Technology expert Randy Johnston, executive vice president at K2 Enterprises, shines a spotlight on what accounting tech can do for you now — and what it's going to do for you in the future.
Transcription:
Dan Hood (00:03):
Welcome to On the Air With Accounting Today. I'm editor-in-chief Dan Hood. With the cloud firmly established and artificial intelligence on the brink of potentially changing everything we know, accounting software is at a bit of an inflection point, and that means it's a good time to take a step back and get a broad view of the state of accounting technology, what it can do for you now and where it might be headed. There are a few people better positioned to provide that kind of view than Randy Johnston. He's executive vice president of K2 Enterprises and one of the foremost thinkers and experts on accounting technology in the field. Randy, thanks for joining us.
Randy Johnston (00:30):
Dan, thanks for the invitation.
Dan Hood (00:32):
This is a huge topic I know, and we're going to try to cover it in 20 minutes. So again, wildly unfair to you, but we're going to cram a lot in here. Let's start by saying it seems, and I haven't actually counted, but just sort of anecdotally, it seems like there's more accounting software products on the market than ever before. Why is that and is it a good thing?
Randy Johnston (00:50):
Well, there are more than ever before, and it's because cloud enablement has made it easy to develop a new platform and you don't have a lot of the infrastructure distribution costs and so forth. So if you can afford some A W s or Azure or Google hosting, all of a sudden you can have a platform up and running with a lot of leverage, and there's libraries that you can work from. And the old rule for me has been find a need and fill it. And that's exactly what's happening across not only the core accounting software products, but the add-ons. And I know in another podcast episode, we'll talk about tech stacks, which is probably more on that line, but bottom line here is the development costs are not over the top. You can build a really good accounting software product for less than a million dollars, but it'll take you 10 million to market it to the US alone. So if you look at the opportunity and you say, well, here's an opportunity, if you can scurry up a million dollars, which used to be a lot of money, but if you can get that, you can actually build a pretty doggone good platform in a relatively short period of time.
Dan Hood (01:58):
So the barriers to entry are pretty low. And as you say, you don't have to mail anything, you don't have to box anything, you don't have to. It's astonishing how many of the things that used to take up a lot of the cash of a tech startup have just vanished with the cloud and with the new distribution channels and new methods and that. So that makes a lot of sense. So here's the question. Is this a good thing?
Randy Johnston (02:20):
The good thing question is, it probably is because now I'll go to the dark side of this, the negative side, a lot of the platforms that are out there were designed over 20 years ago. There's only a few products that are actually modern, and I consider that a problem, number one. Number two, the things why it might not be a good thing. Many of the developers of these systems actually don't understand accounting, and that's a head scratcher to me. It's like
Dan Hood (02:48):
That's a problem.
Randy Johnston (02:49):
You don't understand double entry accounting and you don't understand what a ledger is or supporting transaction journal, and you're writing this for why. See, a lot of these systems are really operational as opposed to having a pure accounting mentality. Now, I don't have a problem with operational systems. We need 'em desperately, but so much of the time, this problems that are being solved are so narrow that you can't use the platforms in general. And part of the issue for those of your listeners that are in public accounting, if you're trying to build a client accounting services practice, there's needs that you have as a firm and needs that the end users have, as opposed to if you're an industry listener, you may have a particular vertical that you're in, manufacturing, distribution, construction, healthcare, whatever, and you have different needs as well. So the issue here is if we're using a general platform, does it solve the problem complete enough? Now, I used to Dan run with the 80 20 rule. If I got 80% of it, I'm pretty good, but now that we're in our third, or some people would count it, fourth generation of accounting software replacements, buyers are much more fussy. Promises from vendors are much less accurate. In other words, they're inflated and it's truly a buyer beware period. And that may not be a good thing too.
