The scenario
The store sells cat and dog food, litter and accessories on its own Shopify site, with a warehouse in Kuwait City and a delivery van that runs three routes a day. Nine out of every ten orders are for cats rather than dogs, which the founder had noticed but never built the marketing around. KNET, Tabby and cash on delivery all sit at checkout, and the twelve-kilo bags and litter cartons are the products that actually carry the margin.
Demand had never been the issue. Kuwait's cat-owning population reorders on a schedule, and the store's own numbers showed a healthy base of repeat buyers. What kept breaking was availability: a specific imported brand would fall out of stock for two or three weeks at a time, and every ad running during that window was still promising a product the store could not fill, sending paid traffic straight into a frustrating out-of-stock page.
- Monthly revenue band
- pending client sign-off
- Average order value
- pending client sign-off
- Fulfilment
- Own warehouse and van fleet, three delivery routes daily across Kuwait
- Team
- Founder, one warehouse hand, one part-time customer-service reply
Every ad was a promise the warehouse could not always keep
When we tagged the previous quarter of spend, one pattern explained most of the wasted budget: the store's four running ads all named a specific imported cat food brand, out of stock for a combined six weeks across the quarter. Every day it sat unavailable, the same ads kept running and cost per purchase kept climbing, and the founder's instinct was to add budget rather than ask why performance had weakened.
The deeper issue was not the stockout itself — every store in this category loses a brand sometimes, and Kuwait's small import market makes it unavoidable. The issue was that nothing in the marketing addressed what a cat owner actually does when her usual bag is not there. She does not wait. She drives to a pet shop in Rai or Shuwaikh that has it, switches to whatever it recommends, or asks in a cat-owner WhatsApp group and takes the first confident answer.
The store had built its entire creative library around one brand's product shots, with none of it built around the moment that actually decides where a repeat customer buys next. The account did not have a targeting problem. It had never given anyone a reason to trust this specific store's judgement when the shelf was empty.
What we did — the creative angles
The losers are here on purpose. A test with only winners was never a test.
The out-of-stock switch, answered before she panics
Winner"Her usual brand is out again? Don't switch blind — ask us first."
- Format:
- Vertical, 16 seconds, one take, phone camera, warehouse shelf in the background
This was the single most frequent WhatsApp message in the store's history, and no ad had ever mentioned it. Naming the fear directly and offering the store's own judgement instead of a discount reframes the brand as a reliable adviser rather than one more seller. It was the clear winner on hook rate and held cost per purchase even during weeks a brand was actually short.
The bag the pet shop won't bring up your stairs
Winner"Twelve kilos. Third floor. No lift. We still bring it to your door."
- Format:
- Vertical, 14 seconds, real delivery footage carrying the bag up a stairwell
The single most Kuwait-specific objection here: local pet shops in Rai and Shuwaikh sell the same bags cheaper but won't carry a heavy carton past the ground floor, and any price gap stops mattering once a customer is expected to haul twelve kilos herself. This creative converted noticeably higher among older buildings without a lift, which the delivery address data confirmed after the fact.
The indoor-summer litter problem
Winner"Your cat is indoors ten hours a day now. Her litter should not smell like it."
- Format:
- Vertical, 15 seconds, before-and-after odour comparison filmed in an air-conditioned room
In summer, Kuwaiti cats move from a balcony or garden to spending nearly the whole day indoors, and a litter tolerable in a ventilated space becomes a real complaint in a sealed, air-conditioned flat. This angle tied a genuine seasonal shift to a specific upgrade rather than a generic pitch, and it performed well enough on hold rate to earn its own summer budget.
The reminder before the bag actually runs out
Winner"We texted her three days before the bag ran out. She never had to remember."
- Format:
- Vertical, 18 seconds, screen recording of a WhatsApp reminder next to a customer testimonial
Framed as proof rather than a pitch, this ad sold the convenience of a WhatsApp reorder reminder timed to when a twelve-kilo bag actually runs out, without ever using the word subscription. Cat owners here resist a recurring charge on principle but readily accept a message that saves them from an empty bowl at ten at night, and this creative held its cost per purchase especially well on retargeting.
Price comparison against the pet shops in Rai and Shuwaikh
Lost"Same bag. Cheaper than the shop in Rai."
- Format:
- Vertical, 12 seconds, split screen with prices on each side
A reasonable hypothesis that lost, and it is worth publishing because the reason is instructive. The store was not consistently cheaper once the local shop's own promotions were counted, and a customer who clicks a price ad and then sees the delivery fee close the gap learns only that the store leads with price rather than the convenience it actually wins on.
