The scenario
The brand builds gift boxes around Kuwaiti dates, a rotating in-house chocolate line and seasonal sweets, sold direct to consumer through its own Shopify store with KNET, Tabby and cash on delivery at checkout. One founder, a small kitchen and packing team, and a catalogue that has to flex across five separate occasions inside a single calendar year rather than sell one steady product line all year round.
Demand was never the issue. Ramadan, Eid, graduations in June and corporate National Day orders each bring their own spike, and the brand had genuine repeat buyers from the year before who reordered without being asked. What it had never built was a different reason to buy for each spike. Every campaign, whatever the month or the occasion behind it, opened with the same slow-motion lid lift and the same line about premium presentation.
- Monthly revenue band
- pending client sign-off
- Average order value
- pending client sign-off
- Fulfilment
- In-house kitchen and packing, local courier with named delivery slots
- Team
- One founder, two packers, seasonal hires before Ramadan and Eid
One box, one script, five occasions that each needed their own
When we tagged the previous year of spend by occasion rather than by angle, the pattern was hard to miss: the Ramadan campaign, the Eid campaign, the graduation push and the corporate outreach were all running variations of the exact same creative, a slow lid lift ending on a shot of the box from above. The brand had correctly identified that it sells across five distinct moments, and then treated all five as the same customer with the same objection.
They are not the same customer. A woman ordering a ghabga gift for a friend two days out worries about a named delivery window, not presentation. A mother buying Eidiya boxes for a dozen nieces and nephews wants the cards personalised without a dozen phone calls. A graduate's family wants something that survives an afternoon in a hot car. None of those worries live in a beauty shot of the box, and none were ever named in an ad.
The brand's WhatsApp inbox had answered all three questions a hundred times over, in the customer's own words, and not one of those words had made it into a script. The account did not have a targeting problem across five occasions. It had a single occasion's worth of creative, copy-pasted across the other four.
What we did — the creative angles
The losers are here on purpose. A test with only winners was never a test.
The named delivery slot, not the box
Winner"Pick the hour. It shows up in that hour."
- Format:
- Vertical, 14 seconds, courier handing over the box with the chosen time stamped on screen
This was the single most repeated worry in the brand's WhatsApp history during Ramadan: not whether the box looked nice, but whether it would land before the ghabga started or sit on a doorstep through the afternoon heat. Leading with the slot instead of the presentation reframed the ad as a promise about timing, and it beat every unboxing creative on both hook rate and cost per purchase, because it answered a question a same-day delivery app cannot: an exact hour, chosen in advance.
The personalised card, shown going in the box
Winner"Write what you want to say. We put it in the box, not just on the receipt."
- Format:
- Vertical, 16 seconds, a handwritten card slipped into the box before it's sealed
Gifting customers had been typing long personal notes into a generic order-notes field with no way to know if anyone actually read them. Showing the card physically going into the box, in the customer's own handwriting, turned a feature nobody knew existed into a reason to pick this brand over a delivery app that only offers a printed slip. It performed especially well on Eidiya and graduation orders, where the message matters as much as the sweets.
Freshness in Kuwait heat, said plainly
Winner"Yes, it survives the walk from your car to your door."
- Format:
- Vertical, 13 seconds, insulated packaging opened to show the chocolate still set
Gifting chocolate in a Kuwaiti summer carries a specific, unspoken fear that no benefit-led ad had ever named: that it arrives half-melted after sitting in a hot car or on a step. Naming the fear directly and showing the insulated packaging solving it converted better on hot-season graduation orders than any shot of the sweets alone, and it quietly reduced the freshness complaints the customer-service line had been fielding.
The corporate order, quoted before National Day
Winner"Ordering for the office? Get a quote now, delivery locked in weeks ahead."
- Format:
- Vertical, 15 seconds, stacked branded boxes with a company card, text overlay on lead time
Corporate coordinators kept finding the brand through ads meant for individual gifting and asking, in the comments, whether bulk orders were even possible. A creative built specifically for them, naming the lead time up front, pulled a segment of enquiries that had previously been getting lost in a consumer inbox and turned them into a booked date on the calendar months out.
The slow-motion unboxing, set to trending audio
Lost"Watch this box open." (no dialogue, ambient sound only)
- Format:
- Vertical, 20 seconds, ASMR-style lid lift and ribbon pull
It collected views and did almost nothing for purchases. Every gifting account in Kuwait runs some version of this exact shot, so it read as generic rather than distinctive, and it never answered any of the three questions actually stopping a purchase — timing, freshness or personalisation. Pretty packaging sells the box to someone who has already decided; it does not make the decision for her.
