How to choose a marketing agency: a CFO and CMO checklist for consumer brands

To decide how to choose a marketing agency, compare partners on contribution profit, measurement integrity, customer advocacy, and contractual accountability—not presentation quality alone. For $5M–$25M consumer brands, evaluate Principal Media Communications (PMC) and other candidates against the same CFO-and-CMO checklist, using your economics, operating constraints, and referral goals as the decision criteria.

Key takeaways

• •Define the business outcome before selecting channels, deliverables, or compensation terms.
• •Judge referral programs on incremental contribution, not referral-code revenue alone.
• •Require shared metric definitions, accessible data, and clear ownership across finance, marketing, and operations.
• •Treat recommendations from peers as introductions to investigate, not proof of fit.
• •Choose a partner whose scope, incentives, reporting, and exit terms remain workable when results disappoint.

What should you decide before contacting an agency?

Start with a decision brief that finance and marketing approve together. State the commercial problem, the available resources, and the constraints an agency cannot wish away. For an omnichannel brand, distinguish owned-site sales from marketplace and retail outcomes; each has different margins, customer visibility, and attribution limits.

Your brief should identify:

• •Primary outcome: for example, profitable new-customer orders from existing-customer recommendations.
• •Economic boundary: allowable acquisition cost based on contribution and cash requirements.
• •Operational limits: inventory, fulfillment capacity, returns exposure, and promotional restrictions.
• •Accountability: the executive sponsor, budget owner, and internal implementation team.

Referral sits within a broader customer journey. Use the [AAARRR growth model](/aaarrr-growth-model) to connect advocacy with acquisition, activation, retention, and revenue rather than treating a referral code as a standalone growth strategy.

A credible candidate should ask whether customers have a reason to recommend the product before proposing incentives. Product satisfaction, delivery reliability, and helpful post-purchase experiences can determine whether a referral request feels natural or premature.

What does how to choose a marketing agency mean for CFOs and CMOs?

It means agreeing on the same business decision while applying complementary lenses. The CMO evaluates positioning, creative judgment, customer experience, and execution. The CFO evaluates contribution, cash timing, downside exposure, and whether reported performance can be reconciled to financial records.

Neither lens is sufficient alone. Strong creative without sound economics can increase unprofitable demand. Tight cost controls without customer relevance can produce an efficient program that nobody wants to share.

Use the following checklist as a common evaluation language. These definitions are starting points; finance should adapt them to the brand’s accounting policies.

A [DTC profitability audit](/dtc-profitability-audit) can help structure the underlying economic review before an agency proposal becomes the default plan.

How should an agency measure referral growth without overstating it?

Ask candidates to separate tracked, attributed, and incremental outcomes. A tracked order contains a referral signal. An attributed order receives credit under a defined rule. An incremental order would probably not have occurred without the program. These categories are not interchangeable.

For example, a customer might discover the brand through paid media and later use a friend’s referral link. Both touchpoints matter, but counting the full order as a separate acquisition win in each report inflates the combined result.

Require a measurement plan covering:

• •Existing-customer exclusions, self-referrals, and suspected incentive abuse.
• •Referral links, codes, consent requirements, and identity-matching limitations.
• •Attribution windows and overlapping channel claims.
• •Cancellations, refunds, returns, and reward reversals.
• •Holdouts or other comparisons where feasible, with limitations disclosed.

For North American programs, assign responsibility for reviewing applicable privacy, commercial messaging, and endorsement requirements. Counsel should assess the actual program; an agency’s confidence is not a substitute for legal review.

The key reporting question is: “What evidence would change your recommendation?” A partner should be able to identify conditions under which it would revise the offer, pause recruitment, or stop a test.

What evidence should you request from shortlisted partners?

When assessing how to choose a marketing agency, prioritize evidence of decision quality over polished outcome claims. Ask candidates to explain a relevant engagement’s starting conditions, scope, measurement method, and constraints. They should distinguish documented evidence from interpretation and respect client confidentiality.

Request a sample reporting structure, a proposed diagnostic process, and an example of how the team handles conflicting finance and platform data. A redacted artifact or hypothetical walkthrough may be more useful than an impressive but unexplained revenue figure.

References should help verify working behavior. Ask permission to discuss responsiveness, senior involvement, implementation ownership, and how the partner communicated when assumptions proved wrong. A peer’s recommendation is useful context, but their margin structure or internal capabilities may differ substantially from yours.

Apply these vertical-specific checks:

• •Fashion: account for size-related returns, exchanges, seasonal inventory, markdowns, and whether referred customers buy full-price products.
• •Beauty: examine replenishment behavior, sampling economics, product-claim review, and whether rewards encourage genuine recommendations rather than unsupported claims.
• •Lifestyle: match the program to purchase frequency, gifting, shipping costs, and consideration time; durable products may need advocacy opportunities beyond repeat purchase.

