: We're, I mean, we're a small company where our risk threshold is obviously higher. [Jacob Matson]: But what everyone was doing was because they already had really good context [Jacob Matson]: for the data, because that, you know, they're sales team. [Jacob Matson]: They're, they, if the data is wrong, you know, everyone's mad. [Jacob Matson]: It doesn't work.
[Jacob Matson]: But like it's, it's tested against the sales data in particular is [Jacob Matson]: tested against reality all the time, right? [Jacob Matson]: Because you're rolling on a call with a customer and you're like, hey, I saw [Jacob Matson]: you used, you know, 10 hours of compute last week, you know, we sell compute. [Jacob Matson]: And they're like, no, we didn't. [Jacob Matson]: You're going to be like, okay, let me go.
[Jacob Matson]: Oops, let me go fix my dashboard. [Jacob Matson]: So they have the stakes are pretty high for them, right? [Jacob Matson]: And for financial teams too, they're really high, right? [Jacob Matson]: They need to be right.
[Jacob Matson]: They need to be speaking about the right numbers. [Jacob Matson]: And so for teams that have developed an intuition for what the data should be, [Jacob Matson]: they're actually pretty good at using LLMs, regardless of language and [Jacob Matson]: understanding like being a data person first, because they have a way to [Jacob Matson]: test it against reality really quickly, right? [Jacob Matson]: They can break down a set of numbers by, you know, like, if you're a CFO or [Jacob Matson]: something, you could probably break down revenue by product and you would with [Jacob Matson]: an LLM and be like, that's right.
[Jacob Matson]: I can tell that it's right because I've seen this before, right? [Jacob Matson]: You already have like the fingertip feel for the data, but if you're someone [Jacob Matson]: else, especially like, this is the hard part for like a data analyst who's [Jacob Matson]: maybe disconnected from the business is they don't have the fingertip feel. [Jacob Matson]: They just have like a question from someone that says, hey, build me this [Jacob Matson]: thing. [Jacob Matson]: It's really, you know, in that case, you really have to, you know, rely on the [Jacob Matson]: context of others to help build the right thing.
[Jacob Matson]: You know, I think like that's, so yes, I think the stakeholder does matter. [Jacob Matson]: You know, I think that that certain teams are more well suited to be data [Jacob Matson]: driven than others just like out of the box, but like that we can definitely [Jacob Matson]: close the gap with like good context, maybe not for everyone, right? [Jacob Matson]: If someone asks a totally off the wall question using the wrong words, like [Moe Kiss]: that's a, you know, we don't have a crystal ball.
[Jacob Matson]: We have, we have, you know, vectors that we're matching at the end of the day, [Jacob Matson]: right? [Julie Hoyer]: With this like approach that you're talking about, like using AI to kind of [Julie Hoyer]: bring up the context for somebody who doesn't have it themselves. [Julie Hoyer]: I have two questions. [Julie Hoyer]: One, and I think, well, you were kind of asking this.
[Julie Hoyer]: So it was like, how do you know when it finds context that it's choosing the [Julie Hoyer]: right context? [Julie Hoyer]: Because again, we've talked about people are defining and talking about the [Julie Hoyer]: same metric differently, especially if it's not one of the highway metrics. [Julie Hoyer]: The other part is it feels like classically the legwork, right, was on [Julie Hoyer]: the individual trying to ask the question, make the query. [Julie Hoyer]: They had to go around and obviously ask and get all that context.
[Julie Hoyer]: But with the new, if we put in the solution that you're talking about, [Julie Hoyer]: where does the legwork now live in that process? [Julie Hoyer]: Do you know what I'm saying? [Julie Hoyer]: We've shifted the hard work of determining what is right for the [Julie Hoyer]: question you're asking of where to grab it, how to think about it, how it's [Julie Hoyer]: defined, and how to define the metric you're trying to query. [Julie Hoyer]: So I'm trying to understand those two pieces.
