We design, build and launch custom storefronts on Next.js for retailers doing $240K–$10M online. Arabic and English from the first line of code, your own code at the end, and no share of your revenue.
A $15,000 reference build against $75k–135k a year for Shopify Plus, with every figure on this page sourced.
We will tell you on the call if rebuilding is the wrong move this year.
Five-year figure is the full build plus hosting at the top of our range. Comparison and sources in section 12.
Three ways this usually goes wrong. You have probably lived at least one of them.
Fast, affordable, and gone. No documentation, no tests, no handover — so the next person has to reverse-engineer what the last one did before they can change anything. You paid twice for the same feature and the second time cost more.
Sold by the person you liked, built by whoever was free. Every question becomes a change request, every change request becomes a quote, and the people who understood your business were never the ones writing the code.
A theme decided your business model. Cash on delivery arrived as a plugin, Arabic as a translation layer, delivery as a flat rate. Your store now works the way the template works, and your operations bend around it.
None of these are unusual. All of them end the same way — a site that is technically live and commercially disappointing.
Nobody replatforms because of pricing. They replatform because Tuesday has become unmanageable.
The catalogue does not model what you actually sell, so every listing needs a manual fix. The workaround becomes the process.
Inventory sits in one system and the storefront in another. A customer buys the last item that sold in-store an hour ago, and you cancel the order.
Every request routes through a roadmap you do not control, and the answer to a good idea is usually no.
Above roughly $800k a month, one major hosted platform converts its fee into a percentage of your revenue, capped at $40,000 a month. The better your year, the larger the invoice.
Every extension is a subscription. Every upgrade is a re-purchase and a re-test. The feature you depend on most sits on someone else's roadmap.
Since 2023, dedicated merchant success support has been withdrawn from Plus merchants below roughly $2M in annual recurring revenue and moved to shared queues. Enterprise pricing, self-service support.
A checkout designed in California asks the wrong questions here. These are not preferences. They are the conditions of trading in this region.
Around 70% of consumers across the Middle East still prefer it. It carries its own rules per product, its own fee, its own reconciliation — and its own cash-flow cost, because money collected on delivery is money not yet reinvested in stock. We model that, rather than hiding it in a plugin.
Local payment methods are not an integration to add later. In Saudi Arabia and Kuwait they are the default, and a checkout without them loses the sale at the last step.
Much of the region lacks the address infrastructure that Western logistics assumes. Delivery is resolved per city, live at checkout, against rates the courier is actually paid.
Not a translation layer. Navigation, cart, search, carousels and arrows all mirror. Tested on a phone first, because that is where Gulf traffic is.
Around 72% of orders in Jordan are placed on a phone, and a mobile-optimised store converts roughly three times better than one that was designed for a desktop and shrunk. We design the small screen first and let the desktop inherit from it.
Saudi e-invoicing is a build requirement. Retrofitting it afterwards means reopening the order pipeline on a live store.
The honest objection to a custom build. Here is our answer, in writing.
If something we built stops working the way it was specified, we fix it. No support contract required, no hourly rate, no argument about whose fault it is. Three years from launch.
The guarantee lapses on code modified by a third party — which is the trade for owning it outright. Everything up to that point stands.
I was genuinely impressed by the results, to the extent that I decided to hire them again to revamp and upgrade my website. Once again, the Umbrella500 team exceeded expectations.
Baha AbdeenCEO, WaqteeWhy this answers the question. Waqtee hired us to build their store, then hired us again to rebuild it when they outgrew it. They came back because they chose to — which is the only proof of a non-lock-in that is worth anything.
Every line below is running in a store we have built. The proof sits next to the claim, because a claim on its own converts nobody.
A gift order has a sender who pays and a recipient who receives. Treating them as the same person is why gift orders arrive at the wrong address, with the price in the box.
The people you meet on the call are the people who do the work. That is the whole difference between this and the last two attempts.
Twenty years as a consultant, technology expert and e-commerce specialist. Co-founder and COO of ShopGo, where she helped more than 1,000 MENA businesses scale online — many of them past $100K a month. She has seen every way a store fails in this region before it happens to yours, and she leads the technology practice.
Co-founder of ShopGo.me, a SaaS platform that helped more than 1,000 MENA entrepreneurs launch online stores and was acquired. He has since supported over 13,000 entrepreneurs across the region, and holds the JCI Ten Outstanding Young Persons award, the NASA App Prize and the European Youth Award. He leads client relationships and the commercial side of every build.
The pairing is the point. Most agencies give you a technologist who cannot advise on the business, or a strategist who cannot build. A store fails when those two people never speak. Here they are the same two people — who built and sold a commerce platform together before this — on your project, from the first call to the last deployment.
