Prescriptive or Descriptive: how to engage a consultant

Short answer: before you engage a consultant, decide whether the engagement is Prescriptive or Descriptive. A Prescriptive engagement tells you what the problem is and what to do about it. A Descriptive engagement tells you only what the problem is, and leaves the solution to you. Choose Descriptive when your own team has to own the fix, or when you want a diagnosis that isn’t shaped by anyone’s preferred solution. Choose Prescriptive when you don’t have the capability or the time to design the fix yourself.

Most consulting engagements that disappoint don’t fail because the advice was wrong. They fail because the client and the consultant had different ideas about what was being bought.

The CEO wanted a clear-eyed read of what’s broken, and got a 60-page transformation roadmap that the leadership team now has to accept or reject. Or the board wanted a recommendation it could act on, and got a well-evidenced list of observations ending in “it depends”. Both consultants did good work. Both clients were disappointed. Nobody said at the start which of the two jobs it was.

I’ve been on both sides of that table: as the consultant brought in to build a strategy, and as the executive accountable for what consultants and vendors delivered. I call the choice Prescriptive or Descriptive. It’s the most useful question to settle before an engagement starts, and it’s rarely asked.

What do Prescriptive and Descriptive mean?

The easiest way to see the difference is a visit to the doctor. A diagnosis tells you what’s wrong. A prescription tells you what to do about it. You can ask for one without the other.

• Prescriptive: the consultant tells you what the problem is and what the solution is. You receive a diagnosis, the options they considered, a recommendation and usually a plan.
• Descriptive: the consultant tells you what the problem is, and stops there. You receive a diagnosis: the real problem, its root causes, the evidence and how big it is. Your team designs the solution.

The names are deliberate. Data teams already use the same pair of words: descriptive analytics explains what’s happening, prescriptive analytics recommends what to do about it. The consulting version works the same way.

Neither mode is better. They’re different products that suit different situations and carry different risks. The mistake isn’t choosing the wrong one. It’s not choosing at all, so the engagement drifts into whichever mode the consultant usually sells.

Prescriptive vs Descriptive at a glance

	Prescriptive	Descriptive
The question it answers	What’s wrong, and what should we do?	What’s actually wrong?
What you receive	Diagnosis, options considered, a recommendation, a plan	Diagnosis, root causes, evidence, the size of each problem
Who designs the solution	The consultant	Your team
Who owns the outcome	You	You
Best when	You lack the capability or the time to design the fix, or a decision-maker needs a recommendation	Your team can design the fix but is too close to see the problem, or has to own the answer
Main risk	Solution bias, and a plan your team doesn’t believe in	A well-documented problem that nobody acts on
How to judge the work	Is the recommendation specific, costed and adopted?	Is the diagnosis accurate, evidenced and accepted by the people who must fix it?

Notice the fourth row. In both modes, you own the outcome. A consultant advises, whichever mode you choose. If you need someone to own the result, that’s a different kind of engagement, which I come back to below.

When should you choose a Descriptive engagement?

Descriptive is the less common choice, and I think it’s underused. It fits when:

• Your team is capable, but too close to the problem. Good people inside a business often can’t see what’s in front of them, because they built it, or because they’ve worked around it for so long it looks normal. They don’t need to be told how to fix it. They need to be told what it is.
• The people who will implement the fix need to own it. Solutions people design themselves get implemented. Solutions handed to them get debated. If the success of the fix depends on a leadership team’s commitment, letting them design it is often worth more than a better answer from outside.
• You want a diagnosis without a sales pitch attached. A consultant who diagnoses the problem and also sells the cure has a reason, conscious or not, to find a problem their cure fixes. Taking the cure off the table removes that pull.
• You already have a solution in mind. Sometimes the leadership team has a plan and wants to know whether it addresses the real problem. A Descriptive engagement tests the problem, not the plan, so it can’t simply hand your idea back to you with a logo on it.
• The board wants an independent read. Boards often need to understand a problem before management proposes the response. A Descriptive review keeps assessment and advocacy apart.

