S12E6: Tate Talks - With Pete Hughes, PSA Insights

Hi everyone, welcome back to Tate Talks, the finale of Season 12 actually, and it was a great episode with my old friend Pete Hughes of PSA Insights. Hope you enjoy. Hi Pete, thanks so much for joining me today for Tate Talks, the final episode of Season 12, so let's make it a good one. How are you doing?

yeah good Chris thanks for having me here especially on your your final episode hopefully we go out with a bang yeah save the best till last as they say so yeah really really looking forward to this mainly because I've known you a long long time in MSP terms not that I can exactly remember the times we first met but you may do and maybe we'll touch upon that as we get along but before we get going for people that don't know you and haven't come across Pete Hughes in the MSP community.

Can you maybe tell us a little bit about yourself? Yeah, sure. So I have been working in this industry for probably 20 plus years now. Started off as a sort of internal techie, IT manager with some internal business and then moved into a more consulting MSP role.

but always had technology and the technical side of this as my background. And yeah, I've been working for MSPs for 20 odd years and more recently building out MSPs and growing them and having lots of fun doing it. Yeah, well, it's quite a long time, quite a long career in the MSP space. And just thinking back, as you were talking there from when we first met, I think this was probably at the stage when Orchid IT, which is where I worked in Derby, were looking to or had recently transitioned from GFI Max to Enable.

And you were with TaylorMade at the time and you were using Enable. So would that kind of be around the time that we first met, do you think? Yeah, I think so. Yeah, I think that's right.

Yeah, we were using, yeah, I think it's through Enable, probably one of the Enable events, the many events that we've been to. And yeah, I've been working with Enable now as a partner of those for almost all of that time. Yeah, OK. Yeah, some good times back then.

So, yeah, I'm pretty sure I can't pinpoint the exact moment, but I know that was something that we did meet. And I spent lots of happy times down at the old tailor-made offices on the South Coast, which was always a pleasure and something that I always look forward to. Yeah, you always enjoy the South Coast, Chris. And if I can, it's nice to get you to take you out on the water whenever possible.

Yeah, we'll hopefully touch upon that later when we talk a bit about what you do to downside. So you touched very, very briefly on your sort of career path there. We'll talk a little about what you're up to now, which I know you've got some sort of exciting new stuff that you've been working, beavering away at sort of quite feverishly, quietly for the last few months, which is I'm really, really excited to hear more about. But so what got you to this point over the last few years?

So, yeah, I think a big part of what's happened over the last probably 10 years or so is the main change that all MSPs are clearly aware of, that move from providing services from a time materials basis up to the vast majority of service providers now offering fixed price services. And the big challenge over those years from a technical point of view of how to deliver that and how to deliver that efficiently, effectively and and profitably and um and so yeah i've spent um many years working with companies like enable um caseya um and dato you know on how can we deliver services most efficiently and building out automations and efficiencies into the services that we as MSPs are delivering into the customers and primarily that's obviously around making sure that the services are safe and secure and you can ensure that they're not costing the earth by ensuring around efficiency and automation but also delivering a great service to the customer And a big piece that we've been working on over the last 10 years is about delivering real proactivity and making sure that that proactive work is more than just an automation that restarts a service if something is broken.

but actual real proactivity in terms of getting out to the customer and turning around that relationship with the customer to make sure the customer knows you're there to help them prevent issues occurring in the first place. because right that helps them as a business to operate and be more effective lawyers effective solicitors dentists whatever they are but from an MSP perspective the less tickets that are coming in on fixed price services the more money you're going to make and yeah that's been a really challenging and interesting journey to to help deliver that and and what really clicked for me quite recently I suppose is you know with the advent of AI over the last few years is is really there's now an opportunity to fix this kind of properly and rather than you know trying to find those patterns in in systems and and trying to identify those problems manually AI is something that can help make this more efficient.

And that's really what's happened for me over the last year is, you know, how can I use this new amazing technology that's being thrown at us all for the best and to help us all grow. Yeah, thanks for that. I made a few notes as you were talking, one of which is around sort of proactivity, because that was, as you probably remember, the birth of the MSP model, I guess, as people were transitioning from time and materials or break fix. That was the buzzword, right?