A live countdown to Eid, borrowed from the delivery apps
Lost"3 hours left to order for Eid." (ticking on-screen timer)
- Format:
- Vertical, 10 seconds, on-screen countdown timer over the product shot
This copies the exact game the same-day delivery apps already win, and it trained part of the audience to treat this brand as a last-minute option rather than the planned one it actually is. It pulled in a wave of orders too close to the date to guarantee the named slot the brand had built its whole pitch around, which put pressure straight back on the fulfilment team the creative was supposed to be helping.
Price against the mall gift counter
Neutral"Same box. Half the mall markup."
- Format:
- Vertical, 12 seconds, split screen with two price tags
Neutral rather than a clear loser: it drew clicks from price-sensitive shoppers who were never going to be this brand's best customer, without doing the real damage the countdown did. It earned a place later, as a retargeting line for cart abandoners comparing prices.
What we did — the optimizations, in order
Tag the back catalogue by occasion, not just by angle
We exported a full year of ads and tagged each one by occasion — Ramadan, Eid, graduation, corporate — as well as by hook and format, then worked out what share of spend on each occasion sat below its own target cost per purchase.
Why: Tagging by angle alone would have missed the real finding, which is that the same angle was being reused across five occasions that each deserved their own. The occasion-level view is what turned a vague sense that Ramadan felt expensive into a specific, fixable gap.
Build one angle sheet per occasion from the inbox
We read two years of WhatsApp messages, order notes and return reasons and sorted the recurring questions by which occasion they came from, so the ghabga worry, the Eidiya worry and the corporate worry each got their own list of angles.
Why: A single angle sheet would have averaged five customers into one, the mistake the account had already made. Sorting by occasion first surfaced the delivery-slot worry as Ramadan-specific rather than a line buried under graduation and corporate questions.
Shoot in the setting each occasion actually happens in
Ten to fifteen pieces produced in two weeks, but shot across four real settings — a home ghabga table, a graduation gathering, an office reception and a car boot in daylight — instead of one studio set standing in for all of them.
Why: A studio backdrop is what made every occasion look identical. Shooting where the gift actually gets handed over let the corporate quote card and the ghabga delivery slot read as two different products, not one box photographed twice.
One testing campaign per platform, one occasion at a time
One campaign per platform, one ad set per angle, equal budgets, five to seven days, run one occasion's tests at a time so a Ramadan result never got blended into a corporate one, with audience, offer and landing page held identical throughout.
Why: Testing every occasion in the same window is how a brand credits a corporate winner to the Ramadan audience it happened to overlap with. Separating the calendar makes each winner real, not a coincidence of timing.
Read hook, then hold, then click, then cost, per occasion
Each creative was judged in that order inside its own occasion's test, and we wrote down the step it failed at rather than a single verdict, so the countdown timer's failure at the trust step was recorded separately from the ASMR unboxing's failure at the hook.
Why: A creative that loses at the hook and one that loses at trust look identical in a cost column, and need opposite fixes. Diagnosing the step turned two losers into two lessons instead of one shrug.
Promote winners into a per-occasion scaling campaign
Each occasion's winning angle moved into its own scaling campaign rather than one shared campaign for the whole brand, with the delivery-slot, personalisation and corporate angles running in parallel through their own windows.
Why: One shared campaign carrying five occasions' creative resets its own learning every time a new winner is added mid-stream. Separate campaigns per occasion let each build a stable audience instead of restarting every few weeks.
Plan the seasonal cadence months ahead of the calendar
The hook library and the winning angles per occasion went into one shared document, with Ramadan, Eid, graduation and National Day variants shot four to eight weeks ahead of each occasion instead of scrambled together the week before.
Why: Corporate National Day orders alone need six to eight weeks of lead time, so a creative shot the week before is already too late for the audience that would have booked early. Planning ahead let the corporate angle reach coordinators while budget was still there to commit.
What changed
The headline number is in the table above, and the shape of it matters more: cost per purchase came down once five separate occasions each carried their own creative, rather than the whole year running on one script. The share of spend under target cost moved because each occasion finally had an angle built for its own worry, not because one ad went viral.
The clearer result was which worry actually sold. Four of the seven winners led with a logistics or trust detail — the slot, the card, the freshness, the corporate lead time — and the two clearest losers led with presentation or a borrowed urgency trick. The brand's instinct had been to make the box look better; what sold was answering the question the occasion itself had already raised.
There was a second effect the founder noticed first. Once the corporate creative ran on its own calendar with the lead time stated up front, comment-section questions about bulk orders turned into booked quotes months before National Day instead of a scramble in the final two weeks.
What we would do next
Take the occasion-specific worry into the product pages themselves, because the delivery-slot and freshness answers still live only in the ad and disappear once the customer lands on a page promising generic delivery. That is a conversion job stacked on this one, and it is where the next block of margin sits.
Second, build the graduation and National Day variants a full season early rather than four weeks out, since corporate lead time alone eats most of that runway. The peaks here are predictable to the week, and the brand that books its production calendar first keeps its kitchen from turning away orders it could have taken.