Tales & Co.

Istanbul — San Francisco

Notes

Diagnosis before solution

6 min read

Clients Choose the Consultant They Can Defend

The scorecard ranks the firms. The sponsor's need to defend the choice picks one. Seeing that second filter lets a buyer decide what it should actually reward.

The stated criteria rank the firms, and a different set of criteria picks one.

Selection processes publish their criteria. A scorecard weights expertise, approach, price and references, and a panel scores each proposal against it. The scorecard is real and it does narrow the field. It rarely decides the outcome. What decides it is a question that appears on no scorecard: can the person who recommends this firm defend the choice if the work goes badly?

That question belongs to the sponsor, usually an executive who carries the engagement's reputation inside the company as well as its budget. A consulting engagement is a purchase whose result arrives late and is hard to attribute. The buyer cannot verify quality in advance, so the buyer verifies defensibility instead. Earlier notes in this series covered how to test fit against a named constraint and what a proposal reveals about delivery. This one describes what the buyer is doing underneath both.

What a buyer is protecting

Under uncertainty about outcomes, a buyer optimises for the choice that survives scrutiny, and the strongest-looking firm is only one route to it. Three things make a choice survivable, and each lowers the sponsor's personal exposure more than it raises the expected quality of the work:

  • A name the board already recognises, which turns a disappointing result into a market-standard disappointment.
  • A method that can be described in a paragraph to someone who was not in the room.
  • A reference from a company that resembles this one closely enough to be cited.

None of these is irrational. Each is a sensible response to a decision that cannot be checked until it is too late to change. The difficulty is that a company can believe it is buying capability while the process is quietly buying cover.

The shortlist is mostly decided before the process begins.

Procurement describes selection as a funnel that opens wide and narrows. In practice the funnel starts narrow. The firms that reach a first meeting are the ones someone already knew: a former colleague's firm, a name from a board member, a partner who wrote something the sponsor read last spring. The formal process then tests a list that was assembled from memory.

Why memory builds the list

Memory favours the firms that stayed visible in the months when nobody was hiring. That is a point about the buyer and about the firms, and it explains why small specialist studios are so rarely on the list however well they would fit. The company did not decide to exclude them. It never retrieved them.

A pattern recurs in mid-sized companies that run a thorough process. Five firms are scored, the panel is conscientious, and the winner turns out to be one of the two the sponsor named before the process opened. The scoring was honest. The list it scored was not neutral, and nobody was responsible for that, because nobody was asked to build it.

What changes the list

A buyer who wants a different result changes the input and leaves the scorecard alone. Writing the constraint in a sentence before collecting any names, as the diagnostic a company can run before hiring anyone describes, produces a different set of firms than asking colleagues who they used. Both lists are legitimate. They answer different questions, and only one of them starts from the problem.

Proposals are read for defensibility, and insight is the lesser test.

A proposal is read twice. The first read is for what the firm will do. The second, usually done by the sponsor alone, is for what the sponsor would say about it to a sceptical colleague. The sentences that survive the second read are plain ones: a stated scope, a named team, a sequence with decision points, a fee that can be explained in a line. The sentences that fail it are the ambitious ones, however good.

Why proposals converge

Firms learn which sentences are safe to sign, and proposals drift toward them. The most interesting idea in the pitch is often cut from the final document because it cannot be defended in one line, and the engagement then delivers the safe version of a good argument. The buyer sees three similar documents and concludes that the market is undifferentiated. The market was differentiated until each firm edited itself for the second read.

The buyer cannot verify quality in advance, so the buyer verifies defensibility instead.

This shapes what a buyer can usefully ask for. A request that rewards the sharpest claim, and protects the firm that makes it, will get a different proposal than a request that rewards the safest one. A buyer can say in the brief that a proposal's single most arguable recommendation should be stated and marked as such. Firms that have been editing themselves for years will notice, and some will write the better document.

The decision usually sits with the person who carries the downside.

The budget holder, the sponsor, the daily counterpart and the end user are often four different people, and a careful process consults all of them. The weight does not fall evenly. The person who answers for the result, to a board, an investor or a chief executive, holds a quiet veto, and it is exercised in a single sentence: I am not sure we can explain that.

Reading the veto early

The veto can be read before it is used. The questions that come back from the senior sponsor in a second meeting are the real criteria, and they cluster around three concerns:

  • Risk: what happens if the first stage disappoints, and who sees it first.
  • Precedent: who else has done this, and how recently.
  • Reversibility: what the company is left holding if it stops after a month.

These are the questions of someone building a defence file, and they are reasonable ones. A firm that answers them in the first meeting, without being asked twice, has understood the buyer's actual position. A buyer who notices that these three questions are doing more work than the scorecard has learned what the process is for.

Making the real criteria explicit is where the thinking becomes work.

The method for a buyer is short. Write the defence before the shortlist: the one paragraph the sponsor would give a board to explain why this firm, this scope and this fee. Then test each candidate against that paragraph and against the constraint the engagement is meant to move. If the paragraph can only be written about the familiar name, the process is a ratification, and it is cheaper and more honest to call it one.

For a firm, the reading runs the other way. A proposal can be written so that the sponsor can defend it without losing the useful idea, for instance by pricing the first stage as its own decision and stating the stop condition in the document. The diagnosis before solution pillar collects the wider material on why the question has to be settled before the supplier is chosen.

Where companies want an outside view on how a selection is really being made, or on the diagnostic that should precede it, that is the kind of work our consulting practice does, and we will say plainly when the better choice is a firm other than ours.