Blog · Personal ·
When a Custom Build Is the Wrong Answer
A studio framework for evaluating bespoke vs. off-the-shelf solutions, distinguishing between friction, structural ceilings, and competitive moats.
When a Custom Build Is the Wrong Answer: An Honest Studio Framework for Bespoke vs. Off-the-Shelf
Every studio says they'll tell you the truth. Almost none of them do, because the truth, most of the time, is "you don't need us for this bit." That's a hard sentence to say when you bill by the build. But it's the sentence that actually earns trust, and honestly, it's the sentence that filters out the clients who'd have been a nightmare anyway.
So let's start with the question everyone gets wrong.
The real question isn't build vs. buy
Founders come in asking "should we build this custom or just use [tool]?" as if it's a binary. It isn't. The actual question is: where, specifically, is your constraint? Not "where is friction" (friction is everywhere, always, forever), but where is the thing that's actually capping your growth, your margin, or your defensibility. Once you know that, the build-or-buy question mostly answers itself.
Most teams skip this step. They feel pain in the workflow, assume pain equals justification, and commission a build. Then six months later they've got a bespoke system that solves a problem off-the-shelf software solved two years ago, and it cost them a fortune, and it's now their problem to maintain forever. That's not a build decision. That's an expensive way of avoiding a Google search.
Three categories of workflow problem
Roughly speaking, every "should we build this" question falls into one of three buckets. Get the bucket wrong and everything downstream is wrong too.
Friction. This is the annoying-but-solved category. Your team hates the CRM, the reporting is clunky, someone's manually copying data between two systems every Friday afternoon. It's real pain, I'm not dismissing it, but it's pain that a thousand other companies have already felt and already paid someone to fix. If the problem is "this is tedious," the answer is almost never "commission a bespoke platform." The answer is buy the tool, configure it properly, maybe pay for an integration. Anyone telling you otherwise is selling you a solution to their revenue problem, not yours.
Structural ceiling. This is different. This is when the off-the-shelf tool technically works but caps what you can actually do, structurally, not just comfort-wise. Maybe your data model doesn't fit any CRM on the market because your business genuinely operates in a way the category wasn't built for. Maybe the workflow you need requires real-time decisions the existing tools physically can't make fast enough. Here, custom isn't indulgence, it's warranted. You've hit a ceiling, not an itch.
Competitive moat. The third category is the interesting one, and it's where studios should be most careful, because it's also where founders most want to hear "yes, build it." This is when the custom build isn't fixing a problem at all, it's creating an advantage nobody else can replicate. Proprietary tooling that makes your delivery faster than anyone in your category. A system that becomes genuinely hard to copy because it's tangled up in years of your own operational data. That's a strategic investment, and it deserves a different conversation, a different budget, and honestly a different level of commitment from you, not just the studio.
The decision matrix
Here's the version that actually works on a whiteboard. Score each candidate problem, one to five, on three things: frequency (how often does this friction actually occur), cost of workaround (what does it cost you, in time or money, to keep limping along with the existing tool), and strategic uniqueness (does solving this well make you meaningfully harder to compete with).
Low scores across the board? Buy. Don't even think about it twice.
High frequency, high workaround cost, low uniqueness? That's your structural ceiling. Custom is warranted, but scope it tightly, and don't let the studio talk you into building the moon.
High uniqueness, regardless of the other two? That's the moat conversation. Different budget, different timeline, different risk appetite required.
Anywhere in the middle, honestly, just wait. Not every ambiguous score needs resolving today.
Red flags in a studio that always says "build it"
If every conversation you have with a studio ends in "we'll custom-build that," that's not confidence, that's a business model. A studio that's actually good will tell you, unprompted, when the smart move is a fifty-a-month subscription instead of a six-figure build. That sentence costs them money in the short term. It's also the single fastest way to tell if they're thinking about your business or their invoice.
Watch for a few things specifically. Vague scoping that never quite defines what "done" looks like, because ambiguity keeps the meter running. A refusal to name existing tools as viable alternatives, even hypothetically, which usually means they haven't actually evaluated them. And the big one: enthusiasm that doesn't move depending on whether the underlying problem is friction, ceiling, or moat. If the studio is equally excited to build you a Slack integration and a proprietary matching engine, something's off. Those are wildly different jobs with wildly different justifications, and a studio that treats them the same either hasn't thought it through or doesn't care.
The best studios lose some revenue by being honest early. That's fine. That's the trade. You want a studio that occasionally talks you out of a build, because that's the studio that, when they finally say "yes, this one's worth building," you can actually believe them.