Aahfil.
Worked example — a composite of patterns we see across Kuwait stores in this category, not a single client engagement. No figures are published here.

Kitchen & homeware · Creative testing

How Kuwait coffee-equipment brands fix ads that never show the coffee

a Kuwait coffee-equipment brand, three years in, selling espresso machines, grinders and accessories direct to consumer

Coffee-equipment brands in Kuwait run polished studio ads that never show the machine actually making coffee, and cost per purchase keeps climbing. The fix pulls real questions out of the WhatsApp inbox, builds an angle sheet around the pour, the plug and the warranty, and tests each one against the studio look. A shaky phone camera usually wins.

A composite worked example drawn from patterns across Kuwait stores in this category, not a single client engagement. No figures are published.

At a glance

Category
Kitchen & homeware
Lever
Creative testing
Platforms
metainstagramtiktoksnapchat
Stack
shopifyknettabbycodmeta-capiga4

The scenario

The brand sells espresso machines, grinders and accessories on its own Shopify store, with KNET, Tabby and cash on delivery at checkout, and a catalogue that opens around the price of a mid-range phone and climbs from there. Three years in, it has genuine repeat buyers among the kind of household that hosts often, but every campaign since launch had been shot the same way: a machine on a marble counter, lit like a catalogue page, with a slow-motion pour dropped in for polish.

Nobody on the team had ever stood in a normal kitchen with a phone and just pressed the button. The founder had assumed a premium machine needed a premium-looking ad, and the account had never actually tested that assumption against anything rougher.

Monthly revenue band
pending client sign-off
Average order value
pending client sign-off
Fulfilment
Own stock, local courier, next-day across Kuwait, in-house technician for warranty repairs
Team
Founder plus one part-time content creator

Beautiful ads, and a machine nobody had watched actually pull a shot

When we tagged the previous quarter of spend, six creatives came back, and all six were the same idea shot from a different angle: the machine, lit well, pouring in slow motion, with a caption about premium taste. Cost per purchase had been climbing, and the founder's instinct had been to raise the budget and let the good-looking ads work harder — the one move that makes an exhausted idea more expensive rather than less.

The brand's own WhatsApp history told a different story. The same three questions kept coming back: does it work on a Kuwait plug without an adapter, what happens when a part breaks, and can it actually pull a shot as good as the one at the demo counter, or is that just how it looks on camera. None of the six ads answered any of them, because none had been built from a question. They had been built from a mood board.

The account did not have a targeting problem or a discount problem. It had a proof problem — six ads that all looked expensive and not one that looked true.

What we did — the creative angles

The losers are here on purpose. A test with only winners was never a test.

  1. The unbroken shot, pulled in a normal Kuwaiti kitchen

    Winner
    "No cuts. This is what it actually pulls like."
    Format:
    Vertical, 22 seconds, one unbroken take, phone camera, real kitchen counter

    This directly answered the demo-counter worry: does it actually pull a real shot, or is that only true under studio lighting with an off-camera assist. Filming it in one take, with the grind, the tamp and the crema visible without a cutaway, turned the ad into evidence rather than a claim, and it won clearly on hook rate and hold rate across every platform.

  2. The plug and the local warranty, shown not claimed

    Winner
    "Kuwait plug, no adapter. And here is who fixes it if it breaks."
    Format:
    Vertical, 15 seconds, close-up on the plug then the technician's workbench

    The second most common WhatsApp worry, and a trust objection rather than a product one: shoppers who had bought an imported machine before and been left with no local service when it failed. Showing the actual technician and plug did more for retargeting cost per purchase than any discount tried, because it removed the reason a saved product page never turns into an order.

  3. One machine, before Ramadan hosting starts

    Winner
    "Get it dialled in before the first majlis of Ramadan."
    Format:
    Vertical, 20 seconds, family gathering setting, machine pulling shots back to back for guests

    Placed the machine inside the exact moment it is bought for in Kuwait — not a solo morning coffee, but back-to-back cups for a house full of guests. It performed strongly in the pre-Ramadan weeks and earned its keep as the anchor for a set offer pairing the machine with cups and a tray.

  4. The price comparison against the demo counter

    Lost
    "Same machine you tried at the counter. Look at the difference."
    Format:
    Vertical, 15 seconds, split screen with the two prices on screen

    A reasonable hypothesis that lost. It drew a large volume of cheap clicks and almost no orders: shoppers who respond to a price-comparison hook in this category are already shopping on price alone and keep looking after clicking. It also invited a comparison the brand could not fully control, since it cannot verify any one store's counter price on a given week.

  5. The influencer unboxing, professionally lit

    Lost
    "Unboxing my new espresso setup."
    Format:
    Vertical, 30 seconds, professionally shot unboxing with music

    Strong on paper and flat in the auction. It read as sponsored content rather than proof, and the polish worked against it exactly as the unbroken kitchen shot's roughness worked for it: this audience had learned to discount a studio-quality unboxing as advertising the moment the music started.

  6. The instalment maths for the higher-end machine

    Neutral
    "Split it with Tabby — about the price of a weekly coffee run."
    Format:
    Vertical, 14 seconds, text-on-screen with the instalment figure over the machine

    Neutral cold and useful further down. Framing the Tabby instalment as a small weekly figure did not win attention on its own, but it earned its place retargeting people who had reached the product page for the flagship machine and left without checking out, which is exactly where the price objection actually sits.

  7. Arabic gahwa and espresso from the same counter

    Winner
    "Dallah on one side, espresso on the other. Same kitchen."
    Format:
    Vertical, 18 seconds, split composition, dallah and espresso machine side by side on one counter

    Tested on the theory that a household hosting Arabic coffee alongside espresso would respond to seeing both honoured rather than the machine positioned as a replacement for the dallah. It performed close to the unbroken-shot winner and confirmed the pattern: this audience is not choosing between the two, and an ad that assumes it has to choose loses trust before it finishes.

