In-house Marketer Versus Agency Decision Checklist

Choosing between an in-house marketer and an agency is a delivery-model decision. It changes ownership, response time, context retention, management load, and the organisation’s ability to learn. Use this checklist before replacing a person, signing a retainer, or assuming that a new supplier will repair an undefined problem.

1. Name the decision and the failure

Write the exact change under consideration: hire, retain, replace, add a specialist, or keep the current model. State the business period, service line, audience, and decision owner. Then describe the observed failure in operational terms: missed follow-up, weak positioning, slow approvals, inconsistent reporting, insufficient expertise, or lack of capacity.

Separate an output complaint from an outcome complaint. “Not enough posts” is an output observation. “Sales cannot use the material in qualified conversations” is an outcome or handoff observation. The remedy can differ completely.

2. Map required capabilities

List the capabilities the next quarter actually requires: research, offer design, channel execution, analytics, creative production, technical SEO, CRM operations, sales enablement, or project management. For every capability, record the required depth, weekly demand, current owner, evidence of competence, and backup owner.

Use the SBA guide to managing a business as a reminder to include people, operations, finances, compliance, and continuity in the operating review. It is a planning boundary, not proof that one delivery model is universally better.

3. Test context and availability

An internal owner may hold valuable product and customer context but lack time or a specialist skill. An agency may provide a broader bench but require more written context and coordination. Record how quickly each model can answer a product question, join a sales review, correct a tracking issue, and approve a risky claim.

Do not count nominal hours as usable capacity. Subtract meetings, review cycles, interruptions, holidays, onboarding, and founder dependency. If the work needs daily judgement but the proposed model offers only a monthly meeting, record the mismatch before discussing price.

4. Verify the handoff path

Map brief, production, review, launch, measurement, lead routing, and learning. Name the owner at every transition. For lead-related work, use the Salesforce lead implementation guide to prompt explicit definitions for qualification, ownership, conversion, disposition, and reassignment. Adapt the fields to the local process; do not import a CRM definition without agreement.

Hold the decision if the agency would deliver assets but nobody owns sales feedback, or if an employee would own strategy but lacks access to the data required to evaluate it.

5. Compare evidence, not promises

Request a small, comparable work sample: a diagnosis, brief, measurement plan, or review of an existing asset. Ask what evidence the provider used, what it could not know, which assumption it would test, and what would make it stop. A polished deck is not evidence of delivery quality.

For measurement work, define the event and parameter before the work begins. Google Analytics event guidance can help structure the event record, but local business meaning, consent, retention, and CRM reconciliation still require an owner.

6. Price the whole operating model

Record cash fee or salary, management time, software, training, recruitment or termination cost, review time, rework, data access, and the cost of delayed decisions. Do not invent a universal “agency premium” or “in-house saving.” The right comparison is the cost of a defined capability at a defined service level.

Add concentration risk. An internal specialist can become a single point of failure; an agency can become a single vendor dependency. Ask who retains the files, accounts, prompts, measurement definitions, rights, and decision history if the relationship ends.

7. Check rights, access, and confidentiality

Specify ownership of creative files, source data, accounts, audiences, dashboards, credentials, documentation, and derivative work. Separate access needed to work from access needed to administer. Record approval for customer data, confidential pricing, regulated claims, and subcontractors.

Hold if the proposed model depends on undocumented personal accounts, shared credentials, or a contract that cannot explain exit and data return.

8. Run a bounded pilot

Choose one business problem, one audience, one owner, one review date, and one evidence path. Define the baseline, expected learning, acceptance criteria, and stop rule. A pilot should be small enough to stop without leaving the core funnel ownerless.

Use a two-column control sheet: “what this model can prove” and “what remains unknown.” Record a credible alternative such as repairing the brief, changing the offer, improving response time, or reallocating internal capacity. If the pilot cannot distinguish the model from the alternative, do not convert it into a retainer or headcount request.

9. Make the operating decision

| Gate | Required record | Hold if | | — | — | — | | capability | skill, depth, owner, backup | need is described only as “more marketing” | | capacity | usable hours and response expectation | availability is nominal | | handoff | brief, review, sales, measurement owners | an owner disappears at a transition | | evidence | work sample, assumption, test | credentials are the only proof | | economics | full cost and displaced work | management or exit cost is hidden | | rights | files, accounts, data, access, exit | ownership is ambiguous | | pilot | baseline, review, acceptance, stop | the change cannot be reversed |

Choose one of five outcomes: keep and repair, add a specialist, run a pilot, change the model, or hold. Record the work displaced by the decision and the next review date. The checklist is complete only when someone outside the selection conversation can reproduce why the decision was made.

10. Preserve the fallback

Document the minimum viable internal capability, the vendor exit path, and the person who can keep critical operations running during a transition. Store the brief, sample, definitions, access map, and decision note in a shared location. A delivery model is resilient when learning and ownership survive a personnel change.

Keep this checklist local and non-indexable until current sourcing, contract, privacy, overlap, and editorial review are complete. It is a decision aid, not a guarantee that an in-house team or an agency will produce a particular commercial result.

Your reaction

How did this article land?

Choose one reaction. You can change it anytime.

Email verification required

Write for Scale Orbit

Turn practical experience into a public body of work

Share useful lessons about revenue, marketing, analytics, CRM, conversion, and growth. Build a visible author profile and learn what resonates with practitioners.

  • Public author profile and publication archive
  • Editorial support for your first article
  • Views, reactions, followers, and topic discovery
  • Free publishing with clear moderation rules

Email verification is required. Every first article is reviewed. Publication, rankings, traffic, leads, and revenue are not guaranteed.

Write

Discover more from Scale Orbit | Full-Service Marketing Management

Subscribe now to keep reading and get access to the full archive.

Continue reading