[Jacob Matson]: The first thing that I always think about is like, well, how do we make it visual? [Jacob Matson]: This was a video podcast. [Jacob Matson]: I could show you a really sweet demo, but maybe I'll just have to send it to [Jacob Matson]: you. [Jacob Matson]: I'll send you a video async that shows you one way that it could look like.
[Jacob Matson]: So I think the first thing is how do we make it so that interacting with the [Jacob Matson]: context is not just typing into a box and then stuff happens and you get an [Jacob Matson]: answer. [Jacob Matson]: We need to make it like you need to be able to understand what that looks [Jacob Matson]: like. [Jacob Matson]: And I think for a lot of this, because I spent so much time in the Excel salt [Jacob Matson]: mines, I think about it a lot like I think about Excel, which is you have [Jacob Matson]: some tab somewhere that says, here's this metric, right?
[Jacob Matson]: Here's our profit for last quarter. [Jacob Matson]: And there's a little button in there that shows trace precedence. [Jacob Matson]: Okay, where does that take me? [Jacob Matson]: Okay, that takes me upstream.
[Jacob Matson]: And then I can keep navigating all the way back until I kind of understand how [Jacob Matson]: this thing comes from. [Jacob Matson]: And then there's another little button on there that says calculate formula, right? [Jacob Matson]: And it just shows me all the little parts. [Jacob Matson]: And it shows how they're adding and subtracting and dividing and multiplying [Jacob Matson]: and how it ends up with the final number.
[Jacob Matson]: I think like we don't have those traceability pieces yet for agent work flows [Jacob Matson]: or semantic layer stuff, really. [Jacob Matson]: I mean, we do, but only for the developer, not for the user. [Jacob Matson]: And so what I'm really thinking about is, you know, the way for me to like manifest [Jacob Matson]: this notion of, hey, you don't need the semantic layer. [Jacob Matson]: It means you need to have something else.
[Jacob Matson]: And part of it is, yes, you need the context. [Jacob Matson]: But you also need a way to visualize it in a way that like people can reason about [Jacob Matson]: and understand and like, you know, agree with how it was calculated, right? [Moe Kiss]: But you're saying visualize the lineage, though, and the way it's calculated. [Jacob Matson]: I think all parts of that, right?
[Jacob Matson]: Lineage, yes. [Jacob Matson]: But like less about like, I mean, I would love to be like, okay, this number comes [Jacob Matson]: all the way from this field in your CRM. [Jacob Matson]: You know, I would love to be able to do that, right? [Jacob Matson]: Okay.
[Jacob Matson]: Or like, you know, some sort of way to say, you know, hey, like we were talking about [Jacob Matson]: one of the active users earlier, like this, here's all the detailed ways that that's [Jacob Matson]: calculated. [Jacob Matson]: I think that's probably actually too hard to reason about. [Jacob Matson]: It's the wrong, it's too detailed, right? [Jacob Matson]: We need to kind of be able to get the proper level of zoom out, right?
[Jacob Matson]: We don't want to be at the 10 foot level. [Jacob Matson]: We don't want to be at the 40,000 foot level. [Jacob Matson]: We want to be at like the 10,000 foot level. [Jacob Matson]: And I think it probably depends on persona too.
[Jacob Matson]: But like you need some sort of way to be able to reason about, you know, those numbers [Jacob Matson]: and how they were calculated to be confident that they're correct, right? [Jacob Matson]: And like SQL is of course one way to reason about it, right? [Jacob Matson]: But like SQL is like verbose and kind of hard to reason about as like a non-technical [Jacob Matson]: person and like even, you know, engineers hate using it. [Jacob Matson]: And there's a whole good set of reasons for that.
[Jacob Matson]: You know, I think like the magic, the magic of Excel continues to be like demystifying [Jacob Matson]: how complex some of this stuff is and doing that with really clever UI interactivity. [Jacob Matson]: And so, you know, how do we, you know, what tools, what abstractions do we need to build [Jacob Matson]: to make that work? [Jacob Matson]: And I've been working on that. [Jacob Matson]: I think the core thing that I've been starting with is just like, how do we take something [Jacob Matson]: like a set of tables and like interweave context in there to show us how the tables relate [Jacob Matson]: to each other, right?