And effectively, a lot of the time it meant, as you pointed out, maybe just having an automation to restart sort of spoiler dot XE or something on a machine. It wasn't massively proactive in the real sense, but it was something that has definitely evolved. And the other point you mentioned around profitability has been a constant point within the MSP space of ensuring contracts are profitable. But you mentioned, I think, how things have changed.

And it's funny timing-wise, because you mentioned you've been in the industry for 20 years. I've been considerably longer than that. And at JumpCloud this week, we launched sort of a Gentic IAM, sort of Identity Access Management. and and just I was just thinking about I wrote on LinkedIn the other day how things have changed where we used to have to manage human beings that was a person in the chair is really straightforward and then devices so servers or workstations but now we're in a position where we've got to try and control non-human identities and agents as well which if you'd have said that to me even five years ago I'd have kind of said this it's just pure madness it's just it's just not happening and you know it's just science fiction but I know that you're someone that's certainly way ahead of me in terms of the um AI curve and kind of being an early adopter and someone that's very very involved in that space you probably saw this coming more so than I did but it's been it's been kind of a huge transition for our industry and I think that like we went from break fix to MSP and, you know, sort of time materials, fixed fee, all these different things.

But the more and more AI that's coming into the workplace is presenting a challenge and an opportunity for MSPs Do you see it that way as well Oh 100 I think it is a challenge in sort of two ways really It a challenge for MSPs in terms of being able to remain relevant profitable and from a commercial perspective, be able to be competitive. because MSPs that are adopting AI and adopting AI into their tool sets, whether that be through custom solutions that they've written themselves or through the adoption of the number of different AI products or where vendors are building AI into their solutions.

And MSPs that are addressing that are able to drive far more efficiency into their business. And potentially that's going to affect their profitability. And MSPs that aren't doing that are going to suffer. And the other side is businesses and their customers want to know about AI.

they often don't know how or what they need to use AI for but they know that it's this this big thing that that everyone's talking about and and and there is something that we need to do to stay relevant and I've had a lot of conversations with business about that and and And the way I generally unpack it for business is to say, look, let's not specifically start with technology. Let's talk about the business challenges you've got. Let's identify how you operate from left to right.

Let's find the efficiencies and inefficiencies in your processes and allow me to then translate that into where AI services might help you. And business owners are then able to go, OK, right, well, that's a way that we can take this forward rather than saying, oh, well, there's this chat tool. Maybe we should just be using that to write my emails or send letters to my customers, et cetera. Taking that kind of approach has been what I found has been far, far easier.

But MSPs need to approach it from both directions, building out AI solutions within their business, but also making sure they're talking to their customers about how they can use AI services, because if they don't start talking to them, someone else will. yeah 100% agree with both those prompts and those approaches and I've often spoke on this podcast I think about how it takes me back and again I am that old that I remember when the kind of internet first came along and people were saying to businesses small businesses the same thing they were probably ringing up our MSP although it probably wasn't called an MSP then saying we've you know we're seeing constant talk on the tv about um or the newspapers remember those um about things like um the world wide web the internet how it's going to transform businesses and people would call us and say well how is that going to affect our 10-man plus merchants which is a sort of organization i always i always reference and because they kind of knew it they knew they needed to do something they were hearing the noise but didn't know where to start.

And I think that is almost identical with what's happening with AI, except it's happening much, much faster. And it's probably got much, much greater scope to change the way they operate their business. That people are seeing huge enterprises talk about AI, the TV commercials are talking about AI. And this plumber's merchant owner is probably sitting there thinking, well, how can I get AI into my organization?

They're probably working with an MSP to keep their printers going and their servers running. But they probably need to be having those conversations. And the point that you made, I absolutely agree with, which is not starting with the AI or the outcome. It's trying to understand more about their business processes and then maybe break them down and see what maybe could be automated or changed along the route.

So I think it's absolutely critical that MSPs are doing that. For the exact reason that you mentioned, I noticed you've mentioned the word twice which i thought was very um sort of insightful which is relevant i think msps if they're going to stay relevant they need to be operating on both of those sort of prongs that you said one to increase the service two to be able to and we'll touch upon this i think hopefully when we get onto psa insights two to be able to maybe demonstrate the service that they're offering to clients and to show the value that they're offering to kind of So it's always been a challenge for MSPs.

