Case studies Philippine ornamental fish and aquarium feed brand, 3 min read
Teaching Videos Reached 8.6x More People Than Sale Posters, So the Sale Ads Were Rebuilt Around Them
Role. Social media specialist. I handle organic content across Facebook, Instagram and TikTok, comments and DMs, ad planning and builds, and reporting to the owner.
- 8.6x
- reach of teaching videos vs product promos
- 1,384
- posts and videos audited
- 2.17%
- click-through on the first sale campaign
Context
The client is a Philippine breeder of competition guppies that also sells its own aquarium feed through Shopee, Lazada and TikTok Shop. Marketplace sale days (9.9, 10.10, 11.11) are its biggest selling moments, and the habit was to promote them with "Shop now" posters and short promo clips.
I had six days before the 9.9 sale when the owner asked for a paid plan. There was no time to shoot anything new, so the question became which existing content deserved the money.
The audit
I pulled every post from the past three months and sorted them by reach: 568 Facebook posts, 429 Instagram posts, and 387 videos I transcribed so I could compare what was actually said in them.
The format we reached for on sale days was dead last. Long teaching videos, where the owner explains a feeding routine or how to grow fry faster, reached 8.6 times more people than the promos and held attention 3.6 times longer. Videos where someone talks to camera also beat silent ones, with a median of 12,392 views against 6,429.
The accidental test
The page had already run the comparison without meaning to.
Same sale, same products, nine times the reach. Part of that gap was a drop in organic reach in August, and I said so in the proposal rather than claiming all of it. But that cut the same way: organic alone could no longer carry a sale.
What I proposed
A plan I called Teach First: the ad spends most of its time solving the customer's problem and only then names the product. I mapped the beats off the page's best-performing video, from the hook in the customer's own words to the point where the brand first appears.
Two other decisions mattered as much as the creative:
- Advertise where the sale closes. Meta can't see a Shopee or Lazada checkout, so marketplace products got demand-building ads and TikTok Shop ads, while products that close in Messenger got Meta conversion budget.
- Record the spend. The page's only past boost had tripled reach, but nobody had written down what it cost, so there was no benchmark. Every campaign since has a cost line.
Results
For the 9.9 sale, re-cut teaching content drove clicks to the store at a 2.17% click-through rate, for about ₱2.26 a click. On TikTok LIVE I switched the campaign goal mid-stream from product clicks, which wasn't getting delivery, to viewer retention, which kept the stream fed with viewers.
What didn't work
Meta tracked a single purchase from 9.9, partly because the checkout it was sending people to was invisible to it, and it rated the ad in the bottom 20% for conversion. Only 15% of the budget landed on the sale day itself, which taught me to front-load sale-day spend instead of spreading it evenly. Those lessons went straight into the 10.10 and 11.11 plans.
Along the way
- Found 42 captions that offered fish "for sale", the wording behind the page's past content flags, and rewrote them so fish posts stay sale-free
- Built a comment and DM log that now holds over 600 conversations, with approved reply templates
- Compiled 765 photos from the national pet expo into a usable library for the season
Takeaway
- Your own page's data beats best practice. The sale-day format was the weakest one we had.
- Teach first, sell second. Usefulness is what earns the length.
- Put the conversion budget where the platform can see the sale.