[Jacob Matson]: In some ways you'd be like, hey, that looks a lot like ERD, right? [Jacob Matson]: It's like, here's the primary keys, here's the foreign keys, whatever, right? [Jacob Matson]: But we can add a richer level of that that says, okay, we calculate these metrics this [Jacob Matson]: way, we use these joins, you know, we use this formula. [Jacob Matson]: That stuff can all, that stuff is also, you know, a level of the graph that's always been [Jacob Matson]: missing when we get the ERD, right?
[Jacob Matson]: We don't know how it's actually used. [Jacob Matson]: We just have a thing that says, here's how the database defines the table. [Jacob Matson]: What we don't have is knowledge of how the application actually uses it. [Jacob Matson]: And so, that's I think the next part too, you know, we can mine like query history.
[Jacob Matson]: In fact, I've done a bunch of work around that, like how do people actually query these, [Jacob Matson]: how do databases or how do applications query these, how do those patterns differ when it's [Jacob Matson]: a human versus an application. [Jacob Matson]: So there's a whole bunch of stuff we can do there. [Jacob Matson]: I'm not trying to be like too abstract or obtuse here, but like, I truly believe that [Jacob Matson]: that part of it is just that we have all of the, we have all of like the Lego blocks, [Jacob Matson]: but like no one has assembled it into something that is like cohesive in a way that's like, [Jacob Matson]: I totally get this now, right?
[Jacob Matson]: It's very much like, even if you're like, you know, high-ranking executive and you like [Jacob Matson]: get some number, there's no like, you're trusting that your team built it right. [Jacob Matson]: And like there's not a good way today. [Jacob Matson]: And like lineage is part of that, but like just being able to say like, all right, how [Jacob Matson]: do all these components fit together and like fitting that in your brain as like, I don't [Jacob Matson]: know, maybe if you're the CMO, you shouldn't do that, but like, I don't know, maybe you [Jacob Matson]: should do that sometimes.
[Jacob Matson]: I don't know, hopefully that's helpful. [Jacob Matson]: Hopefully that's helpful. [Jacob Matson]: That's kind of just how I'm thinking about it at the moment, but like make it visual is [Jacob Matson]: number one. [Julie Hoyer]: I was going to ask, how do you fight the echo chamber effect too from some of this?
[Julie Hoyer]: Sorry, when you were talking about like using AI to search for queries, like if it's just [Julie Hoyer]: always bringing back historical data and you were to like move that into the future. [Julie Hoyer]: How do you do that? [Jacob Matson]: Okay. [Jacob Matson]: So let me just make sure we from this question.
[Jacob Matson]: So you're basically, you're basically saying like, how do you not overfit the history? [Julie Hoyer]: Because they were figuring it out. [Julie Hoyer]: Some of it's not right. [Julie Hoyer]: Some of it is.
[Julie Hoyer]: Or, you know, it's just the bias of like, what was right at that time and you know, how [Julie Hoyer]: do you get things in? [Julie Hoyer]: This was my exact question. [Julie Hoyer]: You have people that aren't going for the context, the old fashioned way they're relying on the [Julie Hoyer]: tool to give them the context, but it's old context, you know. [Jacob Matson]: So I think the first thing that I'm thinking about is like, well, if you had a semantic [Jacob Matson]: layer, you only have the history.
[Jacob Matson]: You don't have anything going forward because someone curated and built that for you. [Jacob Matson]: So now we get a new problem to solve, right? [Jacob Matson]: Now, if we can just forget the semantic layer exists or maybe we have it and it's, you know, [Jacob Matson]: it's the highway, right? [Jacob Matson]: But we want to detect when there are changes, right?
