Commerce · Gulf · Via agency partner · white-label
D2C fragrance retailer:
rebuilt store, rebuilt search.
Platform migration with zero data loss, then eCommerce SEO and paid rebuilt on one roadmap. Revenue up 340% in seven months, with zero data loss in the migration.
A Gulf D2C fragrance retailer grew revenue 340% in 7 months. Delivered white-label through their agency of record: first a platform migration with zero data loss, then eCommerce SEO and paid media rebuilt as one roadmap instead of two competing budgets. The agency kept the relationship and the credit; we kept the delivery.
A Gulf D2C fragrance retailer had outgrown its storefront: revenue flat, marketing split across two suppliers who never saw each other's plans.
- Replatformed the store with zero data loss and every redirect mapped before cutover
- Rebuilt eCommerce SEO on the new foundation
- Re-planned paid and search as one budget instead of two competing ones
Result: 340% revenue growth in seven months, white-label through the agency of record
The store,
in two states.
| Before | After | |
|---|---|---|
| Revenue run-rate | Flat, pre-migration baseline | 340% revenue growth (7 months) |
| Catalogue and order history | Years of records at migration risk | Zero data loss, full history preserved |
| Search and paid budget | Split across two uncoordinated suppliers | One roadmap, one budget conversation |
- Revenue run-rateBeforeFlat, pre-migration baselineAfter340% revenue growth (7 months)
- Catalogue and order historyBeforeYears of records at migration riskAfterZero data loss, full history preserved
- Search and paid budgetBeforeSplit across two uncoordinated suppliersAfterOne roadmap, one budget conversation
A store the brand
had outgrown.
A direct-to-consumer fragrance retailer in the Gulf came to us through their agency of record, under a white-label arrangement: their brand on every call and every report, our team behind it. The brief was blunt: the storefront had been outgrown, and revenue had flattened while the category around it kept growing.
The platform was the first problem. Merchandising changes were slow. The storefront had also accumulated years of catalogue history (products, orders, customer records), and that history made every conversation about replatforming start and end with the fear of losing data. Meanwhile the marketing budget was split across two suppliers: one running SEO, another running paid, neither seeing the other's plan.
That split was quietly expensive. Paid was bidding on queries organic already won, category pages were optimised for neither channel, and nobody owned the question of which products deserved the traffic. The agency asked us for what they couldn't staff internally: one team to rebuild the store and then run search and paid against a single roadmap.
Migrate first,
then compound.
Migration mapped before anything moved
Products, orders, customer records and URL history inventoried and mapped in full before the replatform began. Zero data loss was the contract, not the aspiration. Nothing was switched off until its destination was verified.
The replatform treated as an SEO project
Every URL that had ever earned a ranking got a mapped redirect, and the new information architecture was designed around search demand rather than the org chart. A migration that loses search equity just moves the revenue problem.
eCommerce SEO rebuilt on the new foundation
Category and product architecture, structured data across the catalogue, and content that answers buying questions: the discipline described on our eCommerce SEO service, applied to a store finally fast enough to deserve it.
Paid rebuilt on the same roadmap
Search and paid planned in the same room: paid covering the terms organic hadn't won yet, pulling back where it had, and feeding query data back into the SEO plan. One budget conversation instead of two competing ones.
Reporting under the agency's brand
A weekly review in Prism, written up in the agency's templates and tied to revenue rather than vanity metrics, ready for them to present as their own. The end client never knew we existed, which is the arrangement working as designed.
The decisionsThree forks, and the roads not taken.For the detail readers
Most of what determined this outcome was decided before any code shipped. These are the arguments that actually happened, and what lost.
Replatform first, or run marketing on the old store while rebuilding?
Migrate first, then compound. Search and paid were held at maintenance level until the new storefront was live and verified.
Running a full acquisition push in parallel. It is the tempting option because it shows activity immediately. But it wastes most of the spend, because traffic driven into a store you are about to replace converts on the old checkout, on the old speed, against categories that are about to change URL. Every pound of that spend buys a result you cannot keep.
Treat the migration as an engineering project or an SEO project?
An SEO project that happens to involve engineering. Redirect mapping and information architecture were designed before the build schedule, not retrofitted to it.
The conventional sequence: build the store, then hand the URL list to whoever does SEO. That order is how migrations lose their rankings: by the time anyone maps redirects, the architecture is already fixed and the mapping becomes damage limitation rather than design.
Keep two suppliers on search and paid, or consolidate?
One roadmap, one budget conversation. Paid covered the terms organic had not won yet and pulled back where it had.
Leaving the existing split in place, which is the lower-friction option politically. It is also the one quietly paying twice for the same click: bidding on queries the site already ranked for, with neither supplier able to see the overlap because neither saw the other's plan.
The work,
itemised.
The hard partsWhat this actually cost.A case study with no hard parts is a brochure
The catalogue history was the real work, and it was unglamorous. Years of products, variants, discontinued lines and order records accumulate inconsistencies that only surface when you try to move them: the same product entered twice under different SKUs, categories that exist in the database but were never linked from anywhere, URLs that redirect to URLs that redirect. None of that is visible from the storefront, and all of it has to be resolved before a migration can claim zero data loss and mean it.
The second cost was pace. Holding acquisition at maintenance level during the rebuild is correct, and it feels wrong to everyone watching a dashboard. There is a period where the numbers do not move, and the only honest report is "the work that will move them is not finished yet." That stretch is where migrations get abandoned halfway, which is the worst possible outcome: the disruption of a replatform without the compounding that justifies it.
Working white-label added a real constraint rather than just a branding one. Every recommendation had to be explicable by someone who had not been in the room when it was made, because the agency presented it. That is a useful discipline: a decision you cannot summarise in two sentences for someone else to defend is usually a decision you have not finished making. But it is slower than talking directly to the person who signs off.
When this pattern
transfers, and when it doesn't.
The outcome here came from a specific situation. Saying which parts generalise is more useful than implying all of them do.
This pattern transfers if
- You have real catalogue history and a platform you have outgrown: the migration risk is what is keeping you on it.
- Search and paid are run by different people who do not share a plan.
- The store converts acceptably; the constraint is that growth has flattened rather than that nothing works.
- You can tolerate a quiet quarter while the foundation is rebuilt.
It probably doesn't if
- You need revenue to move this quarter: the sequence here deliberately trades near-term activity for compounding, and a different plan serves an urgent number better.
- The platform is genuinely fine and the real problem is product, price or demand. Replatforming will not fix any of those and is an expensive way to find out.
- You are pre-launch with no catalogue history. Most of the difficulty here, and most of the cost, was in the history.
The services behind this result
Everything used on this engagement is a service you can buy on its own: direct, or under your agency's brand.
Want a result like this
on your own numbers?
A senior strategist records five minutes on your website: speed, search rankings, AI visibility, conversion leaks. Yours to keep, no strings.
Response in 48 hours · No retainer required · NDA standard
Last updated