Managing multiple influencer agencies is mostly a counting problem. A brand that runs creators through two or three agencies, an in-house team and a paid-media partner gets reports in views, impressions, CPM and ROAS, each with its own window and cost scope, and cannot say which partner is actually cheaper per customer. This guide gives you the measurement standard, the shared tracker, the scorecard, the budget rules and the contract clauses to fix that, with tables you can copy.
Why don’t reports from different influencer agencies add up?
They don’t add up because each partner chooses its own unit, attribution window and cost scope, and nothing forces them to match. Brands, agencies and platforms use different KPIs, reporting methods and attribution models, and marketers spend their time reconciling data instead of analysing it, according to PPC Land’s summary of the ANA’s August 2026 report on influencer waste43. In the same work, measurement was rated the hardest step of influencer marketing by 67% of respondents4.
The gaps are predictable:
| What differs | Agency A might report | Agency B might report | Why they can’t be compared |
|---|---|---|---|
| Reach unit | Views (video plays) | Impressions (all placements, incl. paid) | Impressions include boosted delivery; views don’t |
| Cost basis | Creator fees only | Creator fees + its own fee + product | One CPM excludes the agency’s cut, the other doesn’t |
| Window | Views at day 7 | Views at day 30 | Posts keep gaining views for weeks |
| Sales credit | Discount-code orders | Platform-reported ROAS | Platform attribution can claim orders the code also claims |
| Platform mix | Mostly TikTok | Mostly YouTube | Cost per view differs by platform before any skill is involved |
Platform attribution is the easiest place to double count. TikTok Shop Ads credit an order to an ad on a 7-day click or 1-day view window and count orders from livestreams, search and non-affiliate links as well; TikTok states it does not distinguish affiliate from non-affiliate creators7. If one partner runs Shop Ads and another runs affiliate creators, both can claim the same order.
Platform mix matters just as much. In Kolect client programs, the median realized CPM (creator fee per 1,000 views actually delivered) was $6.73 on TikTok, $17.96 on Instagram and $74.24 on YouTube12. An agency that works mostly on YouTube will look several times more expensive per view than a TikTok-heavy one, even if it brings in customers more cheaply.
The industry knows this is unsolved: in the IAB’s 2025 creator ad spend study, measurement, standards and operational tools were the top areas brands said needed improvement5.
What should one measurement standard for all agencies include?
A measurement standard is a one-page document, owned by the brand, that fixes how every partner counts reach, results and cost. It has three parts: definitions, one attribution window with one source of truth, and the cost lines.
Shared metric definitions
| Metric | Definition everyone uses | Source of truth |
|---|---|---|
| Views | Video plays shown on the post at the agreed capture day | Screenshot or platform insight export, filed in the tracker |
| Paid impressions | Impressions from boosting (Spark Ads, partnership ads) | The brand’s own ad account |
| Clicks | Sessions from the post’s UTM link | The brand’s analytics |
| New customer | First order from a customer with no previous order | The brand’s store or CRM |
| Revenue | Net revenue of attributed orders, after discounts and refunds | The brand’s store |
| All-in cost | Every cost line below, for that post or campaign | The tracker |
Keep paid impressions in their own column. Most creator deals now carry paid rights: 80.4% of single-post agreed collaborations in Kolect client programs included paid-usage, whitelisting or ad-code rights12. If boosted delivery is mixed into “reach”, the agency with the bigger ad budget wins every comparison. How boosting works on each platform is covered in Spark Ads vs partnership ads.
One attribution window and one source of truth
Pick one window and use the brand’s own analytics or store as the judge; platform ROAS becomes a secondary column, never the ranking metric. Defaults differ between tools: Google Analytics 4 uses a 90-day lookback for most key events and 30 days for acquisition events8, while TikTok Shop Ads use 7 days for clicks and 1 day for views7. Write the choice down, for example: “an order counts for a partner if it uses that partner’s code, or arrives through its UTM link within the agreed number of days of the post; a code beats a link when both apply.” The method for calculating return itself is in our guide to influencer marketing ROI.