[Jacob Matson]: When there's drift, when there's drift in our metrics, right? [Jacob Matson]: How do we, how do we detect that? [Jacob Matson]: I think this is a really good question. [Jacob Matson]: I don't know if there's like good programmatic ways that I'm like aware of, like I'll talk [Jacob Matson]: my head at the moment, but certainly like there's probably ways to do it programmatically.
[Jacob Matson]: And I think like also some of it is like, well, if there's less humans, you know, doing, [Jacob Matson]: doing some of the, some of the work that's really easy to automate with AI, well, now [Jacob Matson]: what do those humans do next? [Jacob Matson]: I think part of it is like, yeah, maybe we need like a curator person who's like handling [Jacob Matson]: this context potentially, right? [Jacob Matson]: You know, I think I kind of like always jokingly talk about like, hey, we need like more librarians.
[Jacob Matson]: Like we're generating all this context all the time. [Jacob Matson]: And it's like, what do we do with it? [Jacob Matson]: It's like, I don't know. [Jacob Matson]: It just like lives in Slack or like teams or our emails or, you know, DMs on WhatsApp [Jacob Matson]: or whatever.
[Jacob Matson]: And like eventually if it's like, eventually it gets, it makes its way, you know, into [Jacob Matson]: the, the canon of the, of the organization, right? [Jacob Matson]: But like that, just that time is like really, it can be really long. [Jacob Matson]: How do we make that tighter is a really interesting question. [Jacob Matson]: I do think it's like in the other side of that is like, well, what about archival, right?
[Jacob Matson]: What if a metric is now was right and is now wrong? [Jacob Matson]: How do you discard it and like manage the life cycle of it? [Jacob Matson]: So I think like all of those pieces in my mind like kind of fit together, which is like life [Jacob Matson]: cycle management of this. [Jacob Matson]: And it's like, we never even got there because we just get to like, we get to like semantic [Jacob Matson]: layer is published.
[Jacob Matson]: And then it's like, it's such a big lift. [Jacob Matson]: It's like, even get there that it's just like, okay, like I got promoted. [Jacob Matson]: Thank you. [Jacob Matson]: I'm going to go to another company and I'll do this again.
[Jacob Matson]: This is now your problem. [Jacob Matson]: Like, sorry. [Jacob Matson]: And I'm being, again, I'm being sort of facetious, but like, these are hard. [Jacob Matson]: Like everyone wants to build it.
[Jacob Matson]: No one wants to maintain it. [Jacob Matson]: And I think like how do we, I think we, you know, there's probably space for companies [Jacob Matson]: or multiple companies in the space to build products that help us do this better. [Jacob Matson]: But, you know, I don't have anything on top of my head that's like, well, here's how you [Jacob Matson]: catch drift and like reimplement it. [Jacob Matson]: You know, I wish I had the answer.
[Jacob Matson]: You know, it would be, it would be more compelling maybe next time, next time. [Michael Helbling]: Right. [Moe Kiss]: Let me do more. [Michael Helbling]: That can do my next set of research.
[Jacob Matson]: We'll see if we can do it. [Jacob Matson]: You know, we are, we are fully migrating our internal stuff. [Michael Helbling]: So if we wanted certainty, Jacob, we would have just interviewed Claude, right? [Michael Helbling]: So this is great.
[Michael Helbling]: Yeah, exactly. [Michael Helbling]: Exactly. [Michael Helbling]: This is awesome. [Michael Helbling]: Thank you so much.
[Michael Helbling]: Really excellent conversation. [Michael Helbling]: One of the things we do at the end of every show is go around the horn and share a last [Michael Helbling]: call, something that might be of interest to our users. [Michael Helbling]: Jacob, you're our guest. [Jacob Matson]: Do you have a last call you'd like to share?
[Jacob Matson]: I just read a really awesome paper that I will share a link to you to Michael called [Jacob Matson]: skill opt executive strategy for self-evolving agent skills. [Jacob Matson]: I saw it on Twitter this morning. [Jacob Matson]: It is really interesting. [Jacob Matson]: I'll share it and you can put it in the show notes.
