Four checks, none of which appear in a monthly report: whether the rights register is current, whether problems reach the brand from the agency first, whether the creator roster is evolving, and whether reporting has changed shape since month one. Campaign metrics are the weakest signal available.
Table of Contents
- Check one: is the rights register current
- Check two: who tells the brand when something goes wrong
- Check three: is the roster evolving
- Check four: has the reporting changed shape
- What is a weak signal
- What to do with a poor result
- How often to run this
- One check that covers the other four
- Program Delivery and What an Evolving Program Looks Like
- The HireInfluence Model for Ongoing Programs
Judging an incumbent is harder than choosing one, because the evidence arrives pre-packaged. A monthly deck is assembled by the party being assessed, from metrics it selected, against goals it helped set. That is not dishonesty, it is what reporting is. Properly evaluating influencer agency performance means looking at things the reporting cycle does not produce, and all four below are answerable in a single conversation.
Check one: is the rights register current
Ask for it, unannounced, and read it.
A good register lists every creator, the platform, the permission type, the grant date, the term, and the expiry. It should be current to within days, not months, and the brand should be able to obtain it without a week’s notice.
The specific question that separates a real register from a spreadsheet assembled on request: which creators have usage rights expiring in the next sixty days, and what is the plan for each? An agency running this properly answers immediately. An agency that does not have it will produce something within a few days that has been built for the purpose.
This matters more than it sounds. TikTok authorization expires terminally and cannot be reactivated. YouTube conveys no advertising rights through linking at all and enforces no expiry, so the register is the only control that exists. Instagram permission is revocable at any time by either party.
Check two: who tells the brand when something goes wrong
The strongest single indicator of agency quality, and it costs nothing to observe.
In a healthy relationship, problems arrive from the agency: a creator has gone quiet, a permission was declined, an authorization is expiring before the flight ends, a piece of content did not perform and here is the read. In an unhealthy one, the brand discovers problems from elsewhere and raises them.
Every program of any length generates problems. An agency reporting none is not running a smooth program, it is filtering. The useful test is retrospective: over the last quarter, how many issues did the agency raise before the brand noticed them, and how many ran the other way?
Check three: is the roster evolving
A creator roster that has not changed in a year is a sourcing function that has stopped working.
Good programs churn deliberately: creators who performed are retained, creators who did not are replaced, and new candidates enter each wave. A static roster usually means the agency is renewing what is easy rather than sourcing what is right, which is comfortable for everyone and slowly degrades performance.
The question to ask: which creators were added this quarter, which were dropped, and why in each case? Specific reasons indicate a live process. General ones indicate a renewal habit.
Check four: has the reporting changed shape
Reporting that looks identical in month twelve to month one is a signal that nothing has been learned.
A program generating insight produces reporting that evolves: new comparisons appear, weak metrics get dropped, and the questions being answered get sharper. Reporting that repeats the same template with new numbers is describing activity rather than learning from it.
Worth checking alongside this: does the reporting say what did not work? An agency that only reports upward is not giving the brand what it is paying for.
What is a weak signal
Campaign metrics in isolation. Selected by the agency, unbaselined, and heavily influenced by budget and timing rather than agency quality.
Responsiveness. Fast replies are pleasant and uncorrelated with whether the register is current.
Creative quality in isolation. Widely available across the category and rarely the thing that fails.
A pitch-quality relationship at renewal. Attention that increases near a renewal date is describing the renewal rather than the work.
What to do with a poor result
If the register is stale, this is fixable and specific. Give it a deadline and a named owner on the agency side, then re-check unannounced in a month.
If problems are not being surfaced, address it directly and explicitly. Some agencies filter because a previous client punished disclosure, and saying plainly that raising problems is wanted can genuinely reset it.
If the roster is static and the reporting is unchanged, this is a bigger conversation about whether the program has been on renewal autopilot, and it is usually a scope and staffing question rather than a competence one.
How often to run this
Quarterly is enough, and it should be scheduled rather than triggered by dissatisfaction.
A review that happens only when someone is unhappy arrives loaded, and the agency responds to the mood rather than the substance. A standing quarterly check on the four items above is unremarkable, cheap, and catches drift while it is still small. It also gives a good agency a regular opportunity to show work that a monthly performance deck has no room for.
One check that covers the other four
Brands short of time can substitute a single request: ask for a written account of what the agency would do differently if the last quarter were run again.
A good agency produces something specific and slightly uncomfortable, naming a casting decision, a timing call, or a measurement gap. A weak one produces a restatement of results with the word optimization attached. The exercise takes the agency an hour and tells the brand more than a quarter of reporting.
Program Delivery and What an Evolving Program Looks Like
The #CoatYourThroat program for Ricola generated 26M impressions, and the campaign is documented in full in the Ricola case study.
The #OREOShamROCKout program for Oreo and McDonald’s returned 1.7M impressions at $0.06 cost per engagement. For MTV, the #MyMTVStyle program returned 16.1M impressions and 216,600 engagements at $0.01 CPV and $1.50 CPM.
Additional campaign detail is published in the work portfolio.
The HireInfluence Model for Ongoing Programs
Founded in 2011, HireInfluence is a full-service influencer marketing agency built for enterprise brands, headquartered in Houston with offices in Austin, Los Angeles, and New York. The firm runs creator programs for brands including Adidas, Coca-Cola, Microsoft, Target, Warner Bros, and eBay, covering strategy, talent sourcing, branded content production, paid amplification, and performance reporting. Creator selection runs through a manual vetting and validation process rather than database filtering alone, and campaigns are scoped to each client’s objectives rather than sold as fixed packages.
Brands reviewing an incumbent should read the influencer content approval process and how to measure influencer brand lift, since measurement design is where a reporting cycle either learns or repeats. Scoping conversations start through contact.
