flowsk.com
ITP impact

ITP's impact on agencies and their client reporting

Agencies get hit twice: once on their own lead generation, and once every month when a client asks why the platform report and the client's revenue disagree. The second one is the business risk — it is the conversation where retainers are lost.

Quick answer

Uncreditable share ≈ Safari share × P(converts after day 7)

For marketing agencies: 40% Safari × 38% converting after day 7 ≈ 15% of conversions your current window structurally cannot credit.

At an average order value of $3000, that is 15% of revenue attributed to the wrong channel — or to no channel at all.

Safari / iOS share

40%

Convert after day 7

38%

Uncreditable conversions

≈15%

Typical order value

$3000

The buying pattern that breaks

Multiple clients, multiple stacks, multiple platforms, all reporting different numbers for the same period, with no shared ledger to arbitrate.

How it shows up in your reports

  • The monthly report needs a paragraph of caveats explaining discrepancies.
  • Clients compare your platform report to their bank and conclude you are inflating results.
  • Cross-client benchmarks are impossible because every client is instrumented differently.

What to do about it

  1. 01Instrument every client the same way, so numbers are comparable across the book.
  2. 02Report from a first-party ledger and use platform numbers only for in-platform optimisation.
  3. 03Lead the QBR with the journey view — showing the receipt for a specific conversion ends the argument faster than any model explanation.
  4. 04Make the durable id part of onboarding, not a fire drill after the first bad report.

Frequently asked questions

How do I explain the platform-vs-revenue gap to a client?

Do not explain the model — show the receipt. Open one conversion, walk through the ad click, the email, the purchase, and the timestamps. A single concrete journey ends an argument that ten minutes of attribution theory only prolongs.

Should I instrument every client the same way?

Yes. Comparable instrumentation is what turns a book of clients into a benchmark set, and it removes the per-client caveats that make your reporting look defensive.

Who should own the tracking, the agency or the client?

The client owns the data; you operate it. That distinction protects you — the client can never claim you took their measurement with you, and you never have to argue about the numbers being yours.

Does this replace the platform reports in a QBR?

It reframes them. Lead with the first-party ledger for what happened and what it cost, then use platform reports to explain what you did inside each channel.

See the receipt for every conversion.

Flowsk Signals stitches the anonymous click to the email to the purchase — first-party, server-side, de-duplicated on a key you choose. One snippet, $29/mo, and a journey you can inspect event by event.