One of the largest beauty and fragrance platforms in Saudi Arabia — founded in 2014, headquartered in Dammam, over 20,000 products from more than 1,000 brands, more than 2 million shoppers, around 101 employees, and $22.2M raised from investors including Saudi Aramco's Wa'ed Ventures, Raed Ventures and Wamda Capital. It runs on Magento. We have been its growth partner since 2021 — we scaled that marketplace past one million orders on infrastructure and growth marketing, and 72% of them were placed on a phone.
Magento absolutely operates at that scale. What it takes to get there is the part nobody puts in the sales deck: sustained custom development, specialist engineers, and a continuous spend on keeping a heavily customised installation stable through every upgrade. At that volume, the maths works. The platform earns its cost.
Now look at your own numbers. If you are doing $240K to $10M a year, you would be buying the same complexity and the same maintenance bill without the revenue or the funding that justifies either. You would spend year one customising your way to features that should have been in the build, and every year after that keeping those customisations alive.
We know what that road costs because we have been on it. You can start on the other side.
| At that scale | You |
|---|---|
| 20,000+ products | A few hundred to a few thousand |
| $22.2M raised | Funded by your own sales |
| ~101 employees | A team you can name |
| Magento earns its cost here | It would not earn it here |
The company figures — products, brands, shoppers, headcount, funding — are public: Crunchbase, PitchBook, Owler, MenaBytes and Wamda. The order, mobile and tenure figures are ours, from the engagement. The client is not named.
Published 2026 figures, with sources listed underneath. We have not rounded them into something punchier, and we have not picked the worst case.
| umbrella500 | Shopify Plus | Adobe Commerce | WooCommerce | |
|---|---|---|---|---|
| Platform licence | None | From $2,300 / month | $22k–80k / year | None — open source |
| Cut of your revenue | None | A percentage of sales above the threshold, capped at $40,000 / month | — | None |
| Transaction fee | Payment gateway only | 0.2%, or about 2.15% + $0.30 on their own payments | Payment gateway only | Payment gateway only |
| Year one, ~$1M revenue | ~$19k | $75k–135k | $150k–300k+ | $10.5k–24k |
| Year three onward | ~$4k / year hosting | Same or higher | Same or higher | Same or higher, as the plugin stack grows |
| Who owns the code | You do, outright | The platform | Licensed to you | You own the site; each plugin stays licensed |
| Custom checkout | Unrestricted | Top tier only | Possible, costly | Possible, and plugin-dependent |
| Upgrade cost | None — it is your code | Platform changes arrive on their schedule | Re-licence and re-test | Every WordPress and plugin update is a re-test |
Every competitor line is plotted at the bottom of its published range — $75,000 a year for Shopify Plus against a researched $75k–135k at $1–2M GMV, and $150,000 a year for Adobe Commerce against $150k–300k+. The real figures are likely higher. Ours is the full $15,000 reference build plus $4,100 a year of hosting at the top of the $275–340 monthly range.
WooCommerce is cheaper than us for two years, and more expensive from year three. Its licence is free, so its cost is off-invoice and we have estimated it rather than left the column blank: plugin and extension subscriptions $1,500–3,000 a year, hosting for a catalogue of a few thousand products $1,800–4,800, and eight to fifteen hours a month of maintenance at $75–90 an hour — $7,200–16,200. That totals $10,500–24,000 a year, and the chart plots the bottom of it. If your store is small and someone in-house enjoys maintaining it, WooCommerce may never cross. Section 13 says so.
| Cumulative | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 |
|---|---|---|---|---|---|
| umbrella500 | $19,000 | $23,100 | $27,200 | $31,300 | $35,400 |
| WooCommerce | $10,500 | $21,000 | $31,500 | $42,000 | $52,500 |
| Shopify Plus | $75,000 | $150,000 | $225,000 | $300,000 | $375,000 |
| Adobe Commerce | $150,000 | $300,000 | $450,000 | $600,000 | $750,000 |
We will put your real numbers into this table on the call, including the ones that do not favour us.
Most of our clients arrive from one of these four. We have shipped production stores on Magento and WooCommerce, so this is not a competitor teardown — it is what we found out the hard way.
Excellent until you grow.
Fast to launch, reliable, well supported at the start. The problem arrives with success: above roughly $800k a month the fee becomes a percentage of your revenue, capped at $40,000 a month. Checkout customisation sits behind the top tier. And since 2023, dedicated merchant success support has been withdrawn below roughly $2M ARR. You keep paying enterprise rates for a shared queue.
Choose it if: you are under $240K a year, want to launch in weeks, and will never need to change the checkout.
Powerful, and priced like it.
