There’s a particular experience that most CTOs and VPs of Engineering have had at least once, and it goes like this:
You have a real problem. Maybe you’re three months into a platform migration that’s gone sideways. Maybe you’ve been told to “do something with AI” and you need someone who’s actually built production AI systems to help you figure out what that looks like. Maybe your engineering org is scaling faster than your ability to hire leaders, and the architecture decisions being made in the absence of senior judgment are starting to compound.
So you engage a consulting firm. They send a team. The team has a partner, a senior consultant, two associates, and an analyst. You meet the partner during the sales process and maybe once more during a kickoff. From that point on, you work primarily with the associates and the analyst. They’re smart. They’re well-intentioned. They have not personally built the thing you’re asking them to help you build.
Over eight to twelve weeks, they interview your team, review your architecture, synthesize their findings, and produce a document. The document contains recommendations. The recommendations are broadly correct. They are also things your senior engineers could have told you, and in many cases did tell you, before the engagement started. What the document does not contain is anyone who will be there to help you execute the recommendations. Execution is a separate engagement. With a separate scope. And a separate invoice.
This is not a broken process that happens to look like consulting. This is consulting working exactly as designed.
The leverage problem
The traditional consulting model is built on leverage. A small number of senior people sell and scope engagements. A large number of junior people deliver them. The margin is in the gap: senior rates charged for junior hours.
This works fine for certain kinds of work. If you need a regulatory compliance audit, a market sizing exercise, or a standardized ERP implementation, the leverage model is appropriate. The work is procedural enough that experienced juniors can deliver it effectively under light senior supervision.
Technical consulting is different. The value of a technical engagement is almost entirely in the judgment of the person doing the work. When you’re deciding whether to refactor a monolith into services or strangle it incrementally. When you’re evaluating whether your data platform can support the AI workloads your CEO is promising investors. When you’re designing an engineering organization that needs to double in eighteen months without losing its ability to ship.
These are judgment calls. They require the person in the room to have made similar decisions before, lived with the consequences, and learned from the ones that went wrong. You can’t delegate that to someone two years out of a computer science program, no matter how talented they are.
The document problem
The other structural issue is deliverables. Consulting firms sell documents: assessments, roadmaps, strategy decks, architecture reviews. Documents are easy to scope, easy to bill against, easy to hand over in a final readout meeting.
But documents don’t ship software. Documents don’t resolve the tech debt that’s slowing your team down. Documents don’t sit in an architecture review and say “I’ve seen this pattern fail at three other companies and here’s why.” Documents are artifacts of thinking. They are not substitutes for the thinking itself.
The best technical advice we’ve ever given wasn’t in a deliverable. It was in a conversation at a whiteboard where someone said “what if we just don’t build that?” and saved six months of work. That conversation doesn’t show up on an invoice as a line item, but it was worth more than any assessment we’ve ever produced.
What we decided to do differently
When we started Steadfast, we made a few decisions that seemed obvious to us but apparently aren’t standard in the industry.
No leverage model. The person you talk to during the sales process is the person who does the work. We don’t have associates. We don’t have analysts. We don’t have a bench of junior consultants waiting to be staffed onto your engagement. The cost of this is that we can’t serve dozens of clients simultaneously. The benefit is that you never wonder whether the person reviewing your architecture has actually built an architecture before.
No deliverable theater. We’ll produce documents when documents are useful. But the measure of a successful engagement isn’t a binder of recommendations. It’s whether something changed. Did the migration get unstuck? Did the AI system ship? Did the engineering org start making better decisions? If we spent twelve weeks with you and the only tangible output is a PDF, something went wrong.
No discovery as a product. The most profitable thing in consulting is perpetual discovery: “we need to understand your landscape before we can recommend a path forward.” Sometimes this is genuinely necessary. More often, it’s a way to bill eight weeks of interviews before doing any actual work. We’d rather spend two weeks understanding your situation and ten weeks doing something about it.
The discomfort of honesty
We’re aware of the irony of a consulting firm criticizing consulting. We’re also aware that some of what we’re describing will sound like marketing. “We’re not like those other consultants” is the oldest positioning play in the book.
So here’s the honest version: we made this model because it’s the model we wished existed when we were on the other side. When we were the CTOs hiring consultants, this is what we wanted: senior people who’d actually done the work, who’d stay through execution, and who’d measure success by outcomes rather than deliverables.
We didn’t find it. So we built it.
Whether that resonates depends entirely on whether you’ve had the experience we described at the top. If you have, you probably already know what you’re looking for. If you haven’t, count yourself fortunate, and bookmark this for later.
Steadfast Digital does technical consulting without the parts of consulting that don’t work. If you’ve been burned before and you’re skeptical, good. Let’s talk about it.