You know, if it's showing the value, if things are working all the time, people think, well, we can cut costs by stop paying this MSP because we don't know what they do. I think you can help with that. But also adding value in terms of kind of going in and becoming more involved in clients' businesses than I think MSPs have ever been before, whereas it's almost a seat at the table now in terms of trying to get inside all of those business processes, touching all areas of the business to try and help with efficiencies, which is a fundamental change to how MSPs work as I see it.

Yeah, it is. I think it's a big pivot. The provision of IT support is a given. It's something that you have to deliver.

And we've had, as a modern MSP, you know well then i'll take the 20 years that i've had to to get that right um that should be right now you should be able to deliver it um it's now a pivot to having the different types of conversations with your customers they're expecting you to deliver the support services you now need to deliver those additional services around understanding their business and understanding how technology can help their business. I think one of the challenges I've spoken to a number of MSPs about is that they say, well, our customers just don't know that's what we do.

You know, I think there's a big marketing piece there to almost pivot your MSP around that point and try and demonstrate how you can be that service that helps their business, not just fixing their printer problems. And I think it's something I've spoken about at an event I was at recently, but I said half of the MSPs that I talk to don't know how to go and have those conversations. half of them aren't using AI in their businesses already. And so they need to be able to use AI internally before being able to go out and sell those services.

But it's got to be done because AI is not going to replace MSPs for sure, but MSPs that use AI will replace the ones that don't. So they need to make sure they're not in the second group. Yeah, that's a nice way of putting it. And yeah, again, absolutely makes sense to me.

But one of the challenges that MSPs have, and I think that this is something I'd appreciate your sort of thoughts on, is around how are they factoring, how are they pricing for this? Because what a lot of MSPs have done in the past, they've moved to this sort of fixed fee model, which is trying to bundle everything in for a fixed fee and then automate as much as they can to bring down the actual cost of delivering that, which is increasing profitability. But when they start to get involved within business processes and consulting and kind of getting under the hood of these clients, how are you kind of seeing MSPs currently charging for those services?

that okay so i don't feel like that is that is proven yet as the best model that we're all going to be using in five years time to deliver i think everyone's doing it in a slightly different way and i think all these ideas are up there in the cloud they're up there and they're going to converge in the end um my my careful i might say my prediction what i'm seeing is msp's trying to sell fixed price ai services so delivery of strategy pieces strategy consulting you know business consulting to work out where ai would fit into a business and then delivering fixed price pieces of work to build out agents and automations and efficiencies into that business and then charging you know the typical type MSP service on a fixed price to support those delivered services and yeah potentially on the number of services the complexity of them could depend upon the price or the size of the size of the business you're supporting and then seeing other msps that are turning around going well you know we'll own the automation and we'll deliver it to the business and i'll charge on a per use basis so every time that automation fires that's a cost and um you know there are there are so many different ways to provide this and i'm not too sure exactly where this is going to land um what i do know is that the the use of AI services in businesses becoming easier and easier to adopt.

And it's been said before, and it's not my saying, from a Leon actually at ThinkCloud mentioned that, you know, creating agents in the future is going to be like creating PowerPoints. the end users are going to be implementing these agents and creating them. It's not going to be the MSPs. It's going to be our job to make sure that they're doing it the right way.

They can adopt it. They're trained. They can build them. But the vendors are going to allow the agents to be able to be created so much easier.

and yeah maybe one day it'll be about just opening powerpoint and creating your agent like that and you've seen that already through services like um na10 was and it still is incredibly powerful tool but some of the natural more the llms the natural language models are now building in methods of being able to build automations just straight off the bat within within their products just by saying, giving it a natural language prompt and using MCP to be able to build out those integrations directly.

You don't need to be an NAM developer now to integrate one product to another. You just need to be able to ask the AI in the right way. Yeah, it's fascinating. And I love the PowerPoint analogy.

And again, I think you're right. It's going to be something that people are just doing themselves rather than the MSP doing. And MSPs can help with the policies, the guardrails, the training, all of the things to sort of help secure them. But ultimately, it's going to be user-driven, which is fascinating.

So that leads us on a little to what you're up to now, which is I'm really, really excited to hear more about your current thing that you're working on, PSA Insights. And going back to the two pillars that you described for MSPs, whether it comes to AI, does this kind of straddle both sides of that or does this kind of lean more into the MSPs' internal processes? Or is it, mine's, I've had a quick look at it already. I kind of think it's got a foot in both camps, but partly obviously sits within the MSP itself.

