How much does custom software development cost in the GCC

One client received four proposals for the same project: $9,000, $22,000, $48,000 and $95,000. Every vendor was sent the same brief.

None of them was wrong, and none was trying to mislead. Four teams simply read four different things into the word "project".

This article explains what the price is made of, which ranges are realistic, and how to compare proposals against each other.

What the price is made of

Engineering time. The largest part. How many engineers for how many weeks. It is easy to calculate, and it is the part everyone compares.

Accountability. Who carries the risk on scope and deadline. Under a fixed price the vendor carries it; under time and materials you do. That difference shows up in the price, and it is not a saving — it is the risk moving to the other side of the table.

Support. What happens after handover: warranty period, response times, defect fixes. In a cheap proposal this part is often absent entirely.

Risk. A buffer for the unknown. A good team prices it in and tells you so. A weak team does not, and halfway through the project the conversation becomes "that was not in the scope".

Realistic ranges

These are market figures, not marketing ones. If someone offers an "ERP" for $5,000, it is not an ERP.

Five factors that raise the price

1. Number of integrations. Each one takes from one to three weeks. "Just connect the payment gateway" means three stages of work: reading the documentation, connecting, and handling failures. The third is the longest.

2. Data migration. It depends on the legacy format. Clean data takes a week; messy data takes three, plus the cleaning work itself.

3. Number of branches or entities. If each one runs its own process, that is separate configuration.

4. Mobile platforms. The second platform is priced separately unless both can be built from one codebase.

5. Bilingual EN and AR interfaces. Right-to-left layout is not a translation task. Plan for it from the start rather than adding it later, when it costs considerably more.

6. Compressed timelines. Halving the schedule does not halve the cost — it raises it. The more people you add, the more time goes into coordinating between them.

Fixed price or time and materials

A practical rule: if the scope can be written down, use a fixed price. If it cannot, run a discovery stage first and then fix the price. Time and materials makes sense only when you have your own technical lead and your own roadmap.

What is not in the budget

Ask about these separately when you receive a proposal. They are almost never included:

• Servers and hosting, $50-300 a month
• Third-party licences and API fees
• SMS gateway and payment processor commissions
• Domain and SSL certificates
• Developer accounts for mobile apps: Apple $99 a year, Google $25 once
• Content entry and product data upload

The sums are small, but if they are not accounted for, unplanned costs appear at the end of the project.

How to compare proposals

Comparing on price alone is the most common mistake. Level the proposals first:

Once you have answers to these six questions, the $22,000 and $48,000 proposals will look much closer together. Or the reason for the gap becomes clear.

If the budget does not stretch

This is a normal situation and it has a solution. Do not ask for a discount — reduce the scope.

A practical sequence:

1. Identify the two processes causing the most pain
2. Put only those into the first phase
3. Run them for three months and measure the result
4. Fund the next phase from the outcome of the first

This has two advantages: the budget is split into stages, and you test the team before committing serious money.

Working with a team outside the Gulf

If you are considering a team based outside the region, three points decide whether it works:

Time zone overlap. Tashkent sits one hour behind Gulf time, so the working day overlaps almost completely. That matters more than the day rate.

Local practice, not just translation. Arabic support, local payment gateways and local hiring or licensing rules have to be designed in, not bolted on afterwards.

Who stays after launch. Ask what happens in month seven. A team that also runs its own products is far more likely still to be there than one assembled for a single project.

In summary

The price covers not only code, but accountability, support and risk.

Three steps before you collect proposals:

1. Write the scope, or start with a discovery stage. A price without a scope is a guess, not a calculation
2. Level the proposals using the six questions — compare content, not price
3. Ask separately about the costs that fall outside the budget