Genuinely capable, genuinely enterprise. Licences run $22k–80k a year before a single hour of development, and it needs specialists to run. We are Magento-certified and have delivered on it at enterprise scale — which is exactly how we know what it costs to keep a customised installation stable. For most retailers in this region the capability is real and the cost is not justifiable.
Choose it if: you are running multiple brands across multiple regions with a team to operate it.
Free to start, expensive to run.
We built on WooCommerce for years, and it is the reason we stopped. It works well at small catalogue sizes. Past a few thousand products it slows, and every capability you need arrives as another plugin — each one a subscription, a security surface, and something that can break on the next WordPress update. The cost is not on the invoice; it is in the hours spent keeping it standing.
Choose it if: you have a small catalogue, simple needs, and someone in-house who enjoys maintaining it.
Costs more on day one. Costs nothing on day 400.
You own the code. Nobody takes a percentage of your sales. Cash on delivery, Arabic and per-city delivery are modelled in rather than bolted on. It takes eleven to sixteen weeks rather than three, and it only makes sense above a certain size — which is why we will tell you when it does not.
Choose it if: you are doing $240K or more online, the platform is now the thing slowing you down, and you intend to be trading in five years.
If you are under $240K a year, Shopify is probably the right answer and we will say so on the call.
Every tag and pixel wired correctly, and every event passed accurately, so the ad platforms get clean conversion signal and optimise toward people who actually buy. Included in the build rather than sold as an extra.
Meta, Google, TikTok and Snapchat, through Google Tag Manager — installed, tested and verified end to end rather than pasted in and hoped for.
View, add-to-cart, checkout and purchase go client-side from the browser and server-side through the Conversions API. Ad blockers and lost cookies stop being holes in your reporting.
Smart bidding, lookalikes and retargeting all run on the data you give them. Complete signal in means the platform's own optimisation gets better decisions out — lower cost per acquisition, more return on each dinar.
We are describing the plumbing rather than the advertising. A clean signal makes good media buying work harder; it does not turn bad media buying into good.
Press the toggle. The panel below is live markup on this page — direction, alignment, order and copy all change. It is the same switch a shopper uses.
Slazenger Havoc — Silver Stainless Steel, Black Dial (Men's)
64JOD86 JOD
In stockFree delivery across Jordan · Cash on delivery available
Set side by side, the two captures show what a right-to-left build has to get right beyond text: the menu and cart swap sides, the search icon crosses the field, the category toggle reverses, the chip row scrolls the other way and the arrow on the promoted collection points the other way. The panel above carries the translated copy; these two carry the layout.
The order matters more than the length. We build the parts most likely to fail at the start, and show them running before the rest is committed.
We build the risky parts first and prove them before the full build.
Waqtee sells branded watches and jewellery — a leading retailer in Jordan that expanded into the Gulf. Around 9,000 products, every one bilingual, with gifting as the reason most orders exist and cash on delivery as the way most of them are paid for.
Measured on the reference build described below, not projected.
The gift flow in section five is theirs. So is the phone-first sign-in, the per-city delivery pricing and the Arabic search. Every screenshot on this page is that store — no mockups, and real prices in Jordanian dinars to three decimal places.
Yes, outright. The repository is yours from the first commit, in your own accounts. There is no licence to renew and nothing that stops working if you stop paying us.
You keep everything — the code, the data, the infrastructure and the documentation. We will hand over to your team or to another agency, and we will do it properly. A store you cannot leave is a store you do not own.
There is no licence and no share of your revenue. You pay once to build it and then pay for hosting. The comparison table above shows the arithmetic at roughly $1M a year in online revenue, using published 2026 figures.
We do, on a flat retainer — or your own team, if you have one. In any month where nothing breaks, that budget goes into new features instead of sitting idle as an insurance premium.
Eleven to sixteen weeks for a full store, depending on how many systems it has to talk to. The riskiest parts are built first and demonstrated running, so you can stop or change direction before the whole budget is committed.
Roughly $275 to $340 a month, all in — hosting, search, messaging and analytics — and it runs in your own cloud account, billed to you directly. We do not resell it with a margin on top.
That is the point of owning it. Content, products, pricing, specification fields and delivery rates are all merchant-editable without a developer — those were built to be changed. Anything deeper is code, and you have the repository, the documentation and a standard stack, so your own developer or another agency can work on it. We are not the gatekeepers of your store.
One source of truth, read directly by the storefront rather than reconciled overnight. The storefront reads the same stock record the shop sells from, so the last item sold over the counter is gone from the website in the same moment. Where the shop runs a point-of-sale we cannot read live, we hold a buffer per product and surface it — because a cancelled order costs more than a missed sale.
Tell us what you sell and where the current platform stops you. Thirty minutes, no cost, and we will say so if replatforming is the wrong move for you this year.
Or write directly: saleem.najjar@umbrella500.com