Dan Hood (04:11):
So I mean, to a certain extent, we might want to wait for some of these platforms to be shaken out. See which ones actually last for a bit. Because thinking, the last time I saw that we saw this many was in the late nineties and the very, very early two thousands when there was an enormous consolidation and it went from 20, 30 different packages down to something like four, they all got bought up. But it was interesting that point you make about, there was two approaches and there was, I remember at one point the two big players in the field, one was famous for being good technology built by engineers. So the engineering was great, and the other was built for accountants. And so it looked like accounting. So accountants understood that one much better, but it wasn't as strong as the other one because that had been built with good technological underpinnings. I'm wildly oversimplifying that, but that's an issue now is that are these systems what you need them to be? And that maybe brings me to my next question is, and I realize again there's a lot of different packages, but where do you think they are in terms of are we really at that point where, as you say, buyer everywhere and let's wait and see what shakes out or what's the state of accounting software where it needs to be?
Randy Johnston (05:20):
Well, at the risk of hurting new startups, I would suggest you want to ask how long they've been in business and what the installed base is like. So it's got to shake out a year or two. And my old rule of thumb is you had to have at least a hundred installations before you considered any platform. And a lot of these businesses are sustainable with 10 million in revenue, you've got enough money, you can do the development, you can do the support, and you can continue to expand the product. But what you'll discover as you're trying to analyze these systems, they all kind of sound alike. But the other thing that's also, even when you ask about future roadmaps, a lot of things that used to be almost locked down when you heard a future roadmap, now we're thinking about doing this roadmap item, but we may or may not ever build it. So it's become far more of a burden hand versus two in the bush type of problem. You've got to have the burden at hand, it's got to be in the existing system, and it could get deprecated. You could be using a feature that the publisher decides is no longer needed and the feature just goes away. So deprecation is actually becoming more of an issue too.
Dan Hood (06:32):
Gotcha. Well, so this brings up a different way of approaching this question, which is to say, do most of these products do the basics, right? If you buy any accounting software, can you be relatively confident that it will do, I don't know if you want to put it in the 80 20 rule the way you described it, or is I can be pretty confident I'm going to get my basic accounting done, or if it's a tax prep system, can I be pretty confident that I'll be able to produce a 10 40 relatively accurately and without craziness?
Randy Johnston (06:59):
Well, I'm going to fork your question because of its accounting, general edger, AR a P pretty much is going to work in all these systems. Will there be some deficiencies? Ask, but is most of it going to get covered? Yes. Or is the reporting going to be poor? Generally, yes. Now, if it's something more technical, let's just say tax, like income tax preparation or sales tax, those are a heck of a lot more complex because the Byzantine systems that we have in the US and across various states. So the problem is it is doggone hard to get everything right. And even the very big boys, the Thomson Reuters, Walters Klu and so forth, they have issues commonly at the state level. So somebody who tells me they're going to be building a new tax software, whether it's income tax or sales tax, I usually refer to Afghanistan. It's a place that software developers go to die because they do not understand how complex their attempt is and that those guys are going to hang out in the hills, rock treason everywhere and surprise you.
Dan Hood (08:07):
Well, something like eight, no one knows how many tacking jurisdictions there are in the United States. Every time I talk to somebody there's somewhere between eight and 12,000. So yeah, that's a huge ridiculously complex sort of thing. So from that point of view, counting suddenly looks a lot less complex. But so as we look at if in accounting, let's narrow it back to accounting. I brought in tax just because it fascinates me, but we'll narrow it back to accounting and talk about that and as see if you can basically be relied on, it'll get your accounting done. What trends are you seeing them working on now? Are they, where are they going from having the basics set up?
Randy Johnston (08:46):
Yeah, so I am going to talk about what they're avoiding as well as what they're doing. Fair enough. Because one of the things most of the developers avoid is the stuff like inventory and costing. Okay. And where they're going is trying to monetize transactions. So taking the discount rights on payments and things like that. And with the new deployment of Fed now here in July, I think the economic position of trying to make money on transactions, the dollars are going to change on that. But what they're trying to do in general when they're developing these systems is to expand their footprint to more users and they're trying to generally make the products easier to use. And then number the next one up I think, is they're trying to make the interfaces to other systems easier to do so, either done with the application program interfaces or APIs or through a digital plumbing tool like Zapier or Cdata or Booming, one of those types of tools because they've concluded they're not going to build a monolithic system, they're going to instead try to get by with a bunch of third party.
