Influencer agencies mistakes: buyer-side setup lessons

Complaints about an influencer agency often describe something the agency did. Influencer agencies mistakes often start on the buyer’s side: a brief that withheld the real constraint, a scope that counted deliverables, or a payment path nobody tested. Those are the mistakes worth reading about, because they are the ones still in your control.

What to take away

Leaving the constraint out of the brief

A brief is a piece of requirements analysis, and the requirements that get left out are always the negative ones. The brief says who the audience is and what the product does. It does not say that one category of creator is off limits because of a legal position, that a named competitor relationship rules out creators in a vertical, or that a senior person has already vetoed a format that nobody wrote down.

The shortlist arrives, and some options are unusable for reasons the agency could not have known. That rework costs time and credibility, because the next round comes back cautious, and cautious shortlists are boring shortlists.

Write the constraints page before the creative brief. Who has to approve, what can never appear, which competitors are out, what has been tried and failed, and who is allowed to say no at the end. It is a short document and it is the highest-value page you will write for the campaign.

Buying deliverables when you needed a decision

A scope that promises a fixed number of posts and a report is a purchase order. It can be delivered perfectly by people who never once ask whether that output was right, because nobody in the arrangement is paid to argue for less.

The alternative is not vaguer, it is more specific. Name the decision the campaign should let you make: whether this product line justifies another wave, whether this audience responds to demonstration over endorsement, whether a partner is worth a sustained relationship rather than a short burst.

Then write the stopping rule that ends the spend early if the answer arrives early. Agencies rarely refuse this. It changes what they staff the account with. Staffing depends on how the agency is structured, as the pillar guide to influencer agency models explains.

Complaints, and where they usually start

The pattern in that middle column is worth sitting with. None of these are performance failures. They are setup failures that only become visible once performance is being discussed. Row two is the one to fix first: metric definitions belong in your own measurement stack, frozen before anything runs.

Meetings that report activity and decide nothing

The status call is the default rhythm and close to useless for this work. It reports what happened, which you can read, and it arrives when nothing can be changed anyway.

A decision meeting belongs at the point where commitments become irreversible, usually when contracts are about to be signed and the shortlist is final. That meeting exists to kill things cheaply.

A debrief belongs after the final invoice, not after the final post, and it asks what the campaign taught you about the audience rather than what it delivered. Put both meetings in the calendar at the start, because neither gets scheduled once the work is running.

Two suppliers, one undrawn boundary

An agency plus a production company, or an agency plus a separate media team, is a normal arrangement that goes wrong in a predictable way, and where the second supplier makes the footage rather than places it, commissioning short form content sets out what the brief has to specify. Both parties are reasonable. Neither owns the calendar, and the creator receives conflicting versions of the same brief.

Draw the boundary as a list of decisions rather than a list of tasks. A service-level agreement between two of your own suppliers sounds excessive until you see the time it can save. Who chooses the creator. Who signs the contract. Who approves the cut. Who is called when a post goes live with a broken link. Tasks overlap harmlessly. Decisions with two owners are how work gets delayed.

The same problem appears in tooling. Paying for a search seat while also paying an agency to source is common and rarely deliberate. Before renewing anything, check whether the agency already does the job you are buying software for, and whether the record you need is ending up in your own systems or in theirs. What a seat actually buys, and what it cannot, is set out in the guide to creator discovery tools.

When the mistake is choosing between the wrong options

The comparison that gets run is agency A against agency B. A more useful comparison asks what kind of company you are buying from and which parts of the work it owns.

Compare the buying model as well as the company. A retainer buys ongoing capacity; a project fee covers defined work; a platform fee pays for access to a tool; and a managed service bundles execution. Check which model covers creator sourcing, approvals, usage rights, reporting, and payment responsibilities. The label alone does not establish what is included.

Other habits are worth naming. A paid test spread across several agencies divides your attention and deprives each of a real brief.

Treating measurement as something the agency provides means the numbers arrive already framed. Deciding what counts as a result, and where that figure comes from, belongs on your side of the table, and it is worth settling before the first campaign rather than after the third.

Common questions

The agency says our approval process is the problem. Are they right?

Probably in part. Count it rather than argue about it: how many rounds, how many days each, how many people can request a change and how many can approve one. If more people can block than can decide, the process is the problem whoever raises it.

Should we pay an agency to write our brief?

Yes, and say that is what you are buying. A brief written for free during a pitch is written to win the pitch. One that is paid for can tell you the campaign should be smaller.

We inherited an agency relationship with no records. Where do we start?

With rights and dates. Ask for every signed contract and every asset license with its expiry, and put them somewhere you control before you renegotiate anything else. Check that the agreements state the exclusivity window, usage term, paid amplification or whitelisting rights, territory, and whether extensions require written approval. Performance history is nice to have. Knowing what you are still allowed to run is not optional.

Is it a mistake to keep changing agencies?

Only if you are changing for the same reason each time. A repeated failure across suppliers is usually a fault in your brief, your approval chain or your payment process, and another agency will meet it too. Comparing suppliers properly is its own discipline, covered in the guide to running a vendor comparison.