How Often Do Competitors Change Their Ads?

You open a rival's Meta Ad Library page and there are forty active ads. Half of them launched in the last week. The other half are scattered across the past four months. Your first instinct is to screenshot the ones that look good and move on. That instinct skips the most useful thing on the page: the launch dates themselves.

Short answer: How often a competitor changes their ads is a read on their testing intensity and budget confidence. A wall of ads launched in the last week means aggressive testing, they are still hunting for a hook. Three ads running past 90 days means they found a control and are protecting it.

The takeaways

• A wall of week-old ads reads as aggressive testing. Rapid churn, each ad live 6 to 10 days before it disappears, usually says fatigue is hitting fast or the hook has not landed yet.
• A few ads past 30 days is the durable-winner signal. An ad still running after 30+ days has survived optimization cycles and at least one refresh window, so it is a proven control worth studying.
• The library shows launch dates but hides reliable stop dates. You only catch churn by revisiting a page over weeks and noting what vanished, so cadence shows up only when you track a page over time.

What does a competitor's refresh cadence actually tell you?

Refresh cadence is how fast a competitor swaps creative, and it maps to two things: how hard they are testing, and how settled their strategy is. Fast churn means they are still searching for something that works. A small, stable set of long-runners means they already found it and are defending the budget behind it.

The trap is reading speed as strength. A brand cycling through a dozen new ads a week is not automatically winning. Just as often it is a sign of a saturated audience or a hook that keeps dying, so they keep throwing fresh creative at the wall. A confident account usually looks calmer: a proven core running for weeks, with a steady trickle of new tests around the edges.

How do you read cadence in the ad library?

Sort by launch date and look at the whole spread rather than one ad. In the Meta Ad Library each active ad shows when it started. A cluster of same-week launches on one page is a testing burst. A scatter of start dates across several months is a stable rotation with a durable core. Do this across Meta, TikTok, and Google's transparency library so you are reading a brand's rhythm rather than one platform's quirk.

Count variants per concept while you are there. Ten near-identical copies of one ad is one high-budget push behind a single concept, and it tells you where their money is concentrated right now. The pattern of dates plus duplicates is a cleaner picture of intent than any one creative in isolation.

What the ad library won't show you about cadence

The library reliably shows an ad's start date and hides its stop date. An ad can drop out without a marked end date, so you cannot compute a clean "average creative lifespan" from a single visit. The honest workaround is revisiting the same page every week or two and logging what is new and what has disappeared. Cadence is a time series you build up over repeat visits.

Remember what else is missing: impression ranges are wide, and there is no spend and no ROAS. So cadence tells you rhythm, not results. You are inferring how a competitor works and where their confidence sits, never confirming whether any given ad actually made money. Treat the read as a hypothesis about their strategy, then pressure-test it against everything else you can see.

How fast is too fast, and how slow is a warning?

There is no universal number, but the shape reads cleanly. Ads cycling every 6 to 10 days usually mean fatigue is biting hard: a saturated audience, or a hook that is not sticking. A brand refreshing every few weeks around a stable core looks confident and well-organized. Three ads untouched for 90+ days can mean a durable control worth copying the structure of, or a sleepy account that stopped testing altogether.

Read the cadence relative to their spend signals and to your own tempo. A weekly refresh in fast-moving DTC apparel is normal housekeeping; the same rhythm from a B2B advertiser might mean they are flailing. The number only means something next to context.

Turning a snapshot into a rhythm

The moment you are watching more than two competitors, tracking this by hand falls apart, because a snapshot cannot show churn and your memory of last week's page is unreliable. That is the whole point of pulling the ad libraries into one place: Adscalr pulls the Meta, TikTok, and Google transparency libraries daily and sorts them longest-running first, so both the durable winners and the fast churn are readable at a glance instead of reconstructed from screenshots.

If you want the flip side of this, the 30-day longevity signal for spotting a single winning ad covers how to read durability on one creative, and the broader ad-library research workflow covers finding and mapping competitors in the first place. Cadence is the layer on top: once you know who is running what, the rhythm of how they swap it tells you how they think.