(09:59)
And Dan, to your question, that is actually a real key differential of the entry level products, the zeros, the QuickBooks and so forth compared to the mid-markets. The mid-markets have a lot more of the functions and features built in. They still might need third party products, but the entry-level products are depending on all of these add-ons, which add expense. So you have your base subscription and then you have a bunch of add-ons. And what we're finding is that entry level plus add-on is often more expensive than mid-market. And a year ago, Dan and I started using the term, maybe you saw me doing it, process resource operation planning prop as opposed to E E R P. And these systems are a little bigger than the entry level. They have more features in them, but they don't have as much as a mid-market E R P does. And I think a lot of small businesses need those prop capabilities, which include inventory and costing, but also workflow processes. So workflow processes, another thing that's being built into a lot of these systems. And it used to be that workflow was only in the big systems tier 1, 2, 3, 4, 5, the great big systems, but we're seeing it now in four and five, and that's another big change.
Dan Hood (11:13):
Excellent. Alright, this has been a great picture of where we are. I want to take a quick break, but then when we come back I want to talk about a little bit more about where we're going. You pointed the way, but we'll go more deeply into it in a second with Randy Johnston. Alright, and we're back. We're talking with Randy Johnston of K two Enterprises. We talk about the state of accounting software, and you sort of hinted at it, we're really focusing more on the mid-market and the small, there's obviously a whole universe of high level ERP stuff that we're not going too deeply into, but in that universe maybe we talk about where we see things headed in the next several years. On an earlier podcast, we talked a lot about AI. I think everyone probably expects something like that to pop, maybe to pop up here, but where do you see accounting software tools headed in say, the next few years?
Randy Johnston (12:00):
Yeah, so this one, you are correct. There will be AI features added into a lot of these systems, but we're also going to see a lot more automation of transactions. Now, let's face it, we've got a labor shortage, and with the labor shortage, we're trying to automate more of these systems. And I believe what you're going to find is that items that used to be data entry will be auto extracted. Now we already have tools that do that type of work. Sage auto entry, the DExT, formerly Receipt Bank, but now called Dex Prepare Vic AI and products of that ilk. And there's about five of them out there. But you've got to begin to understand that almost all transactions will begin arriving in real time from possibly as far out as internet of things devices, and that we will not be keying transactions into our systems. So that's the big path forward is how do we get labor out of the systems, but still get the operational and financial data that we need to manage these operations.
Dan Hood (13:06):
I think a lot of people are excited about that possibility, as you said, given the talent shortages that are going on. So that's one major thing to expect that automation. Anything else, anything else that accountants should expect from them?
Randy Johnston (13:18):
Yes, there's another pretty big trend that's occurring, Dan, and it has been pathetic for years at the risk of using and overstating it reporting, it's been so miserable.
Dan Hood (13:30):
You mentioned earlier that
Randy Johnston (13:31):
Yeah, and it turns out that the vendors are actually starting to get this pretty right now. First, the standard financials, income statement, balance sheet, cashflow, that stuff good. But what's happening is operational data dashboards and metrics KPIs are starting to surface properly inside the systems, but not only is it in the financial sector, they're actually attaching or bringing in data from other places. Now we see this evidenced in almost every system. Every system has a dashboard, but it's usually too narrow to be useful. Usually it has to be augmented. But some of these systems, and let's just pick on acumatic or Sage intech as two examples, can read data from other systems and present them in a single dashboard view. And if you begin to realize that every role inside your organization and has metrics that would be helpful for them for decision-making, there's some thoughtfulness around this.