Give each partner its own code prefix and UTM value so attribution never depends on who writes the report:
| Field | Convention | Example |
|---|---|---|
| utm_source | Platform | tiktok |
| utm_medium | Fixed value for creator traffic | creator |
| utm_campaign | Brand campaign name | spring-launch-us |
| utm_content | Partner + creator handle | agencyA_creatorhandle |
| Discount code | Partner prefix + creator | AGA-CREATOR |
Cost lines, with creator fee and agency fee apart
Every partner reports the same cost lines, separately. This is where comparisons usually break: one agency quotes “per creator” with its fee folded in, another bills a monthly retainer. In the ANA’s 2026 study of influencer agency compensation, just 51% of marketers said they had full clarity into influencer payments, 31% said their agency used non-transparent compensation methods, and only 25% tied agency pay to specific KPIs21.
| Cost line | What goes in it | Who reports it |
|---|---|---|
| Creator fee | What the creator (or their manager) is paid | Agency, with the creator invoice or contract on request |
| Agency fee or margin | Retainer, project fee, commission or mark-up on creator fees | Agency |
| Usage rights | Paid-usage, whitelisting or licence extensions bought | Agency |
| Product and shipping | Cost of goods and logistics for seeding | Brand |
| Paid amplification | Media spent boosting creator posts | Brand’s ad account |
| Tools and internal time | Software, and in-house hours at an agreed rate | Brand |
How do you set up one tracker for every agency?
Keep one tracker in the brand’s workspace, one row per post, and require every agency and the in-house team to fill the same columns by a fixed day each month. The brand owns the file; agencies get edit access to their own rows.
Copy these columns:
| Column | Column | Column |
|---|---|---|
| Partner (agency / in-house) | Creator fee | Views at capture day |
| Campaign and market | Agency fee or margin | Paid impressions |
| Creator handle and platform | Usage rights bought (and end date) | UTM clicks |
| Post URL | Product and shipping | Code orders |
| Live date (planned and actual) | Paid amplification spend | New customers in window |
| Disclosure checked (Y/N) | All-in cost (sum) | Net revenue in window |
| Brief approved (Y/N) | Reused in ads (Y/N) | Notes |
Keep the data itself in accounts the brand owns:
- Ad accounts. Run Spark Ads and partnership ads from the brand’s ad accounts. In TikTok Business Center an admin can add a partner and set asset- and account-level permissions9; in a Meta business portfolio you assign a partner access to specific assets with full or partial control10. The campaign history stays with you if the agency changes.
- Analytics and store. Agencies get read access or a scheduled export, not a separate property.
- Codes and links. The brand creates codes and UTM links and hands them to agencies.
- Creator records. Contact details, contracts and content licences for your campaigns are filed in the brand’s drive, so a new partner can rebook a creator.
For campaign-level steps before and after this tracker, see how to run an influencer marketing campaign.
How do you compare influencer agency performance fairly?
Compare agencies on all-in cost per new customer from your own data, inside the agreed window, and only then look at efficiency and delivery. Views-based costs are fair only within the same platform and market.
| Scorecard line | Formula | Notes |
|---|---|---|
| Cost per new customer | All-in cost ÷ new customers in window | The ranking metric |
| Cost per 1,000 views | All-in cost × 1,000 ÷ views at capture day | Compare within one platform and market only |
| Content reuse | Posts approved for ads ÷ posts delivered | Content you can run again lowers the next campaign’s cost |
| On-time delivery | Posts live by planned date ÷ posts planned | Late posts miss launches and sales windows |
| Disclosure and brief compliance | Posts passing your check ÷ posts live | The brand stays liable, see below |
| Reporting completeness | Rows with every column filled ÷ rows | Missing cost lines are a finding in themselves |
Engagement rate, impressions and reach are still the KPIs most marketers use to judge influencer content (90%, 87% and 86% of respondents in the ANA study as reported by Marketing Dive)2. Keep them in the tracker as diagnostics, but they don’t tell you which partner is cheaper per customer.
To compare partners working on different platforms, index each partner’s cost per 1,000 views against a platform benchmark rather than against each other; the average influencer CPM by platform gives reference points.
Disclosure belongs on the scorecard because the FTC holds the advertiser responsible: “Delegating part of your promotional program to an outside company doesn’t relieve you of responsibility under the FTC Act,” and advertisers are expected to have reasonable programs to train and monitor their networks6. What to require from creators is in our FTC influencer disclosure guide.