[Jacob Matson]: Definitely worth the read just to understand, you know, what it looks like to actually apply [Jacob Matson]: some of this stuff about what exactly we were talking about. [Jacob Matson]: How do we evolve these things as the systems change? [Jacob Matson]: That's awesome. [Jacob Matson]: I love it.
[Michael Helbling]: Awesome. [Michael Helbling]: Julie, what about you? [Julie Hoyer]: Okay. [Julie Hoyer]: My last call is a little off the wall, but it was just too timely.
[Julie Hoyer]: Okay. [Julie Hoyer]: So, Moe, this is a little bit, a fun fact about me and a question to you because it has to [Julie Hoyer]: do with Canva. [Moe Kiss]: Okay. [Julie Hoyer]: Fun fact.
[Julie Hoyer]: I don't like squirrels. [Julie Hoyer]: Like, I just, I don't like them. [Julie Hoyer]: I don't think they're cute. [Julie Hoyer]: They creep me out.
[Julie Hoyer]: Okay. [Julie Hoyer]: Yeah. [Julie Hoyer]: They kind of scare me. [Julie Hoyer]: They kind of scare me because quick backstory.
[Julie Hoyer]: My dad said, you know, to this little, little girl, if you see possums are at Coons in the [Julie Hoyer]: daylight, they're rabid. [Julie Hoyer]: I thought that meant squirrels. [Julie Hoyer]: I went around for years thinking squirrels were rabid. [Julie Hoyer]: You know how they pop around trees when you're riding your bike.
[Julie Hoyer]: They like hide on the other side. [Julie Hoyer]: I always thought they were going to pop out and get me. [Julie Hoyer]: I don't like squirrels. [Julie Hoyer]: Okay.
[Julie Hoyer]: Fun fact that I don't share with lots of people. [Julie Hoyer]: I get an email from Canva that is squirrel themed and the CTA at the bottom of the email [Julie Hoyer]: said, if you would like to stop getting squirrel themed content like opt out here. [Julie Hoyer]: Like has AI personalized too far? [Julie Hoyer]: Or like, is this, is this a phenomenon out in like the trendy world that I just have [Julie Hoyer]: not been exposed to like our squirrels a thing or do they know this about me?
[Julie Hoyer]: So it kind of freaked me out. [Julie Hoyer]: And there's my fun fact. [Moe Kiss]: Also, like is squirrel a cool hip thing that I also don't know about because I mean that [Moe Kiss]: level of personalization. [Michael Helbling]: Well, if the other cohort was getting emails about moose, then you're something, now you're [Michael Helbling]: getting somewhere, but that's probably a reference that nobody knew.
[Michael Helbling]: Did anybody else get the squirrel email? [Moe Kiss]: Julie, I will follow up and we can, I can share an update with you in our next class [Moe Kiss]: called really people hang in. [Michael Helbling]: Well, there you go. [Michael Helbling]: That's a good last call.
[Michael Helbling]: That just shows the range we can have here. [Michael Helbling]: Amazing. [Michael Helbling]: All right. [Michael Helbling]: Moe, what about you?
[Michael Helbling]: What's your last call? [Moe Kiss]: Okay. [Moe Kiss]: I have been reading a book which Rachel Gerson recommended. [Moe Kiss]: It's called Word Slut, a feminist guide to taking back the English language by Amanda [Moe Kiss]: Moentell.
[Moe Kiss]: And I just like, you know, words are important. [Moe Kiss]: Like you do. [Moe Kiss]: But when you go through this book and you hear about the evolution of certain words and [Moe Kiss]: how they refer to women, it just like, it's, I don't want to say delightful. [Moe Kiss]: It's been surprising in a really great way.
[Moe Kiss]: Like I feel like it's kind of added to me wanting to be more thoughtful about the word [Moe Kiss]: choices that I use, which is entertaining because I don't give too much thought to what [Moe Kiss]: comes out of my mouth. [Moe Kiss]: So anyway, go check it out. [Moe Kiss]: And Michael, over to you. [Michael Helbling]: Yeah.