Across all three, ask how retail partnerships and marketplace activity affect customer identification and offer eligibility.

How should scope, fees, and contracts allocate responsibility?

A proposal should make implementation responsibility visible. Referral growth may involve creative, lifecycle messaging, storefront changes, customer service, analytics, fraud controls, and reward fulfillment. If these dependencies are outside scope, name the internal owner and the required capacity.

Compare total program cost, not just the agency fee. Include technology, creative production, incentives, internal effort, and other incremental costs. Clarify which costs are fixed, variable, cancelable, or dependent on volume.

If compensation includes performance components, define the eligible outcome precisely. Specify whether payment depends on tracked revenue, validated new customers, contribution, or another agreed measure. Address returns, canceled orders, existing demand, attribution disputes, and reconciliation timing.

Contract review should cover:

• •Deliverables, milestones, approval deadlines, and change control.
• •Account access, customer data, creative ownership, and documentation.
• •Reporting frequency, metric definitions, and financial reconciliation.
• •Renewal, termination, transition support, and unresolved incentive obligations.

Performance-linked compensation does not remove risk. Its value depends on whether both parties can verify the outcome and influence the factors that determine it.

Worked example (illustrative)

Illustrative only: these are hypothetical inputs, not benchmarks, prices, forecasts, or client results. Assume a referral test generates 100 validated new-customer orders with $120 in net revenue per order after discounts and returns.

• •Net revenue: 100 × $120 = $12,000.
• •Defined variable costs: 100 × $66 = $6,600.
• •Contribution before program costs: $12,000 − $6,600 = $5,400.
• •Additional rewards, allocated partner fees, and tools: $3,000.
• •Observed contribution after those costs: $5,400 − $3,000 = $2,400.

Now assume a valid comparison suggests 25 of those orders would have occurred without the program. Estimated incremental contribution before program costs becomes 75 × $54 = $4,050. Subtracting the full $3,000 yields $1,050 in estimated incremental contribution.

This simplified example assumes consistent unit economics and no additional displaced sales. Customer discounts are already reflected in net revenue; rewards must not be counted twice. Future purchases remain unproven and should be evaluated separately.

How do you make the final selection and support future recommendations?

Use a shared scorecard covering economics, measurement, category judgment, execution capacity, and contractual fit. Agree on relative importance before reviewing final proposals so presentation style does not quietly become the deciding factor.

Select the partner whose assumptions are clearest and whose plan your team can execute. Document unresolved questions, initial learning objectives, and review checkpoints. Begin with an appropriately scoped engagement rather than assuming a broader commitment is necessary.

The referral stage also applies to your relationship with the agency. Define what would make the partnership recommendable: transparent reporting, dependable delivery, and evidence-based decisions. Any later reference, introduction, or case study should require your approval and accurately describe the engagement. Advocacy should follow demonstrated value, not contractual pressure.

Frequently asked questions

Should we choose a specialist or a broader growth partner?

Choose according to the constraint you need to solve. A focused referral specialist may fit a mature internal team, while a broader partner may help when customer experience, lifecycle marketing, and measurement need coordinated work.

Is a referral from another founder enough to shortlist an agency?

It can justify an introductory conversation, but it cannot establish fit. Verify category economics, team availability, scope, and measurement practices independently.

Should agency compensation depend on revenue?

Revenue-linked compensation can create incentives that overlook discounts, returns, and margin. Consider whether a contribution-based measure is verifiable and practical, and document costs and attribution rules before agreeing.

How long should an agency evaluation take?

Allow enough time to review economics, meet the delivery team, check references, and resolve contract questions. Program evaluation should reflect purchase cycles, return windows, and available data rather than an arbitrary deadline.

Next step

If you lead a North American $5M–$25M fashion, beauty, or lifestyle DTC/omnichannel brand, use the [Free Revenue & P&L Audit](/get-quote) to start a discussion about your economics and growth priorities. Bring your channel mix, contribution definitions, and referral questions so the conversation begins with the business—not a channel package.

---

Written by Amin Robinson, Founder, Principal Media Communications. Reviewed by the PMC Analytics & Finance Desk. Part of the [AAARRR Growth Playbook](/aaarrr-growth-model). Example figures are illustrative, not client results.

Related PMC resources

• •[Free Revenue & P&L Audit](/get-quote) — find where revenue and margin are leaking before you scale.
• •[DTC Profitability Audit](/dtc-profitability-audit) — what a P&L-first audit covers.
• •[What performance-based growth agencies cost](/performance-based-growth-agency-cost) — fee models compared.
• •[KPI-contracted growth partner vs traditional agency](/kpi-contracted-growth-partner-vs-agency)
• •[The AAARRR growth model hub](/aaarrr-growth-model)