Greg Hoy
If you’re in the consulting business and you have a hand in managing projects, you know what scope creep is, and you’ll do whatever it takes to avoid it. The problem is, you also want to maintain goodwill with your client, and that sometimes means extending little scope treats here and there just to show you’re a good person. That’s not actually doing anyone any favors.
For any service consultancy, scope creep doesn’t happen in month six when the client wants one more round of revisions. It happens the second both sides nod along at different bullet points during an early project meeting and assume they mean the same thing.
That assumption is where most project budgets go off the rails.
At the agency I ran for twelve years, we learned that lesson the hard way. An agency lands a project based on a good proposal. The client signs. Discovery happens. Around month three, both parties realize they had wildly different ideas about what a “redesign” or “platform migration” or “digital strategy” would entail. You’re already committed with a portion of the budget spent. There’s only one direction to go: a change order nobody wanted.
It wasn’t better proposals that solved the problem. It was a project definition phase — and it happens before discovery, not during it. Think of it like the definitions section in a contract. Before anyone talks about what you’re building, you establish what you both mean by the words you’re using. Skip that step and the rest is subject to wild ambiguity.
What project definition actually is
Project definition is a separate scoped engagement, at the very beginning of a project, where agency and client agree on scope before committing to timeline and budget.
It’s not a free kickoff meeting. It’s not a discovery sprint that coincidentally happened first. It’s a defined process with a deliverable that both parties own: a roadmap of the work broken into actionable steps, with effort and dependencies defined.
This deliverable could theoretically be taken by the client to any other agency as a scoping document for the whole project. In practice, they don’t. But the portability matters: it means you’re not the one creating the dependency. You’re both figuring out what you’re actually building.
What it involves
It looks different for big versus small projects, but some pieces are always there.
First are stakeholder interviews. Not as a courtesy, but because you’re genuinely trying to understand what success looks like for the people who will use and manage whatever you’re building. They almost always have a different definition of success than the people who signed your contract. Reaching them before you’ve committed to anything is one of the best things you can do.
Then comes a content inventory for any project that involves an existing digital property. Most agencies skip this entirely or leave it until later. That’s how they get surprised to find 4,000 pages when they scoped for 400.
Then comes roadmapping. You break every large piece of work into specific, estimable units and arrange them in priority order against a reasonable estimate of effort. The roadmap is where you start to see the difference between what the client asked for and what they actually need. Better to discover that in month two than month twelve. It’s also what your statement of work should be built from — not a gut estimate, but a documented mutual agreement that both sides helped create.
All of this leads to a final workshop where both teams walk through the roadmap together, reconcile any remaining differences, and sign off on a mutual understanding of scope.
What it looks like in practice
When we were pitching a large nonprofit on a complex website redevelopment, we approached it differently than we had in earlier years. Instead of pricing a project we didn’t fully understand, we proposed a project definition engagement first: five resource weeks of work culminating in a collaborative on-site workshop where their team and ours would walk through a complete project roadmap together and reach genuine agreement before committing a single dollar toward development.
The definition work was $54,500, applied against a full project budget of approximately $275,000. Mathematically, the definition phase is a rounding error compared to the risk of scoping the whole thing wrong.
They said yes.
What came out of that workshop was a roadmap both parties had stress-tested together. The client understood the project — probably better than they’d ever understood a project before. We knew the constraints, the internal politics, the actual complexity. When the full engagement kicked off, we weren’t discovering the landmines. We’d already mapped them.
Why it protects the agency as much as the client
There’s a version of this pitch that’s all about client benefit, and that version is completely true. Clients get a clearer picture of what they’re buying, fewer unpleasant surprises mid-engagement, and a deliverable they own regardless of what happens next.
But project definition protects the agency too. You go into the full engagement with a scope that’s been stress-tested by the people who actually know the client’s systems, constraints, and internal politics. Your estimates are built on real information. Your change order conversations — because they will happen — are easier because you have a documented baseline everyone agreed to. Shared ownership is powerful. It’s a lot better than a he-said/she-said argument over a paper trail nobody can find.
I like to start building as fast as the next person, but the projects that go sideways fastest are the ones where the agency was in too much of a hurry to find out what they were actually building.
When to propose it
Not every project warrants a definition engagement. A small site redesign for a client you know well with a clear brief and a well-defined scope doesn’t need five weeks of definition work.
The signal to propose it is uncertainty. Bullet points that could mean any number of things. Complex integrations. Legacy systems. Multiple stakeholders who might not agree with each other. That’s when assumption-based scoping gets expensive for everyone.
The conversation is simple: we could write you a proposal based on what we have now, but we’d be guessing on some things, and wrong guesses cost more than right ones. Here’s what it costs to get it right.
When most clients hear that, they say yes.