What business process automation software should you choose? It's the question business owners throw around to each other constantly. But in a world where these tools are a dime a dozen, the question I prefer to ask is a different one: who's going to help you choose?
Honestly, finding someone to help you choose is the real work. Search the term online and who knows what you'll end up reading. You feel like you've found the world's best cup of coffee just because the package said so. Every site is the best, every tool is the smartest, every platform is the leader in something. That's marketing. It isn't a decision.
So this article is about how to make the decision well — not which logo to put on your screen, but how to think about the choice, when to build versus buy, and what almost every owner gets wrong about it.
"Best" only means one thing — it fits
What makes a piece of software the best? People answer with quality. With speed. With feature counts. None of those is right by itself.
The thing that actually makes a business process automation software the best is whether it works perfectly for you. That's the only differentiator that matters. You need something that fits you precisely — not something that almost fits, not something that fits a generic version of your industry and forces you to bend around the rough edges. Almost-fits costs you every day.
I'll go further. Something mediocre that fits you exactly will out-perform something high-quality that doesn't. Fit beats fancy, every time. Most software shopping fails at exactly this point, because the buyer is comparing the brochures of two impressive things instead of asking whether either one of them is actually built for the shape of their business.
A story about a problem nobody had a tool for
Let me show you what "fit" looks like in practice with a problem we had to solve a while back.
I was working on something for White Shovel. Their project managers were handling paper receipts. They claimed they kept every one of them and that every receipt was always accounted for. I had no way of knowing whether that was true, and the company had no way of knowing either. The receipts went where the receipts went.
What I actually wanted was for every transaction to be matched to the project it belonged to, automatically — for job-costing reasons. If you can't tie a cost back to a job, your margin is a guess. So I set out to solve it.
That mission led me down a bunch of dead ends. Most people just logged the expenses manually. Nobody else was bothering to get them into their software automatically, because banking is heavily restricted for security reasons and nobody wants your software in there poking around. The off-the-shelf options were nothing. We were screwed.
Then one day I noticed something. The bank had a setting where they would notify me if a transaction crossed a certain dollar threshold. So I dropped the threshold to a dollar, which effectively meant every transaction. Now the new transactions were arriving in my email. So I forwarded them, automatically, into a simple back-end spreadsheet our software could read. And just like that, every transaction was on screen for the project managers. Matched, visible, gone forever as a paper problem.
What had seemed complex was much simpler than everyone was making it. I didn't need a fancy tool at all. I needed a notification setting, an email forward, and a spreadsheet pointed in the right direction.
The view matters more than the tools
That's the entire basis behind looking for new tools. The goal is to find something that works — something that doesn't bog down your current system, that solves the actual problem in the simplest way available. And most of the time, when you stay focused on what works, you don't get pulled into what's cool or what's fancy.
Inside the systems I build, I love using AI to perform tasks. It's powerful, it's flexible, it can handle nuance that no rule-based logic ever will. But sometimes AI is actually inefficient for what a plain comparison process can do for less money and faster. Knowing when to reach for which is part of the job.
The tools themselves don't matter when you're building an automation. What matters is the view — keeping your own opinions and brand preferences out of it, and focusing on the solution. Can this tool solve this problem? Yes or no. That's the question. Not whether the logo's pretty. Not whether you heard about it on a podcast. Not whether your friend's cousin uses it. Just: does it solve the problem cleanly?
This is exactly where digging into the research becomes the real job. Making sure what you're getting is something of actual value to this business, not value in the abstract. (We've made the same argument in the pillar on business process automation — the workflow, not the tool, is what you're really designing.)
The real concern: what if you get it wrong?
My concern as someone doing this work has always been the same one a business owner has: what if my client gets it wrong?
It's something I talk to my team about constantly. As a company, we represent our clients and what they're capable of accomplishing. When we research, we have to think about what if we get it wrong on their behalf. We have to push ourselves to be up to date on every possible tool that could improve a client's operation — not because we use all of them, but because the wrong recommendation from us puts the client behind instead of ahead. The question we ask isn't "what can we do to help." It's "how can we put this client ahead of everyone else in their space."
Business owners face the exact same challenge from the other side of the table. Are you using AI enough? Are you researching new tools enough? Are you actually getting ahead, or just keeping the lights on?
Because the software industry has been swept away by hyper-productive people using agents to elevate their ideas and push themselves further, faster than the old pace allowed. And that wave benefits every industry in some way. We're no longer tied to the tools sitting on the shelf. New tools are showing up rapidly and replacing the old ones, and the new ones are smarter than what they're replacing.
