What a VP of Marketing Should Demand from Their Agency

Some agency relationships produce a steady stream of activity, campaigns launched, content published, reports sent, without anyone being able to say what it did for pipeline. When that happens, it is rarely because one side is incompetent. More often the relationship was set up around deliverables rather than outcomes, with nobody agreeing what success would look like or how it would be checked.

This article is for the person on the client side, whether that is a VP of Marketing, a Head of Growth or a founder, who is about to hire an agency or wants an existing relationship to work better. It covers five things worth asking for. Each is also good for the agency, because each makes good work visible and problems easier to fix early. And each asks something of you too.

Why agency relationships drift

• Scope is defined by deliverables. The brief lists outputs: so many campaigns, so many articles, a weekly report. It says nothing about the commercial result the work is meant to support. The outputs get delivered. The result may or may not follow, and nobody is quite sure whose job it was to notice.
• Nobody agreed how progress would be judged. Without shared definitions of a qualified lead or opportunity, the agency reports what its platforms show and the client measures what the CRM shows. The two never quite meet.
• Problems surface late. If the only serious review happens after months of work, small problems have had plenty of time to grow.

Each of these is fixable at the start, and most are fixable mid-relationship too.

1. A shared commercial goal, and measures the agency owns

Start with the commercial goal the work is meant to support, such as more qualified pipeline from a particular segment. Write it down together. Then agree the measures the agency is directly responsible for, the ones its work can move and that sit on the path to the goal: qualified opportunities from the channels it runs, for example, or conversion through the stages it influences.

The distinction matters. The commercial goal belongs to the business, because sales, product and pricing all affect it. The agency's measures should be things it can genuinely influence. Holding an agency to a number it cannot move produces either excuses or gaming. Holding it to nothing produces drift.

Agree too how the measures will be read. Small volumes produce noisy numbers, and B2B marketing attribution covers how to match the evidence to your volume, from recorded sources to deal-by-deal review and what buyers say. OKRs for growth teams covers how to write measures a team can own.

What you bring: clear definitions in your CRM, honest data on what happens after a lead reaches sales, and a named person on your side who owns the commercial goal.

2. A diagnostic before execution

Before committing to a long programme of execution, ask for a diagnostic: a written view of where the funnel is breaking, why, and what would need to change. Not a sales deck and not a discovery call, but a document you can disagree with.

A good diagnostic shows whether the agency understands your business before it starts spending your budget, and it gives both sides a shared starting point to measure progress against. Sometimes it shows that the main constraint lies outside what the agency would do, in the handoff to sales or in positioning, which is useful to know early. The full-funnel audit shows what that kind of diagnostic covers.

What you bring: access to data, time with the people who sell and serve customers, and willingness to hear things about your own business that may be uncomfortable.

3. Direct access to the data and the work

You should be able to see the ad accounts, analytics, CRM data and work in progress directly, not only through curated reports. Ideally the accounts sit in your name, so what is built and learned stays with you.

Open access helps both sides. Problems surface early, good work is visible as it happens, and conversations are about the same numbers. It also means the knowledge stays in the business if people or partners change. When to hire in-house versus work with an agency makes the same point: whoever does the work, ownership of the goals, measurement and learning stays inside the company.

What you bring: someone who will actually look, and who asks questions rather than just receiving reports.

4. Monthly working reviews and quarterly decisions

Two rhythms do different jobs.

• A monthly working review looks at the real data together: what the agency sees, what concerns it, what it is changing and why. Less polish, more candour. Issues surface while they are still small.
• A quarterly review steps back and makes decisions: which initiatives to scale, which to keep testing, which to stop and what to change in the plan. Quarterly business reviews that change the plan covers how to make that meeting produce decisions rather than a performance.

Agree in advance what happens if a measure drifts well off course: who looks into it, how quickly, and what options are on the table. A pre-agreed response turns an awkward conversation into a routine one. Results from small tests will often be inconclusive, and test-and-learn that compounds covers how to report that honestly.

What you bring: time in the diary, the data the agency cannot see (such as what sales thought of the leads), and decisions when they are needed.

5. One named accountable lead

Agencies usually staff accounts with several people: a strategist, specialists, an account manager. That brings breadth, but it can blur accountability. Ask for one named person who is responsible for whether the work is moving the agreed measures, and who will explain what is happening and what is changing when they slip.

This is easier for the agency too. Everyone knows who coordinates the work, and you know who to call.

What you bring: a matching named owner on your side. Two accountable people talking regularly beat two teams sending each other reports.

When the client is a founder or a small team

Everything above still applies, but the shape changes. A founder working alone, or with a very small team, has little time to brief, review and approve. A good agency adapts its brief, execution and delivery process to that, rather than expecting the client to adapt to the agency. It drafts the brief from a short conversation for the founder to correct, brings only the decisions that genuinely need the founder, and keeps reviews short and focused. Growth support for founders still proving their market covers this in more detail.

If the conversation gets difficult

Some of these requests may meet resistance. That is not automatically a bad sign: the agency may have good reasons, and the conversation is worth having.

• "We cannot be held to a pipeline number." Fair, if sales and pricing are outside its control. Then agree measures it can influence on the way to pipeline, and how both sides will read them.
• "Our tools do not allow client access." Ask what can be shared, and whether accounts can be set up in your name.
• "Monthly reviews are too much overhead." Agree a shorter format. The point is a regular look at real data, not a long meeting.

Partial agreement is common and often fine. What matters is that both sides know what was agreed and why.

Where it sits

These requests turn an agency relationship into a shared operating rhythm: a goal, evidence, regular review and decisions. That is the same rhythm a growth team needs internally, whatever its make-up. Building a growth team covers which capabilities to hire for and in what order, and the broader choice between a consultant, an agency and an in-house team covers the model itself.