In a Descriptive engagement the discipline sits with the consultant. Any experienced consultant will have opinions about the fix by the end of the first week. The job is to keep those opinions out of the diagnosis, so the problem isn’t quietly described in a way that leads to their preferred answer. Agree up front what happens to those opinions. One option is to ask for them in a separate appendix that your team opens only after it has done its own thinking.

When should you choose a Prescriptive engagement?

Prescriptive is the default for good reasons. It fits when:

• You don’t have the capability in-house. If nobody in the business has rebuilt a pricing model across several markets, or taken an AI pilot into production, handing them a diagnosis and asking them to design the fix is slow and risky. You’re paying for experience, so use it.
• Time is short. A raise, an exit, an integration deadline or a program that’s burning cash. When the cost of delay is high, you want the answer, not a starting point.
• A decision-maker needs a recommendation. Boards, investors and lenders often want an independent recommendation, not just an independent view of the problem.
• The problem is a well-known one. Some problems have well-understood solutions. Paying someone to describe one, then asking your team to reinvent the fix, wastes everyone’s time.
• You’re stopping something. A recommendation to stop needs an alternative attached, or the gap it leaves gets filled by sunk cost.

That last point is personal. At a global consumer brand I was the business owner of a US$1M+ digital program that wasn’t going to deliver. Saying “this program won’t work” would have been a Descriptive answer, and it wouldn’t have been enough. I recommended we stop it, proposed an alternative and asked for due diligence before anything was decided. The business adopted the alternative and avoided more than US$1M in capital spend and around US$600K a year in operating cost (the case study is here). Stopping something is only half a recommendation. I’ve written more about navigating the politics of shutting down a project.

What goes wrong with each mode?

Both modes fail in predictable ways. Knowing them is most of the defence.

Prescriptive: the hammer and the nail

• Solution bias. Consultants tend to prescribe what they know and what they sell. A firm with a platform practice finds platform problems. A firm with a restructuring practice finds restructuring problems.
• A rushed diagnosis. When the fee and the deliverable sit in the solution, the diagnosis can become the part you hurry through to get there. A brilliant answer to the wrong problem is still the wrong answer.
• “Not invented here.” A plan that arrives from outside, fully formed, gives the leadership team something to push back on rather than something to own.
• The report in the drawer. If nobody in the business has the capacity to deliver the plan, it waits for a better time that doesn’t come.

Descriptive: a well-documented problem

• “We knew most of that.” A diagnosis that only confirms what the leadership team suspected can feel like poor value, even when the confirmation is useful. The test of a good Descriptive engagement is whether it found something the business didn’t know, or proved something it couldn’t.
• The comfortable fix. Left to design its own solution, a team can choose the fix that treats the symptoms and avoids the hard part.
• Nobody owns the next step. Findings land, everyone agrees, and then there’s no one whose job it is to turn them into a plan.
• Paying twice. If the team can’t design the fix after all, you engage someone again, often the same consultant, who has to rebuild the context.

Why do most consulting engagements default to Prescriptive?

Because consulting is usually sold on the solution. The proposal leads with the answer, the fee is attached to the plan and the success story in the pitch is about what got fixed. A proposal that says “we’ll tell you what’s wrong and then stop” is harder to sell, even when it’s what the client needs.

There’s also an incentive problem, and it’s not about honesty. When the firm that prescribes is also the firm that implements, the prescription and the next invoice point the same way. Most consultants handle that well. But you don’t need to rely on that if you choose the mode yourself, deliberately, before the work starts.

Describe first, prescribe second

For a significant problem, the approach I recommend most often is to run both modes in sequence, with a decision point between them.

1. Describe. Diagnose the real problem, its root causes and its size, with evidence.
2. Agree the diagnosis. The leadership team, and the board if it needs to, accepts that this is the problem. Not the solution, just the problem.
3. Decide who prescribes. With an agreed diagnosis in hand, decide whether your team designs the fix or the consultant does.
4. Prescribe, if you need to. The recommendation is then built on a diagnosis everyone has already accepted.

This works because it gets agreement on the problem before anyone starts arguing about solutions. Most leadership disagreements about “the plan” are really disagreements about the problem. It also keeps your options open: you can stop after the diagnosis if your team is ready to take it from there.