But I guess before we get, I'm digressing, before we get into any kind of detail like that, can you maybe just, and I know this is brand new, so I'm really excited to tell people about this, because I know you've only spoken to a few MSPs so far. But can you tell us a little about PSA Insights and what problems you're trying to solve for MSPs? Yeah, so PSA Insights is a tool that works with your PSA to drag out data and identify patterns and analysis that helps an MSP be able to deliver more efficient services.

You know, tickets in your PSA are pretty much the most honest record any MSP has of what they're doing, what the activities they're doing, where their time is going, what their engineers are working on, the problems that their customers are having on a day-to-day basis. there's a huge amount of data in the PSA and yes it does depend upon on how good the data is in there and you know we we have a facility within insights to help identify whether the data is good as well so you can work with your engineers to to improve on it but the problem has never been that the data hasn't been there it's just about surfacing it and so like i said sort of at the start of our conversation for me this was about a kind of a pivotal moment i realized you know i spent all this time trying to understand this data and so i spent two decades watching other msps fight the same battles around we constantly have these problems occurring or or we're losing money over here, or this customer is costing us way more than they should.

And the use of AI is now allowing us to be able to analyze the data that's in the PSA and surface that to identify where commonality is occurring. So if your customers are regularly having issues with printing, issues with security, issues with user onboarding, offboarding. It could be issues with a specific piece of software that is regularly crashing. And you might not necessarily as a business owner know that this is happening because you're depending upon your team to identify those activities, surface them and then deal with them.

And no matter how hard I've tried in the past, trying to get a wider sort of larger MSP team to be able to, you know, continuously focus on that continuous improvement. It's been a challenge. And that's why I built PSA Insights because it's building out a solution that surfaces that data and helps MSPs find common problems that are costing them money. it connects to the common ms common psas halo autotask connectwise and it's the tool i wish i'd had 10 years ago that's interesting because i just think i was about i was thinking exactly the same because i wrote down um profitability reports and i guess you remember But the way I do that in the early days of PSA, that was something that we're obsessed with generating profitability reports for clients to see where we're losing money.

And then maybe having to have those conversations with the clients to put the price up or fire the client or whatever we did. But we could never really get the real insight apart from very high level ticket categorization. What was causing the problems? Why these contracts were so unprofitable?

and so you're right with my sort of orchid IT hat on this would have been invaluable to be able to dig in and understand what was going wrong with these contracts and then how to how to address it and it yeah that's right and it and the tool works very much for across the board so you can be the deep technical user within the business and want to dive into the analytics of exactly how to go and solve a common problem and so we analyze all the ticket data descriptions ticket titles all the time entries and notes and we're identifying the commonality and then using ai to provide a remediation plan so that can be as technical as you like but equally for the executive or the msp owner to be able to have a dashboard that identifies these other common problems but also be able to use the AI chat tools that we're used to using and ask it questions with natural language.

Where are the common problems I'm having this week? What happened last week? What happened in the last 30 days that you suggest that I deal with so that I don have to deal with them next month or identify data that I can use for the next QBR that I'm going to have for my customer. Being able to use the AI chat to drive that knowledge and efficiency is just ultimately from the background, just helping the MSP deliver those services.

Yeah. And again, it's having access to that data. The possibilities are endless. Like you mentioned around QBRs, that's a fantastic use case to be able to generate that data and reporting before you go and have those conversations with clients.

I think it's fantastic. And I was lucky enough to see a short demo of the platform. And, yeah, to be honest with you, I was sort of really, really impressed. So well done with what you've done so far, Pete.

And I know that you've had conversations with MSPs already that have started to look at this and MSPs that are using this currently. But for people listening today, if they want to kind of get more info, maybe take a look at it, what's the best way of doing that? obviously we have a website um which you can um access at psainsights.ai i hope that all of the information on there is is pretty self-explanatory as to what the product does but directly from that page you can try out the product and a live demo uh with um with fake data so you can actually see how the tool works and how it would surface your data.

And, you know, I'd encourage MSPs to book in a demo and we can guide you through exactly how the solution works. Brilliant. Well, I'll put the website links and your LinkedIn and everything into the episode notes. If anyone wants to get in touch and I would highly recommend that they do, they'll be able to do that.

