The best free signal of what's working: how long an ad has been running

Every advertiser you'd want to learn from guards their numbers. You'll never see their cost per acquisition, their return on ad spend, or which creative is carrying the account. That data is locked inside their ad platform.

But there's one thing they can't hide, because the law makes it public: how long each of their ads has been running. And run-length turns out to be one of the most honest free signals you can get.

Why does how long an ad runs tell you anything?

Because paid media is ruthless, and the people running it are not sentimental. A performance marketer pointing real budget at an ad is watching it daily. If the cost per result climbs or the return drops below target, the ad gets paused — often within days, sometimes within hours. Money does not sit on a loser. It gets reallocated to whatever is working.

The flip side is what makes run-length useful: when an ad keeps spending week after week, month after month, someone with a dashboard you can't see has decided, repeatedly, that it's worth funding. Every additional day of life is a fresh vote of confidence backed by money.

That's the whole logic. You're not measuring the ad directly. You're observing the behavior of a rational spender and inferring the result. Survivors survive because they earn their slot.

Where can you actually see this?

The major platforms run public, searchable ad libraries — most of them born out of political-ad transparency rules, then extended to all advertisers. Each one stamps ads with the dates they were active, which is exactly the data you want:

• Meta Ad Library — the richest of the four. Covers Facebook, Instagram, Messenger and Threads. Search any advertiser, see their active ads, and read the date each one started running. This is usually where you start.
• Google Ads Transparency Center — search by advertiser or domain to see their Search, Display and YouTube creative, each with a first-shown and last-shown date.
• LinkedIn Ad Library — every company's paid posts with their date ranges. The best window into B2B.
• TikTok — no true ad library for organic-style creative the way Meta has, but TikTok's own ad and commercial-content tools, plus a brand's organic posting cadence, tell you which concepts a brand keeps returning to.

When you open one of these, here's what to actually look at:

• Start date and run length. An ad live for many months has cleared a bar that an ad live for three days hasn't.
• Persistence and re-appearance. Look for the same concept relaunched again and again, or many small variations of one core idea. A brand that keeps coming back to the same hook, the same format, the same offer is telling you that angle works for them. Recurrence is as strong a signal as raw duration.
• Concentration. If a brand is running forty ads but the same three structures dominate the long-lived ones, those three are your shortlist.

What does run-length get wrong?

This is a proxy, not a verdict, and treating it as certainty is how you learn the wrong lessons. Be honest about where it breaks:

• Brand and awareness campaigns run long for other reasons. A big advertiser may keep a brand spot live for months because it's measured on reach or recall, not direct response. Longevity there reflects a budget decision, not a conversion result. Direct-response ads — clear offer, clear call to action — are where run-length tracks performance most tightly.
• Big budgets distort the picture. A company with deep pockets can afford to keep a mediocre ad alive far longer than a lean challenger would tolerate. Run-length is most trustworthy when you read it relative to how that specific advertiser behaves, not in absolute days.
• You can't see the spend behind it. An ad running a long time on a tiny budget and an ad running a long time on a huge budget look identical in the library. Duration tells you something survived; it doesn't tell you how much it earned.
• It's correlation, not proof. Ads occasionally stay live out of neglect, contractual commitments, or testing protocols. The signal is strong in aggregate and across many advertisers — it's weaker on any single ad. Read patterns, not one data point.

How do you turn the signal into better creative?

The mistake is to find a long-runner and copy it. You'd be cloning the surface of something whose substance you don't understand — and you'd be late, since they've already been running it for months.

The better move is to study the structure of survivors across many advertisers and find what they share. When you line up the long-lived ads in a category, patterns surface fast:

• The hook. What's happening in the first two seconds or the first line? Survivors tend to open with a problem, a number, or a pattern-break — rarely with a logo.
• The format. Founder talking to camera, fast-cut demo, side-by-side comparison, text-over-broll. Note which formats keep surviving in your space, not just which ones look nice.
• The offer and proof. What concrete claim or guarantee shows up in the ads that last? Specific beats vague, and the long-runners usually prove it.
• The call to action. Where it lands, how hard it pushes, what it promises on the click.

Extract the skeleton, then build your own ad on it with your product, your proof, your voice. You're borrowing the validated structure, not the execution. That's the difference between learning from competitors and imitating them.

Can you automate this?

You can do all of it by hand — open each library, search each competitor, eyeball the dates, build a spreadsheet of survivors. It works, and it's worth doing once to build your instinct for what a winner looks like in your category.

It's also exactly the kind of repetitive scan that's better run for you. Hermoso's Ad Radar pulls a brand's competitors from across the public libraries and ranks their ads by how long and how persistently each has run — so the survivors rise to the top automatically, and you spend your time studying structure instead of collecting dates.

However you do it, the principle holds. You can't see what's working inside someone else's account. But you can watch what they refuse to turn off — and that, read carefully, is most of what you needed to know.

Frequently asked

Does a long-running ad guarantee it's profitable?

No. Run-length is a strong proxy, not proof. Direct-response advertisers kill underperformers quickly, so a long, persistent run usually means an ad is converting. But brand and awareness campaigns run long for reasons unrelated to direct conversions, deep-pocketed advertisers can keep mediocre ads alive longer than lean ones would tolerate, and you can't see the budget behind any ad. Trust the pattern across many advertisers more than any single ad.

Where can I see how long a competitor's ads have been running?

The public ad libraries. Meta Ad Library (Facebook, Instagram, Messenger, Threads) is the richest and shows each ad's start date. Google's Ads Transparency Center covers Search, Display and YouTube with first-shown and last-shown dates. LinkedIn's Ad Library is best for B2B. TikTok lacks a full library but a brand's commercial content and posting cadence reveal which concepts they keep returning to. All are free and searchable by advertiser.

Should I just copy a competitor's longest-running ad?

No — copy the structure, not the execution. By the time you see a long-runner, the competitor has had months of head start, and cloning the surface means imitating something whose underlying logic you don't understand. Instead, line up the survivors across several advertisers and extract what they share: the hook, the format, the offer and proof, the call to action. Then rebuild that validated skeleton with your own product and voice.

Why do advertisers kill losing ads so fast?

Because paid media is a daily optimization game with real money on the line. Performance marketers watch cost per result and return on ad spend constantly, and when an ad slips below target they pause it and move the budget to whatever is working. That ruthlessness is what makes run-length informative: every extra day an ad stays live is a fresh, money-backed decision that it's still worth funding.