(14:30)
Now, a good friend of mine out of the uk, Bernie Smith, I think is the best K P I guy in the world. He's taught us quite a little bit about how to think about KPIs and the developers are beginning to understand some of the KPIs now through the decades, again, we've both done this a long time. I've watched dashboards built for manufacturing, for distribution for C P A firms, and there's a whole lot of fits and starts. If you think about it. Dashboard's first surfaced in 1992 with the Harvard Business Review article on that, and I'm still a Harvard dashboard type of thinker, but I can tell you in my own network management business, we wound up building about 80 KPIs and everybody sees their KPIs every day. And I think that's a pretty good way to run an effective operation regardless of your industry. Sure,
Dan Hood (15:26):
Sure. Well, there's a bunch of other ramifications or potential ramifications manifestations of that. Just as you were talking about, I was thinking about the enormous complexity of E S G reporting and the need to bring in non-financial data and the fact that accountants are going to be at the heart of that and the ability to bring in non-financial data, manage it and connect it with all kinds of other data, and then spin it around for either E S G reporting or for decision-making around E S G issues or any issue that it can affect a business, right? As enormous in accountants, corporate accountants internally, but also public accountants to help often to help the corporate accountants develop that kind of information. There's huge demand for that. Or we hear a lot of accountants saying, I'm being asked to do this, I'm being pressed to do this, or I'm hearing people inside corporations asking accountants to do
Randy Johnston (16:12):
This. And it is very common that that request is coming up in my client base as well. I have an old rule that I've used for decades. If an accountant is keen data, you've got a broken process. And the problem here is you've got all these various systems that we're trying to pull in this statistical or other metric data, and we've got way too many accountants manipulating things in the universal hammer of the profession excel. And my next general rule I've used for a long time is if you're reporting in Excel, you got a broken process. So if you begin to think about the demands of boards of operations manager, I need to see this type of data. It ought to be pretty easy to create it and automate it so it's published on a regular basis. Now, a third old rule of mine, Dan, is I'm a real time reporting guy.
(17:11)
I want the data that I'm looking at right now to be up to the minute, and I realize there's some data that comes in slower, but if I've got up to the minute data, I can slow it down to daily or I can slow it down to weekly or monthly or whatever period I want to report on. But this concept of reporting has led to other reporting platforms. Just throw a few out by name, the power Biss, the Tableaus, the Domos, the cliques, the Acton, how many do you want me to name? Because there's hundreds of them without exaggeration, and they're all trying to solve this problem of gathering and reporting. But a few of the platforms, as I laid out earlier, have some of this inside their platforms and that is starting to work.
Dan Hood (17:56):
Excellent. Very cool. Well, it's good. It's good that it's starting to work because people are really starting to ask for it or starting to need it on the other. The flip side of that, obviously maybe use a final sort thought. You've sort of sketched out a picture of where it is and where it's going, where accounting technology is going. But given all that, given the state of accounting software as you've described it, how should accountants be approaching it? What should their technology strategies look like, particularly around accounting technology? There's a lot of other elements we could bring in, I suppose, but maybe specifically around accounting technology. What's their strategy?
Randy Johnston (18:29):
Yeah, I'm going to simplify this one for you, Dan, because having helped organizations pick their accounting software platforms during the pandemic period, I've actually done more selection process and contracting with firms to get these things deployed. And for our listeners, it's important for you to know that I've maintained independence on this, and I realize there's a lot of people that make their living at this and are quite good at their platform. But my rules here are straightforward. First, you should process map your organization. You need to understand your processes because one way you can select a platform is by gap analysis the processes you have versus the processes they have and what it takes to make those come together. The old way of selecting systems was with feature analysis. Now you probably remember many of the people that developed these feature selection tools. Charles tuning with the accounting library was a classic with 4,500 features in manufacturing software and 2,500 features inside general platforms, you should build a very simple list of what you're trying to accomplish.