But I'm really looking forward to seeing where PSA Insights goes, because even the version I've seen, which is early days, is incredible. I'm just like you said earlier, it's kind of the tool that I wish I had when I was sort of running service desks back in the day. So I'm sure that people would still get huge value from that now. So good luck with that.

I appreciate that, Chris. Thanks. No problem. Right.

OK, well, let's move away from the product a little and move away from MSP a little, actually. And think more about Pete Hughes, the person, the man, rather than the MSP expert, as you're often known as. So what do you do when, obviously, I know sort of firsthand how much time you've been put in with the team on this product over the last few months. So you probably had very little time to do anything else.

But when you're not sort of building new software products, what do you do to relax? Do you listen to podcasts, books, TV, movies? What do you do in your downtime? I clearly listen to take talks every time that comes out to relax.

When I'm not at my desk, I'm either with my kids or I will be out on the water sailing. um sailing has been a big part of my life for 40 years um it's something that i used to do with my family uh when i was when i was small and something that i think i'll do till my last days um so all the way from you know i sail from small boats on the river in in devon um through to you know big super yachts and now my own small yacht that I can sail with family and friends around the south coast of the UK and hopefully one day a little bit further but that's that's what gets me away from away from AI away from away from my desk no one can catch me out there and that's where I go to to think about the next steps yeah it's a fascinating hobby and um i guess it's one of those that uh it definitely takes you away from thinking of ai when you've got kind of more important things on your on your mind when you're trying to um uh let me put the correct phrase uh drive uh steer um navigate whatever you do when you're on when you're on your yacht to get it from a to b safely and they'll say helming but all the cool people still say steering okay there you go okay well that's uh that's a fantastic hobby to have and i remember uh very fondly on the uh the time that you took me out once on um on one of your uh your boats and a little jaunt over to the isle of wight from uh from hamble which was uh very memorable for many reasons a little bit wet i think yeah if not yeah just for my legs have just about dried out and that was uh that was some considerable time ago but it was great fun yeah well we'd do it again absolutely we should do it again so yeah i look forward to that and thanks for saying you listened to uh take talks with the customary uh laugh at the end which everyone else does that answers the question in the same way but uh yeah i appreciate it that's one of the reasons i have so many guests on i think you know if if all the guests listen to the show that's going to increase the numbers that's my my general thinking perfect so um we're down to the end we're down to one final quick question which i ask everyone um about kind of when you're speaking to msp as you mentioned you've just returned from an event um so you must have been chatting to msps of all sizes all flavors um in the bars or in the corridors or in the conference rooms um what are the sort challenges that you're continually hearing and any kind of thoughts how they can sort of overcome them?

I think the biggest challenge that I'm hearing and I think I might have already covered it a little bit is around MSPs knowing how to sell and provide AI services in this new world and stay relevant in this new world. And I think, I suppose my one piece of advice for MSPs would be before you go and sell AI to anyone else, you know, ultimately need to use it on yourself. Pull your last 90 days of tickets, find 10 things your team has fixed more than five times and fix those properly.

if you can't run AI proactively within your business you're going to be less efficient have less ability to be able to go and have those conversations with your customers but that said don't over complicate the conversations with the customers keep those conversations as business conversations and the technology will follow So that's where I feel the best approaches. Yeah, no, and that's great advice. And if I knew how to do it, I would do a word cloud of this episode because I'm pretty sure that relevant would come out as probably one of the bigger words there.

And I absolutely agree with you. I think MSPs have got a challenge around relevance. And I think that taking on some of the sort of thoughts that you had, starting to use AI more within their own organizations and having those conversations with their clients will certainly help them with the stay relevant, which is what we all want to do, what we're all striving to do in our MSP business. So with that, thanks so much, Pete.

It's been a real pleasure to chat as ever. I'm really, really excited to see where you're going to take PSE Insights. I'm also really excited to see where you're going to take me the next time we go out on the water. and hopefully I'll see you again soon.

But thanks so much for taking the time to join us today. I've really enjoyed it. Chris, thanks so much for having me. It's been great to chat through all this.

It's always a pleasure. Let's catch up again soon. Yeah, cheers. Bye.

So there you have it. That's a wrap for season 12. Thanks everyone for listening. It's been fantastic and we'll be back in the summer with season 13.

But until then, take care and goodbye.