A worked example (hypothetical)
Three partners each run a spring brief in the US. The inputs below are invented to show the arithmetic; they are not client results. Costs are in US dollars.
| Partner | All-in cost (USD) | New customers (brand data, same window) | Cost per new customer (USD) | Posts reused in ads | Posts live on time |
|---|---|---|---|---|---|
| Agency A (TikTok-led) | 18,000 | 150 | 120 | 9 of 20 | 19 of 20 |
| Agency B (YouTube-led) | 15,000 | 100 | 150 | 2 of 6 | 6 of 6 |
| In-house team | 6,000 | 60 | 100 | 3 of 10 | 7 of 10 |
Agency B would lose on cost per view (YouTube views cost more), and Agency A’s own report may show a higher ROAS because of platform attribution. On the brand’s numbers, the in-house team is cheapest per customer but late on three posts, and Agency A supplied the most reusable content. That is a reason to test, not yet to move the whole budget.
How do you move budget between influencer agencies?
Move budget in steps, on a quarter of comparable work, using cost per new customer as the main signal and reuse and delivery as tie-breakers. Never move it on one campaign: one creator going viral can swing a single result.
- Freeze the standard for the quarter. No changes to definitions or the window mid-quarter.
- Compare like with like. Group results by market and platform before ranking partners.
- Rank on cost per new customer. Use brand data only.
- Check the tie-breakers. Content reuse, delivery and reporting completeness.
- Run a challenger brief. Give the leading partner a brief in a market another partner holds, with the same standard.
- Shift one step. Move a portion of next quarter’s budget, not all of it, and record the reason in the tracker.
- Tell every partner the rule. Agencies that know they are compared on the same scorecard report more completely.
| What you see after a quarter | What to do |
|---|---|
| Lower cost per new customer and complete reporting | Give the next increment of budget or a new market |
| Cheaper per view but no customer data | Ask for codes and UTM links on every post before adding budget |
| Strong content reuse, average customer cost | Keep for content supply; move conversion budget elsewhere |
| Repeated late posts or disclosure misses | Hold budget flat until fixed |
| Cost lines missing or bundled | Make transparent cost lines a condition of renewal |
How to weigh an agency against software or a hybrid setup in the first place is covered in influencer marketing agency vs platform.
Which contract clauses keep data ownership and reporting with the brand?
Put the measurement standard into the contract and make the brand the owner of the data, accounts and licences, so changing agencies never resets your history. Media buying contracts already work this way: the ANA’s media agency contract template keeps audit rights, defines performance and compliance audits, and adds a transaction data schedule listing the data to obtain11. The ANA’s influencer waste report likewise points to contracts as the place deliverables, ownership, usage rights and compensation are decided3.
| Clause | What it should say |
|---|---|
| Data ownership | All campaign data, reports, creator performance data and post links produced for the brand belong to the brand |
| Account ownership | Paid amplification runs from brand-owned ad accounts; the agency works through partner access |
| Reporting standard | The brand’s measurement standard is attached as a schedule; the agency fills the brand’s tracker by an agreed day each month |
| Cost transparency | Creator fees, agency fee or margin, usage rights and other pass-through costs are invoiced as separate lines; creator contracts available on request |
| Audit right | The brand may check reported fees and results against creator contracts and platform data |
| Content licences | Usage rights are licensed to the brand (not the agency) and stay valid if the agency changes |
| Disclosure | The agency briefs and monitors creators for disclosure and corrects failures within an agreed time |
| Handover | On termination, the agency hands over creator contacts and contracts for the brand’s campaigns, raw files and the full tracker within an agreed number of days |
The clauses that belong in each creator agreement are in our influencer contract template.
Where does the Kolect growth system fit?
The Kolect growth system is built to be run by Kolect’s team, the brand’s own team or the agencies it already uses13. Existing agencies keep working and count results the same way as everyone else, on one shared sheet, so they can be compared like for like (how this works); access to a brand’s data is read-only and revocable13. You can start in one market without replacing current partners.
If you only need to check that a creator an agency proposes has a real audience, the free fake follower checker does that without an account.