SaaS Metrics School · 2026-08-27 · 5 min
Key moments - from our scoring
Substance score
24 / 100
Five dimensions, 20 points each
Traditional ERP dashboards rely solely on general ledger data and archaic formulas, limiting their utility for modern SaaS metrics. Ben Murray argues that AI-powered alternatives, particularly those built on deterministic metrics engines like Software Metrics AI, offer superior flexibility and precision. These new dashboards integrate four key SaaS data sources - financial, HR, bookings, and customer revenue data - rather than just accounting data. The real competitive advantage emerges in customization speed: using Claude, Gemini, or ChatGPT to vibe code HTML into the app creates fully functional dashboards in minutes. More importantly, AI-powered systems allow operators to control period-of-measurement logic; for instance, calculating CAC payback on a six-month basis to match actual sales cycles, something ERP dashboards typically cannot do. Murray also highlights the importance of robust APIs (referencing Saster's API grading tool) and closed-loop data architecture. He describes how AI becomes most valuable at the FP&A process endpoint - generating board narratives and identifying expansion opportunities through rev intel engines that scan for dormant customers - rather than at the data collection phase.
Financial data, HR data, bookings data, and customer revenue data. Traditional ERP dashboards only use financial (chart of accounts) data, missing the other three critical sources.
AI-powered dashboards can be built in minutes by vibe coding HTML into Claude, Gemini, or ChatGPT, whereas traditional ERP dashboards require lengthy canned report development processes.
Closed-loop data ensures data completeness and accuracy across all four key SaaS sources, enabling AI to perform more sophisticated analysis and identify opportunities like dormant customer expansion.
At the endpoint - generating board narratives, creating agendas, and identifying expansion opportunities - rather than at data collection, because this is where subject matter expertise and data interpretation matter most.
A robust API exposes sufficient information and is agent-friendly, allowing downstream analysis and applications beyond the native platform; many APIs exist but provide minimal actionable data.
Our reviewer’s read on each dimension, with quotes from the episode.
There are a handful of non-obvious operational points buried in the rambling - most notably the need for a deterministic metrics engine as a prerequisite for trustworthy AI output, and the argument for customising CAC payback periods to match actual sales cycles - but the episode is so short and stream-of-consciousness that ideas are dropped before they're developed, and promotional asides dilute the density.
this sits on top of my deterministic metrics engine at Software Metrics AI. But the question again, which was great, how's this different from standard ERP dashboards?
CAC payback period. What's our sales cycle? Okay, it's six months. Let me measure CAC on a six month basis. ERP Dashboard's probably not going to do that.
The 'deterministic engine first, AI on top' framing is a mildly fresh architectural take, and the dormant-customer rev-intel angle is interesting, but the core thesis (AI beats legacy ERP dashboards) is a widely circulating claim and nothing here challenges conventional wisdom at a first-principles level.
it's really at the end where now AI can do things even m better than we can ever have done before
my rev intel engine that can look for dormant customers who maybe are ripe for expansion. Right? FP&A is not doing that.
This is a solo monologue by the host promoting his own product; there is no guest whatsoever, and the host's commentary never rises to the level of practitioner depth that would compensate for the absence of an outside expert.
Welcome. My name is Ben Murray
I've been spending a lot of time and also a lot of frustrating time trying to automate my FPA process
Metric names are listed (ARR trajectory, Rule of 40, NRR, gross margin, CAC payback, LTV:CAC, revenue per FTE), and specific tools (Claude, ChatGPT, Gemini, Saster) are named, but there are zero actual numbers, no customer case studies, no before/after comparisons, and no evidence beyond the host's own anecdote about a LinkedIn post.
showed Our ARR trajectory rule 40 net revenue retention, EBITDA margin, gross margin, CAC payback, LTV to CAC revenue per FT and cash balance of course and some trend information
thanks to Saster for their API grading tool, which gets you thinking about is your API agent friendly?
This is an unstructured solo ramble with no guest, no interviewer questions, no follow-ups, and no organised argument; the host free-associates for five minutes and reaches no clear conclusion, making craft essentially non-existent.
So are ERP Dashboards a thing in the past? Well, maybe not. It's going to depend on how they're built
Well, I hope you enjoyed today's edition of SAS Metric School.
Computed from the transcript - who did the talking, and the words that came up most.
Is your ERP dashboard actually built on data that matters - or is it just a chart of accounts dressed up to look useful? In episode #386, Ben Murray breaks down why traditional ERP dashboards are losing ground to AI-generated, prompt-built SaaS reporting and what that means for CFOs and finance leaders right now. If your team is still relying on static dashboards anchored to your general ledger, you're missing three out of four key SaaS data sources before you even start the analysis. The gap between what ERP dashboards can show and what modern AI-native metrics engines can produce is widening fast and the CFOs who close that gap first will be the ones driving the board conversations. Why ERP dashboards are fundamentally limited to chart-of-accounts data - and the three additional SaaS data sources (HRIS, bookings, and customer/revenue data) that actually drive metrics like CAC payback, LTV to CAC, NRR, and Rule of 40. How Ben vibe-coded a full SaaS metrics dashboard in minutes using Claude - covering ARR trajectory, EBITDA margin, gross margin, revenue per FTE, and cash balance - and why a prompt-built report on a deterministic engine beats any canned dashboard.
