Results

How we report results: the published method

The rules every Probative result will follow: definitions, data checks, small-number limits, test status, written client approval and corrections.

Updated October 24, 2026

Probative has no client results to publish yet. When we do, every one will follow the rules on this page. They are the public-results rules of The Probative Standard, written down before the first result exists, so no result can be shaped to fit them afterwards.

In one review of 50 public law-firm marketing case studies, only 6% disclosed a cost per signed case (review). Every result we publish will.

The rules for every published result

  1. Signed cases first. Each result leads with signed cases and cost per signed case, counted from the client’s own records, not with clicks, rankings or leads.
  2. Defined terms. A lead, a signed case and cost per signed case mean what the Standard says, and the version of the Standard is named.
  3. Clean data only. The whole period must pass every data-health check: click and call IDs captured, sources known, our count matched to the client’s CRM. A period that fails isn’t published.
  4. Equal windows. Before-and-after comparisons use periods of the same length, on the same basis, with the prior year shown where it exists.
  5. Small numbers stay small. No cost per signed case for any line with fewer than 5 signed cases.
  6. Credit or cause, labeled. Every result says whether a fair test stands behind it, or says “attributed, not tested.” Test results carry a 90% range (method).
  7. Rounded down, never up. Nothing is adjusted.
  8. Written approval. The client approves the final text in writing. No client is named without written consent.
  9. Connections disclosed. If a client received anything of value, such as a founding-seat discount, the result says so (16 CFR 255.5).
  10. Corrections in public. An error is corrected on the same page, dated, with what changed.

Reviews and testimonials

We ask every client for a review the same way, at set moments, with no incentives and no selective asking. Founding-seat clients are asked only after their discount ends, and a review is never part of any deal. We publish no testimonial we didn’t earn, following the FTC’s rule on reviews and testimonials (FTC). Every claim we make, about clients or ourselves, needs a reasonable basis before we make it (FTC).

What a published result will look like

  • The practice, the state and the channels, without naming the client unless they consent.
  • The period, the Standard version, and the basis (signing date or lead cohort).
  • Signed cases and cost per signed case, before and after, in equal windows.
  • The test status of each channel.
  • What changed, and what the firm’s own teams contributed.
  • The client’s approval date.

Until then, see the founder’s record and the ILLUSTRATIVE sample report.

Questions firms ask

Will you publish results that look bad?
We will publish only with the client's written approval, and we never edit a number to look better. A result that fails our data checks isn't published at all, good or bad.
Can a client withdraw approval?
Yes. If a client asks us to take a published result down, we do.
What if you find an error in a published result?
We correct it on the same page, with the date and what changed. We never silently edit a published number.

Start with your own numbers.

Free. A person records a 90-second video of what clients see when they search for you, by email within 2 business days. No account access, no call. We take a few each week; if we're full we'll tell you the date. Paid work opens in January 2027.

  1. Free video review

    A 90-second video on your public ads and search presence, made by a person, within 2 business days.

  2. The measurement study, $3,000

    Your numbers, from click to signed case, in 14 business days. 7-day refund; credited if you sign within 30 days.

  3. Choose your plan

    A package or single services, every one with a published price. You own every account.