Mandate and diagnosis · Lesson 1

Frame the consulting mandate

6 min read

Lesson objectives

  • Separate business decision, authorized scope, decision owner and evidence threshold; case: owner asks for more followers while margins shrink.
Reading tools

A capable consultant begins with the decision the work will support, rather than the channel named by the client. The owner of a fictional homeware store says, “I want one hundred thousand followers; sales have fallen.” Followers measure an audience. Management's problem might instead be weak qualified demand, falling checkout completion, or declining margin per order. Accepting the request verbatim could yield an excellent posting calendar for the wrong problem. Rewrite the mandate as a testable sentence: “Determine why digital sales contributed less margin last quarter, and compare interventions the team can implement within ninety days.” This sentence does not assume that advertising caused the decline or will cure it.

Separate four elements before opening a dashboard. The first is the decision: should the firm reallocate acquisition spend, repair the buying experience, or revisit product and price? The second is the decision owner, who can approve a budget and stop an experiment; this person may differ from the data owner. The third is scope, such as the website, email, and paid channels in two specified markets, excluding a pricing-policy change outside the team's authority. The fourth is the success criterion, such as better contribution margin on new-customer orders without a rise in complaints. “More sales” alone is insufficient because discounts can increase revenue while reducing profit.

Start the mandate interview with questions that do not steer witnesses toward your preferred answer. What change prompted the request? When did it begin? What does the business call a sale: an order placed or a paid order net of returns? Which decisions are actually available? Who can grant access to platform reports, inventory history, and service records? Ask for disconfirming observations as well: did one product group or country improve while another fell? That contrast may reveal inventory or measurement differences rather than a collapse of all marketing. Record stakeholder accounts as testimony and system extracts as records. Neither becomes a causal explanation merely because it is vivid or numerical.

A consulting engagement is an agreement about deliverables and evidence limits. In a short charter, define the comparison period, segments, source systems, access rights, people to interview, and form of recommendation. Promise “a diagnosis with confidence levels, options with cost and conditions, and an experiment and monitoring plan,” not a guaranteed performance result. State exclusions: no formal financial audit, no legal opinion about consent or privacy, and no guaranteed search rank or advertising return. If returns data are missing, flag a profitability evidence gap and defer a margin conclusion; an available click-through rate cannot substitute for it.

The consultant's role differs from that of a campaign operator. The consultant may challenge the plan, request an event ledger, and compare evidence, but does not automatically have permission to change a live campaign or upload customer data to an outside tool. Specify the minimum access needed for each task, who approves exports, and how restricted copies are handled. If the client demands action before diagnosis is complete, distinguish a reversible, low-cost intervention from one that is hard to reverse. Pausing spend on a demonstrably broken ad can be a temporary measure; moving the full budget on the basis of one week needs stronger evidence.

At the end of scoping, the consultant reads the charter back to the commercial lead, measurement owner, and channel operator. Ask each to correct facts and constraints in writing, then preserve disagreements instead of concealing them in a single average. A sound scope lets each person answer, “Which decision will this study support, and what evidence could change it?” If they cannot, the next task is a decision-framing session, not another dashboard export. Scope can change when evidence reveals an inventory or service problem; record the change and its effects on time and permissions before extending the work.

Set an evidence threshold proportional to the decision. Reproducing a broken ad link and confirming its intended destination may justify a repair. Reducing a limited weekly spend needs comparable days plus checks for stockouts and tagging changes. Retiring an entire acquisition channel requires a longer observation window, credible measurement of direct and assisted effects, and an estimate of opportunity cost. State in advance what you will do if evidence contradicts management's expectation: run a small test or defer the decision? This protects the assignment from becoming a rationale for a predetermined option. Report confidence by conclusion and name missing data, rather than assigning one vague confidence level to every hypothesis.

Lesson summary

A capable consultant begins with the decision the work will support, rather than the channel named by the client. The owner of a fictional homeware st