My diagnostic sprint follows this sequence over about two weeks. The first week is discovery: interviews across teams and partners, and a review of the data, systems, pricing and plans. The second week is diagnosis, a findings workshop with leadership, and then a board-ready report and 90-day plan. That makes it Prescriptive by default, but the findings workshop is the decision point. If you’d rather your team designs the fix, tell me at the start and we’ll agree to stop at the diagnosis.

How to write Prescriptive or Descriptive into the brief

The simplest way to avoid the mismatch is to put the mode in writing before the engagement starts. A good brief covers six things:

1. The mode. One word, near the top: Prescriptive or Descriptive. If you’re staging it, say so and name the decision point.
2. The deliverable. For Descriptive: the problems, their root causes, the evidence for each and their size. For Prescriptive: all of that, plus the recommended solution, the alternatives considered and why they were rejected, the cost, the timeline and the risks.
3. What happens to the consultant’s opinions. In a Descriptive engagement, say whether you want them held back entirely or set aside in an appendix.
4. Who has to accept the diagnosis. Name the people. If the CFO doesn’t accept the problem, they won’t fund the fix.
5. Whether the consultant can bid for implementation. If they can, say so openly, and consider whether a Descriptive diagnosis should come first.
6. What “done” looks like. A board paper, a workshop, a decision or a plan with owners and dates.

Here’s how the opening of a Descriptive brief might read:

Engagement mode: Descriptive. We want an independent diagnosis of why new-customer revenue has been flat for three quarters. Please report the problems you find, their root causes and the evidence for each, sized where you can. Please don’t recommend solutions: our leadership team will design the response. If you form views on the fix, put them in a separate appendix, which we’ll open after our own planning workshop.

That paragraph would have saved both of the disappointed clients at the top of this article.

Questions to ask a consultant before you sign

Whichever mode you choose, these questions show quickly whether the consultant has thought about it too:

• “Is this engagement Prescriptive or Descriptive?” If they hesitate, or answer “both” without explaining how, they haven’t thought about it.
• “What will I have in my hands at the end?” Ask for an example, with the confidential parts removed.
• “If you might also win the implementation, how does that change the diagnosis?” A good consultant has a straight answer to this.
• “What would make you tell us the problem isn’t what we think it is?” You want someone who’ll say so.
• “What happens if we disagree with your diagnosis?” The answer tells you whether you’re buying evidence or a point of view.

Beyond Prescriptive: when you need someone to own it

Prescriptive and Descriptive are both ways of engaging an adviser, and both end with a handover. Somebody in your business has to own the fix. If nobody can, because the seat is empty or the capability isn’t there, a more detailed prescription won’t help. What you need is a person who owns the outcome.

That’s the difference between a consultant and a fractional or interim executive. As I wrote in fractional vs interim executive, a consultant advises but doesn’t own the outcome. A fractional COO, fractional CRO or fractional Chief AI Officer holds the seat, makes the calls and answers for the results.

	Descriptive	Prescriptive	Ownership
What you get	The problem	The problem and the solution	The problem, the solution and its delivery
Who carries the fix	Your team designs and delivers it	Your team delivers it	The executive owns it, with your team
Typical form	A review or diagnostic	A strategy or advisory project	A fractional seat or interim role

The modes can follow each other. At an ASX-listed technology company I was first engaged as a consultant to build a global go-to-market strategy, a Prescriptive engagement. The company then created the role of Chief Revenue & Marketing Officer, and I took it on, owning a global P&L of around A$15M. Writing the prescription and then owning its delivery meant it had to work in practice, not just on paper.

The bottom line

Before you engage a consultant, decide what you’re buying. Prescriptive tells you the problem and the solution. Descriptive tells you only the problem. Choose Descriptive when your team must own the fix or you want a diagnosis free of anyone’s preferred solution. Choose Prescriptive when you need experience you don’t have, or a recommendation someone can act on. For anything significant, describe first, agree the diagnosis, then decide who prescribes. And write the mode into the brief, in one word, before the work starts.

If you’re weighing up outside help and aren’t sure which mode you need, or whether you need an adviser at all rather than someone to own the result, I’m happy to talk it through. My two-week diagnostic sprint starts from A$15,000 + GST, and fractional seats from A$12,000 + GST a month. Book a 30-minute call.