But what if you grab the wrong new tool and you end up behind instead of ahead? That's a real concern, and we understand it because we live in it. And it's not just a tool problem. Hiring someone to help get you ahead can also waste time and put you behind, if the person you hired isn't as determined as you are to stay ahead.
That right there is the real work. It's what separates hiring a company, buying a tool, and trying to do the research yourself. All three options exist. None of them is automatically safe. The thing that makes any of them work is the seriousness of the person doing it — yours, your vendor's, or whoever you bring in.
When buying off-the-shelf is genuinely the right call
I want to be clear about something, because too much writing on this topic is one-sided. We are not anti-tool. The White Shovel story above was solved with a bank's existing notification feature, a free email forward, and a spreadsheet. That's three off-the-shelf pieces working together. We did not build a custom banking integration. We used what was already there in a way nobody had thought to use it.
There are plenty of situations where the right answer for a business is to just buy something. If the function is a commodity — bookkeeping, basic scheduling, generic email, payroll — buy the tool. You're not going to out-build QuickBooks at accounting. Don't try. Save your custom work for the things that actually make your business distinct, the parts of your operation that are different on purpose because that difference is part of how you win.
The honest filter is: is this function a differentiator for your business, or a commodity? Commodity → buy. Differentiator → that's where building, owning, and tailoring earns its keep. (And once you decide to build, owning it outright matters — we wrote about that in why you probably need fewer business process automation tools than you think.)
The question hardly anyone really wants to answer
Here's where this gets uncomfortable, and it's the part that's hardest for a lot of our clients.
Are you ready to go through the development process?
Are you ready to build new tools — actually build them, not just sign up for them?
Most owners say yes immediately. Of course yes. But they don't really understand what that means yet. Building and developing involves tests, and revisions, and getting better. We all want to be better. The question is whether we're willing to put up with the discipline it takes.
It's the same conversation as fitness. We'd all like to be fit and toned. But are we willing to stop eating the sugar? Are we willing to hit the gym regularly, when nobody's watching, on the day we don't feel like it? Are we prepared to do the actual things that being fit requires? Most people aren't, which is why most people aren't fit.
Building a great system takes time. Time spent researching. Time spent failing and rebuilding. Running tests, comparing one approach against another, slowly critiquing what you're doing. Are you ready for the first version of your system to fall short and reveal a huge missing component you didn't know was missing until it shipped? Because that's how this actually goes. It's not a clean, single delivery. It's a process of refinement, and refinement requires that you be okay with the first try not being the last one.
This is the real work of development. Real R&D. And it's the work business owners frequently don't want to do. They want to make money and never lose. They don't want to close their doors for a single split second and miss out on a single dollar. So they push their trucks into the ground. Their buildings never get a fresh coat of paint. Their systems dwindle quietly into extinction, year by year, while they're too busy to look up and notice.
We have to be better than that. We have to be willing to pursue these things and push ourselves through the parts that aren't fun. That's what separates the great companies from the mediocre ones. A new business owner spends a huge amount of energy trying to be better, trying to build something awesome — and most of them, eventually, die down. They settle into the norm. And the norm is the place where good companies go to die.
Bringing it back to the software question
So when you ask what business process automation software you should choose, here is the honest sequence I'd run.
Start with who's going to help you choose. That's the real first decision, and the most consequential one. Look for someone whose job is to put you ahead — not to sell you a particular platform, not to renew a particular subscription, but to genuinely move your business forward. If you're researching alone, fine, but understand what that requires of you.
Then forget "best." Look only for fit. A mediocre tool that fits your real workflow will quietly out-earn a celebrated tool that doesn't. Stop reading feature lists. Start describing exactly how the work moves through your business today and ask whether each option respects that flow or fights it.
Then sort what you're looking at into commodity versus differentiator. Buy the commodity pieces. Don't fight to be a better calendar app. The differentiator pieces — the parts of your business that are uniquely yours — are where custom work, owned by you, pays for itself.
And then ask yourself the question almost no one wants to answer. Are you ready for the work? Are you ready to fail forward, to revise, to discover the missing piece in version one and build version two anyway? Because the right software, chosen well and fit precisely, is still only half of it. The other half is whether the business behind it is willing to keep getting better.
That's exactly what we do in an Assessment. We figure out where your business actually fits, what's worth building and what isn't, and whether you and your team are set up to carry it forward. The tools were never the point. They're not the point now. The work, and the discipline to keep doing it, always was.
