Runna's advertising, taken apart.

What we can see, and what we cannot

Google's Ads Transparency Center and Meta's Ad Library publish what every advertiser runs: the creative, the destination, the advertiser identity, and the date each ad went live. That is enough to reconstruct a strategy.

It is not enough to judge results. Spend, impressions, conversion rates and targeting stay private. So everything below describes decisions, not outcomes – and where we speculate about why a decision was made, we say so.

Search: segmented by distance, not by product

The most instructive thing in Runna's search advertising is the destination, not the copy. Ads for marathon training point at a marathon training page. Ads for half marathon plans point at a half marathon page. The product is identical; the landing page is not.

Headline	Destination	What the split implies
The Ultimate Marathon Plans – Marathon Training Plans	/training	Marathon intent gets a marathon page
Half Marathon Plans – The Best 1/2 Marathon Plans	/half-marathon-training	Distance-specific landing, not the homepage
The Best 1/2 Marathon Plans – Fall In Love With Running	/half-marathon-training	Second angle on the same intent

Source: Google Ads Transparency Center, US region, advertiser verified as The Run Buddy LTD. Checked 11 September 2026.

This matters more than it looks. Someone searching "marathon training plan" and someone searching "half marathon plan" are at different points in a training calendar, with different time horizons and different anxieties. Sending both to a generic app page throws that away, and the cost shows up in conversion rate rather than in anything the ad platform reports as a problem.

Two different headlines against the same half-marathon destination is also worth noting – one leads with superlative ("The Best"), one with emotion ("Fall In Love With Running"). That is a live message test, run where message tests are cheapest.

The trial terms are in the ad, on purpose

Across Meta creative, two lines appear near the top almost every time: "First week free" and "No credit card required."

Putting trial mechanics in the ad rather than saving them for the paywall is a deliberate trade. It lifts click-through and trial starts, because the perceived risk drops to zero before anyone taps. It also lowers trial-to-paid, because a trial started without payment details converts at a fraction of the rate of one with them.

An app that leads with "no credit card required" has decided that volume at the top of the funnel is worth more than conversion rate in the middle – which is a defensible choice for a product whose value takes a week of actual running to demonstrate. You cannot prove a training plan works in three days.

Meta copy written as a FAQ, not as a story

The longest-running Meta creative is structured as a list of questions answered in one line each:

What is Runna? A personalised running coach in an app. Who's it for? Beginners through to marathon runners. What do you get? A plan built around your current fitness and goal.

This is unusual in consumer social advertising, where the default is narrative or testimonial. It is also well suited to a category where the buyer is comparing several similar apps and wants the differences stated rather than dramatised.

Worth flagging for anyone thinking about AI answer engines: copy written as explicit question-and-answer pairs is exactly the shape that gets extracted and quoted. Whether that motivated the choice here we have no idea, but the format pays twice.

Three formats, and the parent brand doing work

Format	How it appears	What it is for
Brand	Posted as "Runna with Strava"	Borrows the parent brand's credibility in one line
Creator	Posted as "Runna with [creator name]"	A real runner's account, not the company page
Direct app install	Standard app-install unit to the App Store	Volume, lowest friction

Source: Meta Ad Library, US, active ads as of 11 September 2026.

The creator format is the one most agencies underuse. An ad served from a person's account outperforms the same message from a company page because people respond to people – and in a category where the product is a coach, a named human is nearly the product itself.

The parent-brand framing is cheaper still. "Runna with Strava" costs nothing to produce and transfers trust from a brand nearly every target already uses.

Creative volume, and what the library tells you about cadence

Several Runna creatives carry Meta's "3 ads use this creative and text" or "6 ads use this creative and text" marker, with start dates clustered in mid and late August 2026. So: the same message, deployed across several ad units at once, refreshed as a batch.

That is a production pipeline running on a schedule rather than a set of ads left up since launch – which is the single clearest signal in any ad library that someone is actively managing the account.

The contrast worth seeing

In the same search results sits a competing running app whose entire Meta presence is one block of copy – the same paragraph, word for word, across roughly ten ad units with start dates from late April through mid August 2026. Not a single line rewritten in nearly four months.

There are innocent explanations. The copy may have won a decisive test. The team may be deliberately holding message constant while testing video. But the pattern matches the most common failure in app advertising: creative is treated as a launch asset rather than a consumable, performance decays quietly, and nothing in the platform reports that assets are the cause.

We are not naming them, because the point is the pattern rather than the company, and because we would be guessing about their results.