(19:41)
I usually recommend that you get a list of 10 to 15 things that you want off of a platform and five to seven things you don't want to lose from your existing platform and use that as your selection methodology. Now, we still believe that rf, qss and demos work in this environment, but for a lot of people at the low end, they're just going to say the 800 pound gorilla is into it, or the best alternative we've got is zero or Zoho at the low end. That might be true, but I can absolutely guarantee you that if you understand your business needs, different platforms will fit you better than others. And there are almost no independent consultants to make these recommendations. So this is very much a buyer beware situation because it may sound good and it frankly just doesn't work. And I won't name this platform, perhaps I should, but I know of a SaaS product that is sold into tier two and three that can't do consolidations.
(20:45)
It's like, what? You've got this class of platform and you can't consolidate locations, you can't do, what is that about? And it's a programming errors. So you need to be very thoughtful about the capabilities that you need and you need to have that demonstrated to you by the vendor. Now, I recommend it to demonstration cycle. A short one just shouldn't get an overview. 60, 90 minutes and then a longer one where you control the demonstrations looking at various elements that you have to have your team watch that. It's partially training, but it's also a way to understand how the vendor thinks, how the platform works, and then kind of vote and score that and you'll be able to hopefully get a relatively good fit. And then the last little caveat on this, maybe two. First, there's a devil of a lot of difference between the capabilities of various partners or installers or their direct teams. And you're almost certainly going to need project management on your side to deal with the project management on their side. Now again, we could go through a whole rant of things to select here, but you got
Dan Hood (21:55):
It right. Well, and that project management thing is true of any software you're going to implement regardless whether whatever you're doing that as is the understanding your own needs and own processes beforehand. I mean, those are great advice for any process you want to do, know what you are and what you do and what you need, and then have a carefully managed project process to go through it. It's particularly important, as you say, for software. And given the primacy of accounting software, it's going to be particularly important there.
Randy Johnston (22:22):
And you're going to find the other big variable, and it too change management. People are comfortable with that. They don't want to change. A lot of times they'll shake their head yes, and inside they're saying no, and you're going to have a devil of a lot of things to change. So correctly selecting accounting software can be a career indie move or enabling move, right?
Dan Hood (22:48):
Yeah. Lot you say, a lot of people are thinking, I've already built a million Excel workarounds for this. I know this system. Lemme stick with this. I just want to throw out one final thought. As you mentioned, there are a lot of the people who might be looking to give you advice are tied to platforms in part because you need to be an expert in a platform to implement and that sort of thing. But I would throw out, I mean, what do you think of as part of that process? Just looking at what other users are saying online, it does seem like there's a lot out there and how much you can rely on it, but at least it gives, can potentially give you some sense of, wow, there's a lot of people complaining about this, or wow, there's a lot of people raving about
Randy Johnston (23:21):
It. So to that point, we think it's instructive to talk to other users in a similar industry and see what their experience is, particularly with the platform and the partners. So you get both of those, but also recognize that the limbing effect also occurs here. A lot of people will say, oh, it's going great, and then you go right over the cliff with everybody else. You need to be an independent selector. But yes, you can look at the evaluations. In fact, I did it just this week, Dan, and I'm looking at these evaluations in the add-on stores and thinking, oh Lord, these are not accurate. Five star ratings have been gained. Right? So just recognize that there's a lot of marketing effort to make it sound better than it is, and you have got to keep your wits about you when you're making these choices.
Dan Hood (24:12):
Alright, that's a lot of very sound advice and a lot of very interesting thoughts about what's coming and what's the future's going to look like. Randy Johnston of K2 Enterprises, thanks so much for joining us.
Randy Johnston (24:23):
Thanks, Dan. I appreciate being here.
Dan Hood (24:25):
And thank you all for listening. This episode of File Air was produced by Accounting Today with audio production by Kevin Parise. Rate or review us on your favorite podcast platform and see the rest of our content on accountingtoday.com. Thanks again to our guest and thank you for listening.