What we did — the optimizations, in order

  1. Tag the back catalogue before making anything new

    We exported the previous quarter of ads and tagged every one by angle, hook type, format and language, then worked out what share of spend had sat on creatives below the target cost per purchase.

    Why: Six creatives that all looked different at a glance turned out to be one idea shot six ways, and no amount of new production would have fixed that until someone counted it. The share-of-spend number became the single metric the whole engagement was judged on.

  2. Build the angle sheet from the inbox, not a mood board

    We read a year of WhatsApp questions, reviews and return reasons and pulled out the seven that came up most, then wrote each one first as the customer's own sentence in Kuwaiti dialect.

    Why: The brand already knew what buyers worried about; it had simply never turned that worry into an ad. An angle sheet built from real messages also settles the argument about whose creative idea is best, because every line carries a receipt.

  3. Shoot rough on purpose

    Alongside the usual studio pass we shot a second, deliberately unpolished round on a phone in a real kitchen, with no lighting rig and no cutaways, because the angle sheet pointed straight at proof rather than aspiration.

    Why: A category sold on demonstration does not get proven by a more beautiful ad; it gets proven by a rougher one that removes the suspicion of a trick. The unpolished round is what carried the winners.

  4. 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 the whole test.

    Why: If the offer or the audience moves at the same time as the creative, the result explains nothing. Freezing everything else is what let us say with confidence that the unbroken shot won because of what it showed, not because of anything running alongside it.

  5. Read hook, then hold, then click, then cost

    Each creative was judged in that order, with a written reason for the step it failed at, rather than a single verdict on the ad as a whole.

    Why: The influencer unboxing and the price comparison both looked like plain losses in a cost-per-purchase column, but they failed at different steps for different reasons, and only one of them was worth trying again in a rougher cut.

  6. Bundle the winner into a set offer

    Once the Ramadan-hosting angle proved out, we paired the flagship machine with cups and a tray at a stated saving on the product page, rather than surfacing the bundle only as a checkout upsell.

    Why: A winning creative that only sells one unit leaves margin on the table in a category built around gatherings. Showing the set where the buying decision actually happens raised the average order value the ad had already earned.

  7. Write the hook library down

    Every hook tested, its result and the reason went into one shared document, along with the seasonal variants to shoot before the next Ramadan and Eid.

    Why: The asset that outlives the engagement is not the winning ad, which will fatigue with repetition. It is the written record of what proof this specific audience needed before it believed the machine, which stops the brand paying to relearn the lesson next year.

What changed

The headline number is in the table above, and the shape matters more than the size: cost per purchase came down while spend went up, because the share of budget on creatives under the target cost moved from a small minority to a clear majority. No single ad went viral — the winners simply kept converting as spend increased, which the old studio ads never did.

The clearer finding was about what proof looks like in this category. Three of the four winners were built around showing something rather than claiming it — the unbroken pour, the real technician, the dallah beside the machine — and the two losers read as advertising rather than evidence, however well they were lit. The instinct had been to make the ads look more expensive; what sold was making them look less like ads.

Because the winners pre-answered the plug and warranty questions, the volume of pre-purchase WhatsApp messages the founder fielded by hand dropped, and the ones that still arrived were closer to a decision than a doubt.

What we would do next

Take the same rough, unbroken-take format into the air fryer and serveware side of the catalogue, since the pattern this category rewards — proof over polish — is unlikely to be specific to coffee alone.

Second, build the Ramadan and Eid variants of the winning set offer a month early rather than the week before, because hosting season in this category is predictable to the week, and the brand that has its proof-led ads ready first collects the audience everyone else is still shooting.

take it and use it

Steal this

  • In a demonstration category, a rougher phone shot beats a polished studio shot. Proof outsells production value.
  • Your best angles are already in your WhatsApp inbox. Read a year of it before you book a shoot.
  • Do not position a new machine as a replacement for the dallah. This audience hosts with both.
  • Bundle the winning creative's product into a hosting set on the page itself, not as a checkout afterthought.

Frequently asked questions

What sells a coffee machine online in Kuwait?+

The shot being pulled, in one unbroken take, in a normal Kuwaiti kitchen — then the warranty, the local service and what happens when a part needs replacing. A studio-lit slow-motion pour reads as marketing; a shaky phone shot reads as proof, and proof is what this category is actually buying.

How many creatives should a Kuwait kitchen-appliance brand test at once?+

Six to eight angles in one round, one ad set each, equal budgets, everything else frozen. Fewer than six leaves too much of the objection space uncovered; more than eight thins each ad set's budget below the point where a week gives you a real answer.

Why did our polished ads stop converting?+

Usually because polish and proof are different things, and a demonstration category buys proof. If every ad is lit like a catalogue and none shows the product actually being used in a normal home, the audience that still doubts the demo-counter version of the product has no reason to change its mind.

How do you raise average order value in homeware?+

Sell the set, not the item. Cups with the machine, a tray with the dallah, the second basket with the air fryer — bundled on the product page at a stated saving rather than offered as an upsell after the decision is already made.

How long before a creative test gives you an answer?+

Five to seven days per round for the attention metrics, and long enough after that to gather a meaningful number of orders for cost, which for a higher-ticket item in a Kuwait-sized market usually means running the winner for a further two weeks before trusting its cost per purchase.

Same method, other categories

Want this run on your store?

Tell us where your store is now and we'll come back with the first thing we would test, and why.

WhatsApp us+965 51557699
Message us on WhatsApp