A generic happy-pet lifestyle shot
Lost"Everything your pet needs, in one place."
- Format:
- Vertical, 20 seconds, stock-style montage of dogs and cats with upbeat music
This was the closest thing to the store's original creative, and it stayed flat on every read. It says nothing specific to a Kuwaiti cat owner, ignores that nine in ten of this store's customers own cats rather than dogs, and a mixed-species montage answers no question anyone actually has — a reasonable explanation for why it never lifted past the account average.
A bundle: food, litter and an accessory together
Neutral"The whole restock in one box, one delivery fee."
- Format:
- Vertical, 15 seconds, unboxing a mixed carton at the door
Neutral in prospecting and useful further down. It did not win cold attention on its own, but it earned its place as a retargeting creative for customers who had already ordered food once, where the pitch is no longer about trust and is purely about not paying for two separate deliveries a week apart.
What we did — the optimizations, in order
Tag the back catalogue by brand availability, not just angle
We exported the previous quarter of ads and tagged each one by angle, hook, format and language, then cross-referenced every flight date against the store's own stock log to see how much spend had run while the named product was actually unavailable.
Why: In most categories a baseline audit only needs the ad data. Here it needed the stock data too — a large share of the wasted spend was not a bad creative, it was a fine creative pointed at an empty shelf, which no hook testing fixes on its own.
Build the angle sheet from the WhatsApp inbox and the delivery log
We read a year of customer messages and cross-checked them against delivery addresses, which is how the stairs-and-no-lift objection surfaced alongside the out-of-stock switching fear and the indoor-summer litter complaint.
Why: None of these three worries would show up in a normal review scrape, because they are about logistics and timing rather than the product. An angle sheet built from real operational data finds the objections a customer never writes in a public review but will mention on WhatsApp.
One testing campaign per platform, everything else frozen
One campaign per platform, one ad set per angle, equal budgets, five to seven days, with the audience, the offer and the landing page held identical across every ad set in the round.
Why: With a stock position that changes week to week, it would have been easy to blame or credit a creative for something the warehouse did. Freezing everything else, including which products were actually in stock during the test window, is what let us trust the read.
Read hook, then hold, then click, then cost
Each creative was judged in that order, with the reason it failed written down at whichever step it actually failed, rather than one blended verdict for the ad as a whole.
Why: The price-comparison creative and the generic lifestyle creative both failed on cost per purchase, but for different reasons — one attracted the wrong click, the other never earned attention at all. Diagnosing the step is what told us which one was worth a rewrite and which one simply had no future.
Promote winners and pause them the moment a brand runs out
Winners moved into a scaling campaign, and we linked the ad manager to the store's own stock alerts so a creative naming a specific product pauses itself the day that product runs out rather than three days later.
Why: The single biggest source of wasted spend before this engagement was a good ad running three or four extra days into a stockout because nobody was checking daily. Automating the pause is a small piece of plumbing that protects every creative that follows it.
Write the hook library down, tagged by season
Every hook tested, its result and the reason went into one shared document, tagged separately for the indoor-summer months and for the rest of the year, along with the National Day and winter grooming variants to shoot ahead of time.
Why: The litter-odour angle only makes sense for roughly a third of the year, and shooting it out of season wastes a good idea on an audience it does not apply to. A hook library sorted by when it actually applies is what stops the store re-running a summer ad in December and wondering why it went flat.
What changed
The headline number is in the table above, and the shape of it matters more than the size: cost per purchase came down even in weeks when a brand was genuinely out of stock, because the ads running then had stopped promising the missing product and started promising the store's judgement instead. The share of spend on creatives under the target cost moved from a minority to a majority once the out-of-stock and delivery angles took over the prospecting budget.
The more useful finding was which fear actually sells here. The three clearest winners were about logistics and trust — the switch, the stairs, the reminder — and the two clearest losers were about price and a generic benefit. The founder's instinct had been that this crowded market is won on discounts. What actually won was being the store a cat owner trusts when her usual plan breaks.
A second-order effect showed up in the delivery log: once the stairs-and-no-lift ad had run a few weeks, a noticeably larger share of new orders came from older buildings without a lift — an audience the store had never deliberately targeted before.
What we would do next
First, put an in-stock filter and an automatic alternative recommendation on every product page, because the same out-of-stock switching question the ads now answer is still an unanswered dead end once a customer actually lands on the site.
Second, shoot the National Day and winter grooming variants a month early rather than the week before, and pre-load a small budget for the switching angle specifically, since a supply gap on any given brand is predictable in frequency even if the exact week is not.