Marketing Oops published on 30 July 2026 a write-up of a Wisesight Research session reporting that Thai social audiences pulled back from promotional and educational posts in the first half of 2026. Comparing 2025 with the first half of 2026, Wisesight put entertainment content up 7%, product-promotion content down 18%, community content down 21% and education content down almost 28%.
The session, "Social Media Trends: The Springboard to Your Next Move", was delivered by Thammaporn Phuengtua, Lead Research Intelligence Strategist at Wisesight Research, at the Wisesight Cheat Day no.2/26 event. Wisesight is a social-listening and data vendor and this was its own stage, so every figure below is Wisesight's own research rather than independent measurement. The conclusion the numbers lead to, that brands need finer engagement measurement, also describes what Wisesight sells. They are still Thai data about Thai brands, which is rare enough to be worth the reading.
Education content fell almost 28%, but recipes and DIY grew 32%
Wisesight's pillar comparison, setting 2025 against the first half of 2026, shows the decline is not uniform inside the education category. Recipe and DIY content grew up to 32% and fact or how-to content grew 12%, while warning-style posts of the "เตือนภัย / รู้ยัง" kind fell 45% and comparison reviews fell 58%. The drop is therefore a drop in a particular register: alarm and adjudication lost ground, while content someone can go and do held up.
Engagement per post fell 67% while post volume fell only 6%
The most usable finding in the Wisesight session, reported by Marketing Oops on 30 July 2026, concerns the Voxpop street-interview format. In early to mid 2025, engagement on Voxpop content fell 67% while the number of Voxpop posts fell only 6%. That is a different measurement window from the pillar comparison above, which sets 2025 against the first half of 2026, and the two should not be read as one trend line.
The gap between those two numbers is what to keep. Creators and brands went on publishing the format at close to full volume long after the audience stopped rewarding it. A dashboard reporting total engagement would have shown a gentle slope, because volume propped the total up. Engagement divided by post count would have shown a collapse. Wisesight did not present this as a measurement recommendation, but the arithmetic supports one: engagement per post, tracked per format, is a leading indicator of format decay, and total-engagement reporting hides it. If your measurement setup only rolls up totals by channel, a format can die inside it with nothing on the chart looking wrong.
What earned engagement in the first half of 2026
Wisesight reported that posts in which a brand used an influencer to promote a product generated 10 times more engagement than posts in which the brand promoted the product itself. For celebrity campaigns, launch-day engagement rose more than 7 times, and brands that kept publishing recap or fan-sourced content for one to three days after the event grew engagement a further 88%. In both cases the brand is not the one doing the talking.
Giveaways moved the same way. Wisesight found that giveaway posts offering experience-based prizes averaged 6,300 engagements per post, ahead of both discount vouchers and high-value prizes such as gold, and that giveaway engagement grew 2.2 times year on year. The cheaper prize outperformed the expensive one.
The Thai campaign results Wisesight cited
Wisesight named four Thai campaigns in the session reported by Marketing Oops on 30 July 2026, with these results:
| Brand or campaign | Result reported by Wisesight |
|---|---|
| Sushiro | Brand mentions up 62%, with mentions holding a higher baseline after the campaign ended |
| Cafe Amazon | Brand mentions up 73%, also holding a higher baseline afterwards |
| Lenskart, "วิ่งแลกแว่น" | Engagement grew 2.6 times |
| Potato Corner, "เติมไฟ" | Engagement grew 61%, to 4.5 million engagements |
Wisesight framed the whole set through a formula it presented as Social Performance = (Relevance x Participation x Utility) x Trust.
What this means for Thai marketers
The first job is to check whether your own numbers say the same thing, because Wisesight has not published the basis of these figures. The source does not state sample size, platform coverage, category mix, or whether the percentages are medians or averages, which makes the direction more usable than the magnitude. A 62% lift in mentions for Sushiro is not a benchmark you can hold your own campaign against without knowing what was counted.
The direction is consistent, and it is Thai. Promotional posts published from the brand account lost engagement, while the same product sold through an influencer earned it back many times over. If your paid social and organic calendars are still weighted towards brand-voice product posts, the data argues for moving the selling into other people's mouths and keeping the brand account for entertainment. For content planning, the education finding is the more precise instruction: keep the how-to and recipe formats, and look hard at the alarm posts and comparison reviews that fell 45% and 58%.
Frequently asked questions
How large was the sample behind these figures?
The source does not say. The write-up of the Wisesight session does not state sample size, which platforms were measured, which product categories were included, or whether the percentages are medians or averages. Treat the figures as directional.
Is this independent research?
No. The figures come from Wisesight Research, a social-listening and data company, and were presented at Wisesight's own Cheat Day no.2/26 event by Thammaporn Phuengtua, its Lead Research Intelligence Strategist. Wisesight sells the kind of measurement its findings argue for, which is worth holding in mind when reading them.
Does the 67% engagement drop cover the same period as the content-pillar changes?
No, the two cover different periods. The 67% fall in Voxpop engagement, against a 6% fall in Voxpop post volume, was measured in early to mid 2025. The pillar changes, entertainment up 7% and education down almost 28%, compare 2025 with the first half of 2026.
Should Thai brands stop posting educational content?
No, and Wisesight's own breakdown argues against it. Education as a pillar fell almost 28%, but recipe and DIY content inside it grew up to 32% and fact or how-to content grew 12%. What fell hardest were warning-style posts, down 45%, and comparison reviews, down 58%.
Formats decay before volume does, and volume is usually the thing being watched. If you want to know which of your own formats are quietly dying, start by tracking engagement per post by format, month over month. Talk to us if you want help putting that in place.
