Working · 2026-11PrimaryAcquisitionPaid

Test brand-keyword bidding with a holdout

Pause brand-term ads in a controlled test to see whether the sales move to organic results before paying for brand defence.

Also called: brand defence, brand protection bidding

Claim and evidence

The claimbrand-keyword ads have no measurable short-term benefits.

eBay · nber.org · 2014-05

What the numbers show

No published number for this tactic has passed our word-for-word check yet, so there is no chart.

Why it works

Diagram, not data
Brand search campaignPaused in testregionsKept running incontrol regionsCompare paid plusorganic conversions
Brand search campaignPaused in testregionsKept running incontrol regionsCompare paid plus organicconversions

People searching for your name often click the organic result anyway, so brand ads may only capture demand you would get free. A controlled pause measures true incremental value.

How to run it

  1. Check whether competitors or affiliates bid on your brand
  2. Run a geo or time-based holdout on brand campaigns
  3. Compare total (paid + organic) clicks and conversions
  4. Keep ads where competitors appear above you
  5. Redo the test when competitors change

Ways to run it

Design options to test, not promised results.

  1. Option AGeo holdout

    Switch brand ads off in matched regions and compare total clicks and conversions with regions that keep them.

  2. Option BTime-based pause

    Alternate periods with and without brand ads and compare paid plus organic results across them.

  3. Option CDefend only when attacked

    Keep brand ads only where competitors or affiliates bid above you, and redo the test when that changes.

Timeline

  1. 12014 first documented: Blake, Nosko and Tadelis (eBay field experiments)
  2. 22014 eBay — Field experiments turning off search ads for brand and non-brand keywords.
  3. 32026-11 status checked: Working

Dated examples

  1. 2014

    eBay

    Field experiments turning off search ads for brand and non-brand keywords.

    Brand-keyword ads had no measurable short-term benefit; non-brand returns averaged negative as frequent buyers dominated spend.

    “brand-keyword ads have no measurable short-term benefits.”
    ✓ quote checked word-for-word on the page

    nber.org · 2014-05

Does it still work?

Workingas of 2026-11

Academic result for eBay still stands as a test method. Nov 2026 B2B paid trend piece lists a shift to branded search among software companies as AI Overviews cut Google inventory, so test, don't assume. source

Risks

  • Generalising from eBay's brand strength
  • Competitors can bid on your name once you stop
  • Short-term test window

Sources

  1. nber.org · NBER working paper 20171 · Tom Blake, Chris Nosko, Steven Tadelis · 2014-05
    “brand-keyword ads have no measurable short-term benefits.”
    ✓ quote checked word-for-word on the page
  2. growthunhinged.com · Growth Unhinged · Kyle Poyar · 2026-11-01
    “AI Overviews are squeezing out inventory on Google, leading to more competition for fewer clicks.”
    ✓ quote checked word-for-word on the page

Change log

  • 2026-11-20 — fact-check: paraphrase replaced with page text (https://www.growthunhinged.com/p/state-of-b2b-paid-ads)
  • 2026-11-20 — harvest v0.1 (acq-paid)