It Was Scoped as a Four-Day Job. The Real Problem Needed Six Months, and It Came In Under Budget.

In 2015, Collective Health booked me for four days. I was the lead consultant and project manager at Adaptavist at the time, and the job was a quick information architecture review, mostly about organizing a few Confluence spaces. Straightforward scope.

Within the first few hours of talking with their team, it was clear the real problem was much bigger than that.

Collective Health was a health insurance startup with about 35 employees, tens of millions of dollars in venture funding, and a plan to grow from a pilot of about 100 members to 40,000 within six months. The industry average for customer satisfaction in health insurance was 35%, and they'd raised their money on a promise to double it, to 70%. The information architecture work was necessary, but it was a band-aid on a deeper wound. Their tools weren't built for the growth they were about to go through, and the people answering members' questions were going to feel that first.

So I had a choice that every advisor eventually faces. I could do the four days I'd been hired for, do them well, and leave. Or I could tell the client the job they'd bought wasn't the job they needed.

The scoped work wasn't the client's real problem

The four-day version would have been easy to defend. It was what the statement of work said, it would have produced a clean deliverable, and nobody would have complained. It also would have left a support team about to handle a 400-fold increase in members with tools that couldn't keep up, in an industry where the moment someone calls is often a moment they're anxious about a medical bill.

I flagged it early, laid out what I was seeing, and proposed expanding the engagement. That was a real risk for Collective Health, too: a much bigger investment and a much longer timeline than the one they'd planned for. They agreed, and a four-day review became a six-month project.

That conversation is harder for an independent advisor than it looks from the outside, because there's no firm behind you. When you're the one whose name is on the work, proposing a bigger engagement can feel indistinguishable from an upsell, both to you and to the client. The temptation is to stay quiet, deliver what was scoped, and hope the bigger problem becomes someone else's.

Why saying it builds trust instead of eroding it

The research on client relationships suggests clients read it the other way. In a study of consultant–client relationships in small businesses, Gary Noble found that clients' trust didn't rest primarily on the consultant's perceived expertise. It rested on what the clients themselves called the "honesty factor": the consultant's willingness "to speak the truth" rather than mask their views about where the business really stood. It's a small qualitative study, but it lines up with larger ones. When researchers surveyed 171 executives at large Australian companies about what makes consultants successful, integrity and honesty ranked among the top five capabilities, alongside the ability to listen to and understand the client.

Listening is the other half of it, and it's what makes a re-scope credible rather than self-serving. The proposal came out of those first hours of conversations about how their team actually worked and where members' questions were going to land, so I could show them the gap in their own terms instead of asking them to take my word for it.

What the bigger job actually was

The six months went into three connected systems. The first was a knowledge base in Confluence organized around member scenarios and the questions members actually asked, rather than around the company's org chart, with review cycles and owners so it would stay accurate as Collective Health added employers and plans. The second connected their in-house phone system to member profiles and plan details, so a support rep knew who was calling and what their plan covered from the moment they picked up, and could escalate an issue into Jira with everything attached. The third was a bug-tracking workflow in Jira that kept protected health information properly restricted without making the process so painful that engineers worked around it.

I designed the architecture, but I didn't build all of it. An Adaptavist engineer built the phone-system integration, two other consultants built the HIPAA-compliant bug tracking, and all three were in London while I was in Missouri. I flew out to Collective Health's office in San Mateo every other week for six months. I was at their Thanksgiving party. When I started there were about 35 people, and when I left there were close to 500.

Coming in under the bigger scope

The project came in under budget and ahead of schedule, and that's the part I'd point any advisor to, because it's the proof that the re-scope was about the client. Delivering under budget meant Adaptavist made less money. We were okay with that, because the priority was Collective Health.

And the outcome mattered more than the invoice. Collective Health's members rated the company at 95% satisfaction, against a promise of 70% and an industry average of 35%. I'd credit that partly to the systems and mostly to a phenomenal team of really smart people, whose commitment to member experience is what drove those numbers. Good tools in the hands of a mediocre team produce mediocre results, and this team was anything but mediocre. The systems gave them a foundation for accurate answers delivered fast, which is what drives satisfaction when people are anxious about their health insurance.

The principle I took from it, and still work by, is to be aggressive about identifying the real problem and disciplined about solving it.

Scope expansion isn't scope creep

Those two things get confused, and the confusion is why so many advisors avoid the conversation. Scope creep is uncontrolled: the engagement drifts, nobody writes down what changed, and the client discovers at invoice time that they bought something different from what they agreed to. Scope expansion, when it's deliberate and justified and well-managed, is the opposite. It happens early, out loud, with evidence, and with the client deciding.

The client is the one who can tell the difference, and the research suggests they notice. Fiona Czerniawska found that going back over a consulting firm's original proposal was ten times more common among unhappy clients than happy ones. Clients rarely reread the scope when the work is solving their problem. And in a 2024 survey by the CMCE, a UK consulting body, 73% of respondents named agreeing client expectations as the most important factor in whether consultants delivered value. A re-scope done well is exactly that: agreeing, again and out loud, on what the client actually needs.

So if you're a few hours into an engagement and the job you were hired for isn't the problem, say so. Say it early, show your evidence, and give the client the real choice, including the choice to keep the original scope. And if they choose the bigger job, deliver it as though the smaller one were still the budget.

An honest re-scope is a finding you sign. Coming in under it is how the client knows you meant it.