An ad can be clever, widely shared, and still leave people remembering the joke instead of the brand. That is what interests me about marketing fails: the gap between a good-looking idea and what the audience actually takes away.

This article looks at creative ads and guerrilla marketing through that lens. A surprising execution is a starting point. It is not evidence that a campaign worked.

What counts as a failure here?

I am judging the creative execution and the problems it can create. Without campaign objectives, costs, and results, I cannot claim that an example lost money or failed commercially.

An image can show an awkward message or a weak brand connection. It cannot, by itself, prove the outcome of a campaign. Keep that distinction in mind when looking at any roundup of marketing failures.

A creative ad is not automatically a good guerrilla marketing example

A door handle shaped like a product or a billboard with a visual trick can be an interesting ad. To understand the marketing, you still need to look at the placement, the audience, and the response it was meant to create.

I tend to favour ideas where the product fits naturally into the experience and the distribution does some of the work. That is a preference, not a universal definition. Visible branding does not automatically disqualify a guerrilla campaign, and subtle branding does not automatically make one effective.

Before adding an execution to your inspiration folder, ask five questions.

1. What is the audience supposed to do next?

Share the idea, try the product, visit a location, or remember the name? Pick an intended response before deciding whether the execution supports it.

“People will talk about it” is incomplete. What will they say, to whom, and what does that have to do with your product?

2. Does the brand belong in the idea?

If another company could swap in its logo without changing anything, the idea may be more memorable than the product. Subtle branding can work, but invisible branding is not an achievement on its own.

Show the execution to someone who has not seen your brief. Ask what they remember and which company they think it promotes. Do that before explaining the joke.

3. What makes the distribution unusual?

Look at the placement, participation, timing, and route through which people encounter the campaign. A surprising visual alone tells you little about how the campaign reached anyone.

A stunt photographed once for an awards submission and an experience encountered by your potential customers are different things. Find out which one you are looking at.

4. Is the payoff worth the cost?

Production is only part of the bill. Include placement, staffing, follow-up, and the work needed to handle reactions. Cheap execution and effective execution are separate things; a screenshot rarely gives you enough evidence to judge either.

Decide what you will measure before you start. Visits, trials, enquiries, sales, or a change in brand recall each tell you something different. Counting shares is useful if sharing was part of the plan, but it does not answer every other question.

5. What might people take away instead?

Could the joke obscure the product, suggest something unpleasant, or send people toward a competitor? Could participation feel intrusive to someone who did not agree to it?

Test those interpretations with people who have not sat through your pitch deck. You need their first reaction, not their agreement after a five-minute explanation.

A documented example: The North Face and Wikipedia

In May 2019, the Wikimedia Foundation criticized a campaign by The North Face and its agency, Leo Burnett Tailor Made. According to the Foundation, Wikipedia images had been replaced with promotional photographs. It rejected the suggestion that Wikipedia had collaborated and reported that volunteers removed the images or cropped out the logos.

The Wikimedia Foundation’s account documents the objection and the removal of the placements. It does not establish the campaign’s sales impact.

What went wrong with the idea?

My criticism is that the distribution depended on treating a community’s editorial space as available advertising inventory. That gave the people maintaining the space a reason to undo the placement.

The brand connection was clear. The unusual distribution was clear. Neither solved the problem of whether the people running the platform wanted the campaign there.

What can you take from it?

Ask who controls the space you intend to use and what they get from the idea. If your plan depends on them not noticing what you are doing, include the likely removal and reaction in your assessment.

There is a useful distinction between offering something a publisher may choose to use and altering someone else’s editorial content for promotion. The free stock-photo marketing guide explores the first approach: create a useful asset and let the publisher decide.

Put the criticism to work

Before launching an idea, write down the response you want, the role your brand plays, the cost of delivery, and the reaction you least want to create. Then show the execution to someone who does not already know what it means.

For another angle, read how large corporations use guerrilla marketing and compare the distribution tactics with the creative itself. The 2018 marketing stunt roundup gives you a historical set of examples to examine with the same questions.

A good idea should survive those questions. If it only works while nobody asks them, keep working on it.

Editorial update, October 2026: the earlier image roundup has been replaced with a sourced example and a practical assessment framework.