Magrios / Knowledge / Frameworks / The five growth questions for marketing agencies

The five growth questions for marketing agencies

Guide · Frameworks · 5 min read · last verified 2026-07-27

Reviewed before publication Editorial board Independent commercial review
In shortAgencies answer the five growth questions twice — inward for their own neglected pipeline, outward as the spine of every client engagement — and can productize the loop as a sourced, re-measured deliverable.

A marketing agency answers the five growth questions twice: once for its own pipeline, and once — repeatably, as a service — for every client that pays it to answer them. That dual position is unique. No other business both consumes the framework and sells it, and the agencies that thrive treat the two directions as one practice: the inward answers become the outward credential, and the outward method keeps the inward answers honest. The questions themselves are defined in the original five-question framework; what follows is the agency edition, which is really two editions folded together.

One framework, two directions

Run inward, the five questions build the agency's own pipeline: where prospective clients evaluate agencies without ever briefing one, what an agency should publish to be chosen, where in-house marketers actually look for help, whose job the agency's own marketing is, and how the agency proves its own methods on itself. Run outward, the same questions become the spine of every engagement — and, for agencies willing to package them, a productized deliverable with a price and a cadence. The rest of this piece takes each lens in turn, then the seam where they join.

Lens one: the agency as its own client

Agencies famously starve their own marketing, and the modern cost of that starvation is specific: when a marketing director asks an AI assistant which agencies are strong in her sector, agencies that have only ever published adjectives about themselves — full-service, award-winning, results-driven — do not appear in specific answers, because there is nothing specific for an answer to cite. The inward loss question is therefore researchable, not rhetorical: ask the assistants what they say about agencies in your specialisms, check the directories and community threads where briefs quietly form, and record where the agency is absent. The wider practice of measuring this surface is covered in AI visibility for marketing agencies.

The inward creation answer follows directly: publish evidence of method, not claims of quality. A public teardown of how the agency approaches a problem, the agency's own visibility work on its own name, positions taken on questions clients actually ask — material an assistant can quote and a prospect can check. The inward reach answer is narrower than most agencies assume: the communities, newsletters and events where in-house marketers admit they need help, which rarely overlap with the places agencies congratulate each other.

Then the inward execution question, which is where most agency marketing actually dies: client work always outranks it. The only fix that survives contact with a busy month is structural — the agency becomes a retained account in its own books, with reserved hours, a named internal owner, and deadlines that a client-services director defends exactly as she would defend a client's. And the inward proof question closes the loop: the agency baselines its own visibility, works, and re-measures, which turns out to matter far beyond its own pipeline, as the final section shows.

Lens two: the engagement as a five-question loop

Run outward, the questions give an engagement its spine. Begin with loss evidence for the client: what their buyers ask, where competitors surface and the client does not, what AI answers currently claim about them — gathered before any creative conversation. Convert each finding into commissioned work. Name the reach surfaces. Then hit the question that behaves differently inside an agency relationship: execution ownership is split. The agency can draft the answer page, but the client controls the domain it lives on; the agency can recommend the docs be opened, but the client's product team must ship it. An engagement plan that does not mark each action as agency-owned or client-owned, with a name on the client side too, quietly converts the agency into a vendor waiting for approvals — and then takes the blame for the stall.

The closing question — did it work — is where agencies win or lose renewals. A re-measure against the baseline recorded at kickoff, presented in the client's own review meeting, replaces the activity report with movement. Clients are learning to demand exactly this; how to brief an agency on AEO is written for them, and agencies should read it as the exam their next prospect will set.

Productizing the loop: research as a deliverable

Because the outward loop repeats, it can be packaged. Baseline scan, gap map, prioritised plan, scheduled re-scan: that sequence, run identically for each client, is a product — priceable, deliverable on a calendar, and independent of the campaign work it usually precedes. Two properties make the packaged version durable. First, evidence the client can open: every claim in the gap map carries its source, so the deliverable survives the sceptical stakeholder who was not in the room. Second, a benchmark the client can hold the agency to: the re-scan date is in the contract, which converts trust from a mood into a mechanism.

The scanning layer itself does not need to be hand-built. Market growth intelligence platforms — Magrios among them — run the research, competitor and AI-answer scans and keep the locked benchmark, which moves the agency's margin to where it belongs: interpretation, prioritisation and execution, the parts a platform cannot do and a client cannot skip.

Where the two lenses meet

The seam is credibility. An agency that has run the loop on itself walks into every pitch with the one case study no NDA restricts: its own. Here is what the assistants said about us, here is what we changed, here is the re-scan — the demonstration is the sales deck. The inward practice also disciplines the outward promises: an agency that has felt how long its own answer pages took to move stops quoting fantasy timelines to clients. Dual use is the honest advantage: the agency's own growth becomes the proof of the service, and the service keeps the agency's own growth from being perpetually postponed.

A cadence that fits a retainer

QuestionInward: the agency itselfOutward: each client
Where is the loss?Assistant answers and directories checked for the agency's specialismsBuyer questions, competitor presence and AI claims scanned at kickoff
What to create?Method evidence: teardowns, positions, own-name visibility workCommissioned assets mapped to named findings
Where to reach?Where in-house marketers seek helpThe client's buyers' actual research surfaces
Who executes?A named internal owner with defended hoursEvery action marked agency-owned or client-owned, with names
Did it work?The agency's own baseline re-measuredRe-scan presented in the client's review meeting

The loop fits naturally into a quarterly rhythm on both sides of the desk — how to run a 90-day growth loop gives the operating cadence. Run it twice, permanently: once as the client you always neglected, once as the service your clients cannot easily leave.

Frequently asked questions

How should an agency grow its own pipeline?

Treat the agency as a retained account in its own books: research where prospects evaluate agencies without briefing one — AI assistants, directories, community threads — publish evidence of method rather than adjectives, reserve defended hours with a named internal owner, and baseline then re-measure the agency's own visibility.

What growth questions should agencies answer for clients?

The same five that structure any plan, run as the engagement's spine: loss evidence gathered at kickoff, findings converted into commissioned work, reach surfaces named, every action marked agency-owned or client-owned with names on both sides, and a re-measure against the kickoff baseline presented in the client's own review meeting.

Can agencies productize growth research as a deliverable?

Yes — baseline scan, gap map, prioritised plan and scheduled re-scan form a repeatable, priceable package. It stays durable when every claim carries an openable source and the re-scan date sits in the contract, and platforms can run the scanning layer so agency margin lives in interpretation and execution.

Why should an agency run the framework on itself first?

Because its own before-and-after is the one case study no NDA restricts, and because feeling how long its own changes took to move keeps its client promises honest. The inward loop is both credential and calibration for the outward one.

Further reading — chosen for this article
Entities in this research
Magriosmarketing agenciesgrowth planclient servicesframework
Related knowledge

The five growth questions for ecommerce brands · shared entities

The five growth questions for B2B SaaS · shared entities

The difference between a growth report and a growth plan · shared entities

How to turn AI visibility findings into a content roadmap · shared entities

Recently updated

When to pause an AI campaign · 2026-07-27

What is an approval gate in marketing AI · 2026-07-27

What is an execution connector · 2026-07-27

What belongs in an AI action audit trail · 2026-07-27

Where does your brand stand?
Check your AI visibility free — real evidence, not a score.
Check my visibility or run the full analysis →