[Michael Helbling]: So mine is from back in May. [Michael Helbling]: James Hawkins, the CEO of post hog wrote an article about what he was thinking about [Michael Helbling]: as sort of their next chapter of vision of the future. [Michael Helbling]: Just so your book, if listeners aren't aware, post hog is sort of both an analytics tool [Michael Helbling]: as well as other tools for digital and website operations. [Michael Helbling]: Anyway, it, I don't know that I agree with everything he wrote in terms of like where [Michael Helbling]: products should go, but I thought it was very thought provoking.
[Michael Helbling]: It definitely worth some time because all of us in analytics and in measurement spaces [Michael Helbling]: were facing a lot of change. [Michael Helbling]: And it's good to see like, okay, here's a company who's got a lot of customers who's [Michael Helbling]: doing a lot of work, especially with AI. [Michael Helbling]: Here's how they're looking at the future. [Michael Helbling]: So I think it's kind of a good read just to develop and gain perspective.
[Michael Helbling]: So that's why I would recommend that. [Michael Helbling]: All right. [Michael Helbling]: Once again, Jacob, thank you. [Michael Helbling]: Thank you so much for coming on the show.
[Michael Helbling]: This has actually been a really cool conversation and kind of like we, we, we stayed pretty [Michael Helbling]: high, but I feel like there's also some really amazing kernels for people to pick up on and [Michael Helbling]: drill into their orgs with that I think will actually bear a lot of fruit. [Michael Helbling]: And also because it relates to songs like so now in my head, I'm like, life is a highway. [Michael Helbling]: Okay. [Michael Helbling]: Wow.
[Michael Helbling]: Yeah. [Michael Helbling]: See, that's what, that's what happens to me when we talk about stuff. [Michael Helbling]: Anyways, so thank you again. [Michael Helbling]: And obviously to our listeners, yeah, I'm sure you've got thoughts and questions and we'd [Michael Helbling]: love to hear from you.
[Michael Helbling]: And there's a great way for you to do that. [Michael Helbling]: You can reach out to us on LinkedIn or on the measure slack chat group or via email contact [Michael Helbling]: at analytics hour.io and wherever you listen, you can also leave ratings and reviews and [Michael Helbling]: we read all of those as well. [Michael Helbling]: And Tim Wilson who's not here today would like you to know you can also request a sticker [Michael Helbling]: for your laptop.
[Michael Helbling]: Just go to analytics hour.io and there's a form that you can fill out and we will mail [Moe Kiss]: it to you even internationally. [Michael Helbling]: All right. [Moe Kiss]: This has been great.
[Michael Helbling]: I think there's about three more of these that probably need to be done. [Michael Helbling]: AI is changing so fast. [Michael Helbling]: But Jacob, thank you once again. [Michael Helbling]: And I know I speak for both my co-hosts, Moe and Julie when I say no matter where you [Michael Helbling]: are in the jungle, keep analyzing.
[Announcer]: Thanks for listening. [Announcer]: Let's keep the conversation going with your comments, suggestions and questions on Twitter [Announcer]: at analytics hour, on the web at analytics hour.io, our LinkedIn group and the measure [Announcer]: chat slack group. [Announcer]: Music for the podcast by Josh Grohurst.
[Announcer]: Those smart guys wanted to fit in so they made up a term called analytics. [Announcer]: Analytics don't work. [Charles Barkley]: Do the analytics say go for it no matter who's going for it? [Charles Barkley]: So if you and I were on the field, the analytics say go for it.
[Charles Barkley]: It's the stupidest, laziest, lamest thing I've ever heard for reasoning in competition. [Jacob Matson]: I think actually like, if you moved your cable around, I wonder if the cable was just like [Jacob Matson]: giving you some weird feedback or something. [Jacob Matson]: Because I was don't touch it. [Julie Hoyer]: Don't touch it now.
