The Problem With Agency Deliverables: Turning a Client Brief Into a Commercial Decision

Richard Rumelt's work on strategy emphasizes diagnosis, guiding policy and coherent action rather than treating aspiration as a substitute for a plan. This principle matters particularly to advisory work because clients may arrive with a preferred solution already in mind. An outside team needs enough independence to test the premise without appearing to dismiss the client's experience. The strongest diagnosis usually combines leadership objectives, customer evidence and the constraints observed by people executing the work.

A practical discovery stage should answer four questions. What changed in the business? Which customers, competitors or operations are implicated? What is the cost of staying where we are? Which facts contradict the first proposed explanation? Interview the people who handle sales, support, implementation and finance, not only the sponsor who commissioned the work. Look for where their definitions differ. That disagreement often identifies the real work more quickly than a detailed request for proposals.

Brand is sometimes treated as a world of impressions, sentiment and visual choices isolated from sales. That is too small a view. Positioning influences which buyers can recognize a relevant offer. Reputation influences whether they trust the claim enough to evaluate it. Customer experience influences whether the original promise feels credible after purchase. The financial result depends on many other functions, but that does not erase the commercial mechanism through which brand and marketing can contribute.

Write a causal hypothesis rather than a slogan. If the market confuses this product with an adjacent category, qualified buyers may overlook it or arrive with incorrect expectations. A category clarification could improve qualified evaluation. If public incidents stem from repeated service failure, clearer responses alone are unlikely to restore lasting trust. A joint operational improvement may reduce recurrence. Both hypotheses can be examined, but neither should be sold as guaranteed new revenue before the evidence exists.

Start with direct alternatives. Customers might compare the client to obvious brands, internal processes, freelancers or the decision to do nothing. Compare how each alternative frames the job, whom it serves, what proof it offers, how it charges and how difficult it is to replace. Then examine the customer's actual buying or service journey. Search trends and social discussion can show language and demand patterns, but they are imperfect proxies for the decisions people will make with their budgets.

Useful research connects evidence to a decision. A competitor matrix should inform differentiation, not become a large appendix no one uses. Interviews should reveal how a buyer chooses, not simply collect attractive quotations. Social listening should distinguish unique problems from the amplification of the same incident. For every research activity, record which strategic choice it could change. If it changes nothing, question whether the activity belongs in the scope.

A project scope is a set of expectations about decisions, rights, time and risk. Define what the agency recommends, what it produces, which approvals the client retains, and who owns execution beyond the contract. That is particularly important when the work touches product roadmaps, customer service or revenue reporting. A communication agency cannot authorize a product release. A marketing consultant should not quietly inherit responsibility for sales closure because the brief includes pipeline aspirations.

Convert each major deliverable into an acceptance condition. A positioning system is accepted when the agreed buyer groups, evidence and product boundaries are represented correctly. A campaign is accepted when tracking and routing work, not merely when advertisements appear. A research project is complete when the leadership decision and supporting assumptions are documented. A contract can protect the working relationship by making the boundary of responsibility clear without reducing the commercial ambition.

The number of check-ins says little about whether a client engagement is moving. What matters is whether the right people can answer the difficult questions. A marketing manager may approve copy but lack authority to change a product promise. A customer-operations leader may own a service failure but have no visibility into the agency's incident analysis. A finance owner may understand margins but never see the proposed acquisition assumptions.

Build a responsibility map with the decision, the accountable owner, the people whose evidence is required and the condition for escalation. Keep a decision log that records the options considered, why one was chosen and what would justify revisiting it. That record protects relationships when teams change or the original sponsor becomes less available. It also gives the agency a practical way to distinguish delay caused by missing data from delay caused by unresolved authority.

A content calendar is a scheduling tool. It is not a commercial operating model. Begin with a small number of meaningful hypotheses, determine how they will be tested and specify what evidence can invalidate them. For example, a product-specific landing page may be intended to reduce category confusion among a defined buying committee. The early measures could include correct inquiry routing, relevant engagement and sales acceptance. Page views alone cannot demonstrate that the confusion disappeared.

Review progress as a set of decisions. Continue a program because observed behavior supports its premise. Revise it because a specific assumption has failed. Stop it because the problem no longer matters or the investment cannot be justified. An agency that recommends stopping a poorly performing initiative may create more value than one that keeps every activity active to demonstrate effort. The client is paying for judgment as well as the work needed to carry that judgment out.

A good advisory process listens outside the planning deck. Customer escalations, community conversations, field-sales objections and reviews can identify gaps between intended positioning and experienced reality. But the raw volume of comments is a weak decision metric. One severe and credible issue may deserve more attention than hundreds of repetitive low-impact mentions. The analysis needs issue family, first known occurrence, actual business exposure and a route to a person who can change the condition.

If the engagement includes reputation risk, define who can approve public responses, who owns the incident, and where technical or operational facts are verified. Do not manufacture reviews, promise the deletion of legitimate criticism or convert anonymous incidents into recognizable client stories. The business value of good intelligence lies in giving accountable people time and evidence to act. Reporting that cannot change an operating decision is expensive observation.

The scorecard should mirror the causal chain agreed at the beginning. It can include competitive differentiation adopted, the decision cycle improved, relevant accounts engaged, sales-qualified opportunities progressed, service failures reduced, retained clients or value protected where genuinely measurable. Alongside those outcomes, document the assets and activities used to influence them. Do not declare company-wide revenue personally sourced by a campaign when sales, product and procurement also played material roles.

Preserve the distinction between leading indicators and commercial results. Search growth, share of voice, content volume and response time can explain how a strategy operated. They do not establish new ARR or lower churn without supporting evidence. Use time periods and denominators, record countertrends and state limitations. The best scorecard should survive a client's most skeptical finance or operations colleague because every headline has an appropriate definition.

The work should leave the client more capable of making the next decision. Deliver a documented diagnosis, a coherent set of choices, accessible research, reusable messaging, responsible process ownership and a test plan. Explain what changed and what remains uncertain. If the team needs the agency merely to understand its own data, the engagement may have failed to transfer knowledge. A stronger outcome is a client who can carry the operating principles forward.

This is the distinction I care about in strategic work. A polished deliverable can look impressive on the day it is presented. Durable commercial value appears when the business knows whom it is trying to win, what promise it can defend, which risks it must manage and how it will measure progress. It is acceptable for an engagement to discover that the original problem was misidentified. That discovery can be the most important decision the project produced.