Khaadi is the brand every Pakistani fashion founder benchmarks against, which makes it the most useful teaching case in the market. It started in 1998 as a single handloom shop in Karachi's Zamzama and now runs around sixty stores across roughly thirty Pakistani cities, plus stores in the UK, UAE, Qatar, Bahrain and — since 2024 — the United States. Its Experience Hub at Dolmen Mall Karachi runs to 32,000 square feet with a café built into it, which tells you the brand stopped thinking of stores as places to transact a while ago.
What interests me is not the retail footprint. It is the gap between how well Khaadi runs its brand and how unevenly it runs its paid media. When we pulled the Meta Ad Library, the Google Ads Transparency Center and TikTok side by side, a pattern showed up that I see in almost every large Pakistani retailer I am brought in to fix.
The short version
Khaadi runs brand-building on Meta and hard-sell discounting on Google, with almost no coordination between the two. It has no TikTok advertising at all in a market where TikTok is the fastest-growing product discovery channel. The brand is strong enough to absorb this. Your brand probably isn't.
Meta: disciplined, consistent, and running out of road
At the point of analysis, Khaadi Pakistan was running roughly a dozen active ad units on Meta, all launched on the same day, all built from one creative concept for its ready-to-wear line. Around nine were near-identical studio images on plain backgrounds with different garments swapped in. One was a well-made 33-second video. One was a boosted page post pointing at Facebook rather than the store.
The discipline is real: one idea, one layout, one call to action, full placement coverage across Facebook, Instagram, Messenger, WhatsApp and Threads. A shopper who sees any single unit knows immediately whose ad it is.
The cost of that discipline is that there is nothing to optimise toward. Identical copy across every unit means no hook testing, no urgency, no offer, and no description tailored to the specific garment on screen. A customer who sees three of these in a week has seen the same ad three times. Frequency climbs, click-through decays, and the account has no fresh creative queued to rotate in.
Three further gaps are worth naming, because they are the same three I find in most large fashion accounts here:
- No carousel or collection ads. For a catalogue business, these are among the highest-performing formats available, and they were simply absent.
- No visible retargeting. No cart-abandonment messaging, no dynamic product retargeting, nothing aimed at warm audiences. Every rupee was working on cold traffic.
- No vertical-first video. The assets read as feed crops rather than content shot for Reels and Stories, which is where the attention actually is.
Google: the same brand, playing a completely different game
Google is where the contrast becomes instructive. The Transparency Center shows roughly 200 creatives run under Khaadi's legal entity in Pakistan, with data going back to 2018 and activity continuing right up to the pull date. The format mix skews heavily to display, with a substantial block of search text ads and a meaningful video library behind it.
And the messaging is almost entirely promotional. Search headlines lead with discount percentages. Display banners have sale graphics burned into the creative itself. Even the video asset is framed around a specific promotion. Sitelink extensions branch into sale pages, fabric sale pages, ready-to-wear sale pages.
So the same brand runs quiet, discount-free brand advertising on Meta and loud, percentage-off direct response on Google. Used deliberately — social for brand, search for harvest — that is a legitimate structure. But there was no evidence of deliberateness. The visual languages do not match, the messaging does not hand off, and a customer moving between the two channels gets two different impressions of what kind of brand this is.
The clearest miss on Google is Shopping. For an apparel retailer with a catalogue this size, product listing ads are normally one of the highest-return formats available, and Khaadi was running almost none of them in Pakistan. Its own UK operation runs a large Shopping presence, so the capability plainly exists inside the group — it just has not been pointed at the home market.
TikTok: the hole in the strategy
Khaadi has roughly thirty thousand TikTok followers and, at the time of review, no official TikTok advertising whatsoever. Meanwhile third-party resellers were actively running TikTok ads on Khaadi product.
I want to be precise about why this matters. TikTok in Pakistan is not primarily a conversion channel — it is where product discovery now begins, particularly for buyers under thirty. A brand with no presence there is not losing a sales channel so much as losing the first step of the funnel, and handing the framing of its own product to resellers who have no interest in protecting its positioning.
International: three markets, three different playbooks
The overseas picture is the most revealing part of the whole study. The UK operation, when active, is aggressively promotional — real "was £X, now £Y" pricing, Boxing Day and Eid sale waves, Clearpay and Klarna instalment messaging, and explicit "ships from the UK" copy that answers the customs-and-delay objection before a shopper raises it. The UAE campaigns localise the same master creative with Tamara and Tabby instalments and a free shipping threshold in dirhams. Pakistan and the US run the same evergreen, discount-free new-arrivals campaign in parallel.
What this shows is that the group knows how to localise. It builds one master creative kit and swaps the offer, the payment providers and the trust signals per market. That is a genuinely sophisticated operating model, and it is the single most transferable idea in this case study for any Pakistani brand selling to diaspora customers.
What I would change first
- Introduce two or three copy variants per campaign so Meta has something to optimise between. This is a week of work, not a quarter.
- Build a real retargeting layer. Warm audiences are the cheapest revenue in the account and nothing was pointed at them.
- Turn on Google Shopping properly in Pakistan. The feed already exists for the UK.
- Take TikTok seriously as a discovery channel before resellers finish defining the brand there.
- Decide consciously whether Meta is the brand layer and Google the performance layer — then make the creative reflect that decision instead of drifting into it.
The pricing ladder is the quiet lesson
Underneath the media, Khaadi runs a genuinely well-built pricing structure: entry-tier unstitched lawn that pulls in first-time and price-sensitive buyers, mid-tier stitched ready-to-wear for the convenience shopper, and a premium festive line that protects margin and brand prestige. Sale events are used to clear aged stock without permanently marking down the core range.
That ladder is why the brand can hold a two-and-a-half thousand rupee lawn suit and a thirty-thousand rupee formal piece under one name without either cannibalising the other. Most Pakistani brands I work with have either no ladder at all or one so compressed that every discount eats directly into the core range.
Where Khaadi is exposed
Three risks stand out. Raw fabric is largely imported across the category, so rupee volatility squeezes margin industry-wide, not just here. Acquisition costs on Meta and Google keep climbing as more brands bid for the same audience. And global fast fashion is competing for exactly the same urban, mobile-first wallet.
Against that, Khaadi's real advantage was never the clothes. It was turning a craft story into a retail system — stores that work as brand theatre, an eCommerce operation built market by market rather than copy-pasted, and a price ladder wide enough to hold the whole audience. That combination is hard to copy, and it is the actual lesson here.
If you are building a fashion brand in Pakistan and want this kind of channel audit run on your own accounts, that is a large part of what I do as an eCommerce consultant in Pakistan.
