00:05:34 · Steve Stagner, Miguel Armaza
And I worked on a couple of platforms that are very similar to what we're doing today in the HVAC space and the roofing space and the veterinarian clinic space. And that's where I met Jake and Frank who are the founders of Crete. And so really have a broad background here to help us navigate the structure and the strategy and the scale needed to build this dream that we have. [00:05:57] S1: We've talked in the past and you mentioned a core principle for you is people, process and technology. [00:06:09] S0: Correct. [00:06:10] S1: Can you double down, like you share kind of why have you developed? How did you arrive to this process, to this order? How did that apply to your retail background with 10,000 employees? And how does that apply to today?
00:06:28 · Steve Stagner
[00:06:28] S0: Yeah, great question. So it's interesting. I've been through moving from Excel level, Excel based inventory systems to all the way signing a deal with Steve Ballmer with Microsoft Dynamics on an enterprise system with a public company and thousands of stores. And it's interesting in every single case where we've done some sort of big technology integration, what happens is it starts off with excitement that we're going to get this new system and it's going to work great. But then very quickly you end up in the valley of despair, which is what happens is it's a lot of work just transforming and people get frustrated. It's mainly because people have ingrained norms in the way that we do things. And when you change, change management is really the art of it. And so for example, when I was CEO of Mattress Firm, we'd already taken the company public and we launched an enterprise system, signed a deal with Microsoft Dynamics and spent like $12,000,000 on it, started launching in the first test markets And what happened was my people were very frustrated. There was a lot of conversation and chatter around how we made a big mistake. And so I sent a team out there and we all went out there and got involved and very quickly realized surprise surprise it wasn't the technology, was the training. And in other cases I've learned that bad data plus other good data doesn't equal great data. So to your point I've just kind of developed a formula of people process and technology that the technology is an enabler and it's always great. But if the people don't understand how to use it, they're not going to use it. And if the data is not clean and mapping very clean and clearly, then it's not going to be productive and give us the quality of the result that we want. So what I'm applying here at Crete is we're on the precipice of the greatest transformation of technology in probably the history of the world with AI entering the space. And so it's super exciting, but as we start putting these tools in, what I'm really challenging our team to do is to make sure that we're, to make sure that it's done in a way that will be adopted and with quality is that we're making sure that we're process mapping everything, that we're bringing people along, we're testing small, we're learning and we're adopting. So I call it crawl, walk, run.
00:08:54 · Steve Stagner, Miguel Armaza
We're crawling before we walk, we're walking before we run and then we're running. And that's really the approach that we're taking and over my career that approach has been super successful. [00:09:06] S1: A lot of people are excited about AI but also a lot of people are afraid especially as it relates to the labor market and the consequences that AI might have, this new LLM wave, of course. But you're saying that technology, including AI, is just an enabler, right? I call it an amplifier and not so much a replacer. How is it different this time? If not, why? And if it is also, I'd love to hear your point of view.
00:09:34 · Steve Stagner
[00:09:34] S0: You know, look, mean, I think if you go back over time, when the printing press came out, there was concerns about how that would change the world and the production line and the internet and now AI with the newest technology here. AI doesn't invent anything. What it does is it basically predicts and it predicts based off of the data that it has. I think that there will be humans. I'm not going to say here that I don't think that humans won't be at risk of losing some jobs. I think that humans that don't embrace AI will be at risk of losing jobs. And so I think that you have this PhD in your pocket now. You have this opportunity to kind of leverage technology as a thought partner and then do the more valuable work that is possible. So if you take like tax returns, over the past three years I've had a pretty complicated tax return. It's roughly 500 pages. And I have a top 10 firm tax preparer that is a partner that works on my account. I won't mention the name or the firm. But as a client, what happens every year is I get a list of things to do to turn in and I get a bill and I pay that bill and it's expensive. What hasn't happened is any tax planning. And this is a partner in a top 10 firm. And so I do sense the pressure coming up on April 14 of getting everything turned in so that we can make the assessments. But what I do think is possible is that so much of that work, whether it be on the business side or the personal side can be automated because just transferring numbers is not necessarily value add. And if AI can be a tool that allows people to shift the mundane work because the flip side is they're also very stressed. They're going through a lot of stress trying to, it's very time based as we know in this industry, it's very timeline based and they're very stressed to get everything done and then to get it done right and there could be human error. And if leverage you AI to take all of this data, normalize it, validate it and tick and tie it up against the current laws and everything and then allow the professionals to do what they do best which is to start focusing on the insight. And the exciting thing about where we could develop the partnership and grow the business or grow or do tax planning on the individual level. I think that that's really what is so exciting about AI.
00:12:03 · Steve Stagner, Miguel Armaza
But if you look at it as something that's just going to take away what you do, I just feel like you're looking at it the wrong way. I think we've got to, I think we all as a society have to look at how we can leverage it to create more value and do more valuable things. [00:12:21] S1: So let's bring it specifically to Crete, right? Yeah, sure. That is the segment that you are at least initially focusing on and that's accounting firms, right? Have been the quickest wins that you've identified that you're starting to transform?
00:12:38 · Steve Stagner
[00:12:38] S0: Yeah, so if you zoom out, if you think about the framework, what we're trying to do is go in and look at actual technologies and systems that they're already using and then go in and see if we can accelerate the mundane activities inside those activities versus just creating like an AI agent necessarily just to do all the work. So we're trying to make the existing tools more productive. And so one way, like take for example, you could take the trial balance. So we could go into a business and we've developed tools that can go into a business to take messy messy data and a trial balance and normalize that data which becomes the foundation for doing the business tax returns and doing the adjustments. It comes to the foundation for doing audit testing and then opens up the door for the advisory work. So if you think about it, if we map the messy data in such a way that we could probably save 30 to 50% of their time and some of our early tests are showing exactly that 30 to 50% time savings on mapping that data and normalize it so that they're focusing on, okay, this is all accurate, this makes sense and they're focusing on the intuition and the quality of it. Then when we get into the audit testing, now you can pull that data in there very, very quickly which audit testing is pretty straightforward. So to do the substantive testing that is required, that can be done in a very fast way, probably save 25 to 30% of time so that the human can do the evaluations on what is not accurate or not or is different. And then when we go into the tax software, once we've got the audit testing done, can, as you know, much of what's filled out on a tax return is very, very straightforward. So we can just pull that data in there, automate that, and then the tax professional can even, I think the next game changer is to work with AI to look for and spot based off of tax law and based off of other business insights and look for areas to advise the client in a clever way as a thought partner. Of course, you would never do that without validating and verifying because you don't wanna rely on it 100%. I don't think it's there yet. But if it can save you time on all of that mundane work, I think it does a lot of things.