Stop Online Harm · March – June 2026
The best-performing thing we published had no creative in it at all. Working out why changed how I plan content.
Thiha Htun · August 2026 · 4 min read
Between March and June I ran organic and paid social for four country pages across five languages, on a total ad budget of $522. A fifth market, Bangladesh, was still in launch, so it is not in these numbers.
The best-performing thing we published was not a campaign asset. It was a plain list of every helpline channel we run, in one post: 113,500 people reached, and 164 of them clicked through to a helpline.
I did not expect that. I expected one of the campaign assets — something we had planned, briefed and designed — to be the thing that drove meaningful engagement.
The list had no creative to speak of. No hero image, no strapline, no narrative arc. It was information, formatted to be read fast and screenshotted faster. That turned out to be exactly the point. Our audience is not browsing. They are looking for help quietly, often on a cheap phone and an unreliable connection, often needing to act before someone else notices. A list is scannable, screenshottable, and shareable in a private message. A campaign asset is none of those things.
I want to be precise about what this does and does not prove, because the same quarter gave me a second result that cuts against the neat version of the story.
The Karen-language launch post reached less than half as many people and drove roughly twelve times the clicks. So utility spread further; the launch converted harder.
The lesson is not "stop making campaigns." It is that reach and action are different jobs, and utility content wins the one that matters when someone is already in trouble. A person in crisis does not need to be persuaded. They need a phone number.
First, the obvious thing: I pinned the list to the page, so it stopped being a post that scrolled away and became something permanently findable.
Then I stopped treating it as a social post at all. If a plain list of channels was the most useful thing we had made, the job was to get it in front of people who were never going to see our page. So I took it to earned media — approaching three outlets to republish the helpline information for their own audiences: a major Myanmar media platform, Karenni People’s Voice Media, and Radio to Karen. Ethnic-language outlets reach exactly the communities our smaller pages struggle to, and they carried it because it was useful to their readers, not because we paid for it.
That cost nothing but a decade of relationships in Myanmar media. It is the part I would not have thought to do if I had read the numbers as a performance report rather than as a signal about what people actually wanted.
Utility content will not shift attitudes or build recognition. It reaches people who already know they need help. Everything upstream of that — the reason someone trusts us enough to click at all — still comes from the campaign work. You need both, and the mistake is assuming the polished thing is automatically the useful one.
The audience that came back tells its own story: 57% women, led by 18-to-34s. Which is precisely who the service exists for.
The full Stop Online Harm case study has the rest of the quarter’s numbers.