Choosing a Consulting Firm Means Choosing a Delivery Model
Proposals describe how a firm sells. The engagement runs on staffing, references and fee structure, and those can be tested before anyone signs.
A proposal describes how a firm sells, and the engagement is run by a different part of the firm.
Companies that compare consulting firms tend to read the proposals as if they were samples of the work. They are samples of something else. A proposal is the product of the firm's sales process: the partner who writes well, the framework that has been polished across forty pitches, the case study chosen because it reads cleanly. The engagement that follows is staffed, scoped and managed by people who may never have been in the room.
The gap between the pitch and the delivery
A pattern recurs in mid-sized companies that run a proper selection. Three firms present, one is chosen on the strength of a partner's judgement in the room, and the work begins with a team the company has met once. Eight weeks in, the partner appears at the steering meeting, and the analysts are doing something competent that does not quite resemble what was discussed. Nobody lied. The firm sold its best people and delivered its standard model.
The selection question that matters is not which firm is strongest but which firm's delivery model fits the work the company actually has. That is a question about structure, and it can be tested before signature if the company knows where to look. Earlier notes in this series covered how to test fit against a named constraint and when the timing is wrong. This one assumes both are settled and looks at what the proposal and the firm's references reveal about delivery.
The team that is named in the pitch is the first thing worth testing against the team that will turn up.
A proposal usually lists a lead partner, a principal and a set of consultants, with biographies attached. The biographies describe the firm's depth. They do not describe the allocation, and the allocation is where the cost sits.
Who does the work, and how much of their week
The useful questions are plain and rarely asked, because asking feels like doubting the firm's good faith. It is closer to the opposite: a firm that has thought about its staffing answers them in a sentence.
- Which named individuals will be on the engagement, and what share of their week it takes.
- Who writes the first draft of each deliverable, and who reviews it before the client sees it.
- What happens when the lead partner is unavailable for a fortnight, and who the client calls.
A firm whose model is a senior person with a small team of specialists will answer concretely. A firm whose model is a senior person selling and a rotating pool delivering will answer with language about flexibility. Both models exist and both can be right. The cost is in not knowing which one is being bought, because the company then discovers it in the second month and pays to adjust.
The firm sold its best people and delivered its standard model.
A reference check is informative only when it asks what the firm declined to do.
Most references are arranged, and most are positive. That does not make them useless. It means the questions have to be ones that a positive reference can still answer with substance.
What a reference can still tell a buyer
The question 'were you satisfied' returns courtesy. The more revealing set concerns the points where the engagement could have gone the easy way and did not. Did the firm ever tell the client that a requested piece of work was not needed? Did it change the scope after the first month, and who raised it? What was left in the company when the engagement ended that the client still uses a year later, and what has been quietly dropped?
A client who can name something the firm refused to do has seen the firm under pressure, and that is the nearest available evidence of how it behaves when the company is the one pushing. A client who can only describe the deliverables has described the proposal again.
The last question is the most useful of the set. Work that was handed over in a form the client's own people could run, such as a decision calendar, a set of thresholds or an ownership map, tends to survive. Work that lives in a slide deck and a model only the firm can update tends to decay once the firm leaves. The distinction says a lot about whether the firm is building the client's capacity or its own repeat business.
The fee structure shows what the firm expects to find.
Fees are discussed as a cost, and compared across firms as one. They are also a statement about how the firm sees the problem, and the structure is more informative than the number.
Fixed scope, open scope and a first stage
A fixed fee against a long list of deliverables says the firm believes the problem is already understood. That is appropriate where it is, and a warning where it is not. An open-ended time-and-materials arrangement says the firm expects to discover the scope as it goes, which is honest for an unclear problem and expensive for a clear one. A bounded first stage, priced separately and ending in a decision about whether to continue, says the firm is willing to be wrong about the size of the engagement.
That third structure is the least common in proposals and the most useful to a buyer. It is also close to the discipline of the diagnostic a company can run before hiring anyone, moved inside the engagement. A firm that prices its own first stage as a separate decision has accepted that the company may stop, and that acceptance is a stronger signal of fit than any credential on the slide.
None of the three structures is wrong in itself. The question is whether the structure matches how well the problem is understood, and a mismatch in either direction means one side is carrying risk the other has not priced.
Where the selection becomes work.
Selecting a firm is a decision with a hidden owner, a hidden dissent and a hidden review date, in the same way as any other contested decision in a company. It is useful to treat it that way. The diagnosis before solution pillar collects the surrounding material on why the question has to be settled before the supplier is chosen.
The practical version is short. Write down the delivery model the company needs: who must be on the work, what must be left behind, and what stage gate the company wants. Score each proposal against that page rather than against each other. Ask each reference about refusals and about what survived. The result is a decision that can be explained to a board in a paragraph, and that does not rest on the memory of a good meeting.
Where companies want an outside view on that selection, or on the diagnostic that comes before it, that is the kind of work our consulting practice does, and we are willing to say when the honest answer is to run the first stage internally.