FINDING more ads ≠ better hit rateLEAN under 20 live ads → ~half prove outHEAVY 20+ live ads → ~a third prove outDTC runs more ads, wins a smaller shareSNIPER some run <20 ads, 100% provenDATA built from the public Meta + Google ad librariesFINDING more ads ≠ better hit rateLEAN under 20 live ads → ~half prove outHEAVY 20+ live ads → ~a third prove outDTC runs more ads, wins a smaller shareSNIPER some run <20 ads, 100% provenDATA built from the public Meta + Google ad libraries

// creative velocity

Running more ads doesn't buy a better hit rate.

We counted every ad running right now behind ~140 growing companies, 3,800+ live ads in total, and asked a simple question: do the companies that flood the ad libraries actually land a higher share of winners? They don't. Companies running 20+ live ads prove out about 41% of them; lean advertisers prove out about 53%. Volume buys more winners in raw count, not a better strike rate. Here is the whole distribution, honest about where the ad libraries cap us.

25
median live ads per company
how big a roster the typical company keeps running
43%
of live ads are proven
still running 60 days or more
53%
hit rate, lean advertisers
companies running 20 or fewer live ads
41%
hit rate, heavy advertisers
companies running 21+ live ads
Section A · volume vs hit rate

Does testing more ads mean a higher share prove out?

Each bar is a group of companies by how many ads they keep live, and its height is the pooled hit rate: the share of that group's live ads that have survived 60+ days. Run time is the market's verdict, a company only keeps paying for a creative that converts. The bars step down, not up. More volume comes with a lower hit rate, not a higher one.

0%15%30%45%60%52%1–521 cos49%6–1520 cos48%16–3038 cos40%31+62 coslive ads a company runs →

The heavy end holds even when you drop the companies pinned at the library caps: the pattern is the more ads you run, the smaller the share that prove out. That is what real testing looks like, most of what you launch is a bet, and the winners come from placing more bets, not from a magically higher strike rate. Absolute winner count still rises with volume: a company running 45 ads at 41% lands far more proven creatives than a lean one, it just wastes more spend finding them.

Section B · SaaS vs DTC

The same pattern splits cleanly by business model

DTC brands live and die on creative, so they run bigger rosters and churn them hard. Founder-led SaaS runs leaner and lets proven ads ride. The split is exactly what the volume-vs-hit-rate finding predicts: more ads, lower share proven.

DTC / e-commerce
36 median live ads
Runs the bigger roster. Only 39% of those ads have proven out, across 51 brands.
Founder-led SaaS
19 median live ads
Runs leaner. A higher 46% of its ads have proven out, across 90 companies.
Section C · the extremes

Snipers, sprayers, and the rare company that does both

Two ways to run a live ad account. Open any company for the full teardown behind the numbers.

Section D · the roster size

How many ads a growing company actually keeps live

Context for the finding above: the size of the live-ad roster across the set. Most growing companies are not running one or two ads, the median keeps 25 live at once, and a large group runs far more.

211–5116–102311–203221–355436+live ads (companies in each band)

Read the top band as a floor, not a ceiling. The public ad libraries are collected up to a per-platform cap (25 Google, 50 Meta), so 32 of the 141 companies are pinned at that ceiling, their real roster is bigger than we can see. That is also why we don't publish a “top X% run Y% of all ads” concentration figure: the tail is censored, so any such number would be made up. It does not distort the hit rates above, which are shares within each company's live ads.

Method: 141 companies with at least one live ad, drawn from the companies we've broken down. Every ad counted is live right now (still marked active) in the public Meta or Google ad library, bound to a company by verified page or domain, never by name. A “proven” ad has run 60+ days, run time is the market's judgment that it converts. Hit rate is a company's proven ads over its live ads, pooled within each group. Ad counts are capped per platform (25 Google, 50 Meta), so the busiest advertisers' totals are a floor. We do not estimate a new-ads-per-month cadence: we can only see ads that are still live, which would undercount everything a company has already killed. Figures round and update automatically.

Run a free growth audit on your own domain →