[Michael Helbling]: Don't touch it. [Michael Helbling]: Hands up. [Julie Hoyer]: Tim sent me a mic that he knew made a buzz. [Julie Hoyer]: That was like an evil trick.
[Julie Hoyer]: Sabotage. [Julie Hoyer]: Yeah, my toddler now uses it as her microphone. [Moe Kiss]: I was like, sure, you can have this. [Michael Helbling]: Does she do a podcast because you do one?
[Michael Helbling]: Because that would be the most adorable thing in the entire world. [Michael Helbling]: No, but she does like to sing. [Julie Hoyer]: She just breathes into the mic. [Michael Helbling]: So that's what I do.
[Moe Kiss]: I'm going to be triggered a lot with every time the word, the S word is mentioned. [Moe Kiss]: You know, usual. [Michael Helbling]: I do the exact same thing, but I look over because a lot of times I leave crap on my [Michael Helbling]: chair back here. [Michael Helbling]: And so then I look, I'm like, oh, clear.
[Michael Helbling]: I don't. [Moe Kiss]: My husband gets dressed in here in the morning and decides to leave whatever pants he didn't [Moe Kiss]: want to wear in the background. [Julie Hoyer]: Trust me. [Julie Hoyer]: You don't want to see what you can't see behind this cover.
[Michael Helbling]: All of the mess is my responsibility. [Moe Kiss]: I do get asked very frequently, though, why I have so many remotes on the wall behind [Moe Kiss]: me. [Moe Kiss]: And I'm like, it's a fair question. [Moe Kiss]: Well, there's another one that's missing.
[Moe Kiss]: So it's totally fine. [Michael Helbling]: I don't know. [Michael Helbling]: You're just, you're just a very successful person. [Moe Kiss]: And if people can't deal with that, I'm just, I'm afraid to do it.
[Michael Helbling]: I'd be afraid of that level of success is what my answer is. [Michael Helbling]: Five remotes. [Michael Helbling]: I know. [Michael Helbling]: I don't have the lifestyle governance required.
[Michael Helbling]: All right. [Michael Helbling]: Let's get into it. [Michael Helbling]: So I'll give us a five count and all right. [Michael Helbling]: All right.
[Michael Helbling]: Let's stop moving our microphone. [Moe Kiss]: It's the problem with the YOLO one. [Moe Kiss]: It like pains everybody. [Michael Helbling]: That's right.
[Moe Kiss]: That's right. [Moe Kiss]: We're very serious. [Julie Hoyer]: Rock flag. [Moe Kiss]: And if you build it, who will maintain it?
Other episodes covering the same guests and topics, from across The B2B Podcast Index.
#291 Why Most AI Projects Fail to Deliver ROI, Sinohe Terrero, CFO and COO, EnvoyGrowCFO Show · on Claude91 / 100
Why a $1.2B exit felt like his biggest failure, and the customer-obsession thesis behind AgencyThe GTMnow Podcast · on Claude86 / 100
Is Your AI Actually Worth What You're Spending? with Parker ConradStrictlyVC Download · on Claude86 / 100
Is Your Business Invisible to AI Search? (And How to Fix It) ft. Ray YoungRevenue Science · on Claude85 / 100
Episode 018: Season 2, the $75 Consult and the Frankenstein StackAI Tools for Practicing Lawyers · on Claude84 / 100
Agentic Engineering for Testers: How to Automate Your Way to the Top with Amit RawatTestGuild Automation Podcast · on Claude82 / 100#299: AI Can (Help) Build the Dashboard. It Can't Build the Buy-In.66 / 100#298: Listener Questions Answered Live from Marketing Analytics Summit!66 / 100#297: Durable Wisdom in an Age of AI Slop66 / 100#296: Avoiding Major Oopsies: Twyman's Law, Intuition, and Valuing Accuracy Over Precision75 / 100#295: Research and Analytics: the Peanut Butter and Chocolate of Data?73 / 100