A buyer’s guide · Dubai and the UAE
Most software decisions in Dubai are made the wrong way round: a product is chosen, then the business is bent to fit it. This is the guide we would want an owner to read before spending anything — including the parts that argue against hiring us.
Buy vs build · What drives cost · Questions to ask any vendor · What goes wrong
The short version
If your process is unclear
Fix that first. Software will only make a confused process run faster.
If a product covers most of it
Buy it. A subscription is a much smaller commitment than a build.
If the gap is where you compete
Then custom is worth it — and worth doing one workflow at a time.
We build custom software for a living, so read the above knowing that. It is still the advice we give in the room, because a badly chosen project is worse for us than no project.
The decision
Every software conversation should start here rather than at a demo. Work out which of these three you are actually in, because the cost of getting it wrong is not the software budget — it is a year spent on the wrong problem.
Cheapest, and more common than anyone sells
A surprising number of software problems are agreement problems. Two people own the same step, nobody owns another, and a form gets filled in twice because that is how it has always been. Automating that just makes a confused process run faster.
Signs you are here
Right answer most of the time
If a product covers most of what you need for a monthly fee, buy it. Accounting, CRM, payroll and helpdesk are mature categories with good products serving the UAE, and a subscription is a far smaller commitment than a build.
Signs you are here
Right sometimes, and worth it when it is
Custom earns its cost when the part no product handles is the part your business actually competes on, or when staff are being paid to carry data between systems that will not talk to each other.
Signs you are here
Try these first
This list costs us work, and it is still the right advice. If one of these closes your gap for a monthly fee, take it — and spend the money you saved on the part of the business that actually needs building.
| What you need | Look at | Why |
|---|---|---|
| Accounting and VAT | Zoho Books, QuickBooks, Tally | All well established with UAE users and accountants who know them. There is almost never a reason to build this. |
| CRM and sales pipeline | Zoho CRM, HubSpot, Pipedrive | Mature and inexpensive at small scale. Worth exhausting before commissioning anything custom. |
| Field service and jobs | Established field-service products | Several handle scheduling and job cards well. The gap is usually local specifics like LPO handling or contract structures. |
| E-commerce | Shopify, WooCommerce | Selling online is solved. The operation behind it — stock, allocation, delivery — is where custom work tends to be justified. |
| Internal documents and approvals | Your existing office suite | Shared drives and approval flows in tools you already pay for cover more than people expect. |
| Everything, in one system | Treat with suspicion | A product claiming to run your whole business usually does several things adequately and none of them the way you do. |
Product names are given as starting points, not recommendations or endorsements — we have no commercial relationship with any of them, and you should evaluate anything against your own requirements before committing.
On price
We will not publish a price for custom software, because a figure that does not know how many workflows you have, what it must integrate with or how messy your data is would be dishonest. What we can do is tell you exactly what moves it, so you can read anyone’s proposal properly — including ours.
Where work genuinely is productised, we do publish the price. Our website packages carry figures because the scope is fixed.
See the packages that do have prices →What actually moves the number
How many workflows are in scopeOne is a project. Five is five projects wearing a trench coat.
What it has to talk toAn integration with a system that publishes a proper API is routine. One that does not is its own small project.
The state of your existing dataReliably the most underestimated line in any proposal, including the ones we have written.
Where it has to workA desk is easy. A phone in a plant room with no signal is not.
How many kinds of userEvery distinct role needs its own screens, permissions and testing.
Before you sign anything
Including us. How a vendor answers these tells you more than a portfolio does, and the last one is the most revealing.
The answer is not always you. Ask who holds it, who controls hosting and the domain, and what you would be handed if you stopped working together tomorrow.
A proposal that describes a system rather than a first deliverable is a proposal you cannot hold anyone to.
Migration is where these projects overrun. A vendor who has not asked about the state of your existing records has not priced the hard part.
Projects stall internally more often than technically. If nobody has told you the time commitment expected from your team, the plan is incomplete.
A vendor with no answer has never turned work down, which tells you how the recommendation is being made.
What goes wrong
Without one person inside the business empowered to decide, every question waits for a meeting and the build stretches around the gaps.
Each department adds a reasonable request. None is unreasonable alone, and together they double the timeline before anything ships.
Duplicate customers, three names for one product, stock figures that drifted years ago. Always found late unless deliberately looked at early.
If the new screen is slower than the spreadsheet for the person doing the job, they will keep the spreadsheet and you will have paid for both.
None of these are technical risks, which is why choosing a developer on technical grounds alone is not enough. Ask how they handle these four, because they will decide whether the project works far more than the choice of framework will.
Common questions
Buy first, almost always. If a product covers eighty per cent of what you need for a monthly fee, that is usually a better decision than commissioning a build, and we will tell you so. Custom becomes the right answer when the twenty per cent it cannot do is the part your business competes on, or when you are paying several people to move data between three products that will not talk to each other.
Any figure given before someone understands your workflow is a guess dressed up as a quotation. What we can tell you is what moves the number: how many distinct workflows are in scope, whether it needs to integrate with systems you already run, how much existing data has to be migrated, whether it needs to work offline or on phones in the field, and how many kinds of user need their own screens. We scope and fix a price for a first phase, and treat anything after that as a separate decision you make with evidence.
A reasonable test is whether you can name the specific thing that would improve and how you would know it had. "We need a system" is not ready. "We lose two hours a day re-entering orders from email into the accounts package, and I cannot tell which customers are unprofitable" is ready. If you cannot yet describe it that precisely, the discovery conversation is worth having before any build is discussed.
Four things, in roughly this order. Nobody inside the business owns the project, so decisions stall. The scope grows because every department adds a wish. The existing data turns out to be far messier than anyone remembered. And the people who were meant to use it were never asked what would make it faster than what they do now, so they quietly carry on with the spreadsheet.
You should, and with us you do. This is worth asking any developer before you sign anything, because the answer is not always yes. Ask specifically who holds the source code, who controls the hosting and the domain, and what you would be handed if you stopped working with them tomorrow. A vendor who is vague about that is telling you something.
We build on Laravel, which is a mainstream open-source framework, so any competent PHP developer can pick the work up. That is deliberate: the value should be in the system, not in your inability to leave. Handover includes the source, documentation, credentials and a working deployment, whether or not we keep working together.
You should. A first phase aimed at one workflow that is genuinely costing you money will tell you more about whether this route is right for your business than any proposal document, and it is a far smaller thing to be wrong about. Businesses that try to digitise everything at once usually end up back on the spreadsheet.
A focused first phase is typically a matter of weeks rather than months, and the benefit should be visible almost immediately or the scope was wrong. If the first thing we build takes a quarter to deliver and then needs explaining before anyone can see the point, that is a warning sign about the project rather than a normal part of it.
If you are still not sure
Including when the answer is a product you can subscribe to this afternoon, or a conversation with your team rather than a developer. We would rather be the company that told you not to spend than the one that took the project anyway.
Keep exploring
If the guide pointed you toward building something, these show what that actually looks like in practice for businesses like yours.
If the guide points you toward building: how we scope, deliver and hand over a system, and what a first phase covers.
Read more ServiceWhere repetitive work can be automated without a full system, and where introducing AI is the wrong first move.
Read more IndustryWhat this looks like for a workshop: bookings, job cards, parts and invoicing as one connected record.
Read more IndustryQuotations, LPO tracking, dispatch and maintenance contracts for companies working on the client’s site.
Read more IndustryStock allocation, partial deliveries and credit holds for trading and distribution businesses.
Read more Next stepThe process from a business problem to a working system, and the decision points along the way.
Read more