Transcribed and scored by The B2B Podcast Index.
Speaker A: ERP dashboards a thing of the past. Well, let's find out in today's edition of SAS Metric School. Welcome. My name is Ben Murray so posted a pic of a vibe coded not it wasn't a dashboard but just a report. Just a quick report summarizing some financial data and had a great question response to that post in the in the dashboard pic is and the question was how is this different from the standard ERP dashboard? And that's a great idea and I'll put the link in the show notes to this LinkedIn post. But it was a sample, you could say dashboard but it was just created in a couple minutes using Claude that showed Our ARR trajectory rule 40 net revenue retention, EBITDA margin, gross margin, CAC payback, LTV to CAC revenue per FT and cash balance of course and some trend information and just really quick way to build out some basic metrics. Of course this sits on top of my deterministic metrics engine at Software Metrics AI. But the question again, which was great, how's this different from standard ERP dashboards? And you know, it was real off the cuff. Just my immediate thoughts I said well one, this screenshot, it's built on four key SaaS data sources, generally your ERP dashboard. And if we think erp, probably thinking accounting as a CFO that that's just sitting on top of your chart of accounts. So it's only one of the four key SaaS, uh, data sources and one it's not a dashboard, it was really just built with a prompt. And then also I mentioned this before, it sits on top of a deterministic engine so I can create anything I want really once I have that deterministic engine in place because I can trust the output versus an ERP dashboard just sitting on top of your general ledger data and some very archaic formulas and rules behind that to try to display some useful information Next it's easy to modify. I can change this any which way. And that's why our ERP dashboards are the thing in the past. And as I've been building out canned reports in my app I'm like well why am I really doing this? Because now we can create anything we want. I have a feature where you can vibe code in ChatGPT Gemini Claude, bring that HTML code back into the app and now we've got a custom built dashboard in five minutes. So canned reports probably a thing of the past. Now with my FP and a process, they're always reports. I want my MRI waterfall retention schedule FP and A reports, bookings, reports. So maybe those will still be canned. Uh, but then also why is this better in the ERP Dashboard, I can control the calculations, I control the period of measurements. Hey, CAC payback period. What's our sales cycle? Okay, it's six months. Let me measure CAC on a six month basis. ERP Dashboard's probably not going to do that. And it's also using a robust API endpoint and thanks to Saster for their API grading tool, which gets you thinking about is your API agent friendly? Is it exposing enough information that you can actually do things with it? Just because someone has an API doesn't mean it's actually going to provide some useful information out uh, of that API that then you can take and even do more with it than you can in that application. So are ERP Dashboards a thing in the past? Well, maybe not. It's going to depend on how they're built, what data they're sitting on top of. Is it sitting on top of a chart of accounts or closed loop data? And that's what we need for AI to be really effective is closed loop. And again my four key SaaS data sources, financial H R s, bookings data and customer revenue data. Now bookings data, if you don't speak the bookings language, your self service PLG, we're going to derive that from our MRI waterfall. We still need that to calculate pillar five go to market efficiency in my five pillar SaaS metrics framework. So things are moving fast and I've been spending a lot of time and also a lot of frustrating time trying to automate my FPA process. Then have AI write the board narratives, AI write the board agenda, and so on and so on. And really you could think kind of my closed loop FPA process of where I can insert AI and really it's not at the beginning, right? We don't need AI ah for that. It's really at the end where now AI can do things even m better than we can ever have done before. For example, my MRO schedule. Now I've got a rev intel engine that can look for dormant customers who maybe are ripe for expansion. Right? FP&A is not doing that. But now with AI we can do that because this is where subject matter expertise really plays a part of how can we really leverage the data that we have in place today. So our ERP dashboards dying, gone. Well, we'll see. But it's going to get harder for them to stay up to date. Now some of these new AI native accounting solutions. Now, maybe we'll see. Right. Built on top of modern technology, but, uh, we'll see how that goes. Well, I hope you enjoyed today's edition of SAS Metric School.
Other episodes covering the same guests and topics, from across The B2B Podcast Index.
Paul Graham On Startups, Ambition, and Great FoundersY Combinator Startup Podcast · on ChatGPT88 / 100
The AI-Native Law Firm, with Ryan Walker of General LegalMeeting of the Minds · on Claude88 / 100
How SSW turned AI into ½ their pipeline - Ulysses Maclaren, COO of SSWSaaS Stories · on Claude86 / 100
657. Waziri Garuba, CEO of Harlem Labs, Introducing G.R.I.O.TUnleashed · on Claude80 / 100
Unresolved.cx - When Feedback Has To Matter - Paul TuckerUnresolved.cx · on Claude80 / 100
321 | How to Build an AI App Without Coding: From Idea to App Store with Replit & Claude by Bryce Rattner KeithleyLeveraging AI · on Claude80 / 100
How Much Are Tech CFOs Actually Making in 2026?50 / 100The Latest GRR Benchmarks66 / 100Why AI ARR Alone No Longer Lifts Your Software Valuation52 / 100Here's What Separates the 9 Public SaaS Companies that Trade Above 10x63 / 10012 Steps to Creating an Outcome-based Pricing Plan63 / 100