Blog Overview
Quotes range from SAR 5,000 to SAR 150,000 because they do not measure the same thing. Here is what you actually pay for, the published bands, and the costs missing from the first proposal.
“What does a client portal cost?” is not a question about the market. It is a question about shape. A portal for two hundred clients who upload a document and sign an approval is not the same as an operational portal used daily, integrated with an ERP and ZATCA, carrying identity data.
This article breaks down what you are actually paying for, gives three scenarios at different budgets, and names the costs that are missing from the first proposal. The numbers below are Rasad’s published bands, not estimates for a project we have never seen.
The short answer
We work in three shapes. The band is what the shape costs; where your project lands inside it depends on the number of integrations and the volume of data and content. All prices are in Saudi riyals, excluding VAT, with third-party costs billed separately.
| Shape | From | Most projects | Delivery |
|---|---|---|---|
| Portal — first release | SAR 8,000 | Under SAR 20,000 | 6–12 weeks |
| Portal, integrations and launch | SAR 15,000 | Under SAR 30,000 | 12–20 weeks |
| Care and retainer | SAR 1,500 / month | Under SAR 8,000 / month | Same-day response |
The difference between SAR 8,000 and SAR 30,000 is not design quality. It is how many systems must talk to each other, how much data must be migrated, and how many decisions must be tested before a real user touches the portal. See the pricing page for the full breakdown of each shape.
Why quotes vary this much
Because each proposal measures something different. One counts screens, another counts team weeks, and a third gives a low number to open the door, then adds integrations later. Before comparing numbers, make sure you are comparing the same work.
Screens are not the work. One workflow may need five screens or fifteen; the difference is in the logic, not the layout.
Integrations are priced on assumptions. “ERP integration” can mean a nightly file import, or two-way sync across seven document types. The first is days; the second is weeks.
Legacy data. Migration is not import. The work is cleaning records, reconciling them, and testing what was lost.
Security and compliance. Permission review, audit logs, and penetration testing are not free line items. If they are absent from the proposal, they are deferred, not removed.
What you are actually paying for
In a typical portal project, the work splits into eight parts. Some appear in the proposal; some appear only when delivery starts.
Discovery: who uses the portal, what decision they make, and what happens when a form is incomplete.
UX: information order, navigation, Arabic direction, and empty and error states.
Identity and permissions: sign-in, roles, who sees which document, and what the system logs.
Workflow: request, review, approval, rejection, resubmission, and notifications.
Integrations: ZATCA, Nafath, ERP, Mada, or single sign-on. Each has its own registration and test path.
Testing: real devices, two languages, different permissions, and failure cases. Not a click on every button.
Launch: production environment, backups, monitoring, and a rollback plan.
Handover: code, database, documentation, and training. This is what makes leaving possible.
Three scenarios at different budgets
The numbers below are illustrative examples inside our published bands, not quotations. A real project needs discovery before the number is fixed.
| Scenario | What it includes | Expected band | Delivery |
|---|---|---|---|
| Document portal | Email sign-in, document upload, one approval, status dashboard, Arabic and English | SAR 8,000–20,000 | 6–12 weeks |
| Operational portal | The above + ZATCA + ERP + multiple roles + reporting | SAR 15,000–30,000 | 12–20 weeks |
| Enterprise portal | SSO + data migration + multiple integrations + security review + training | Custom scope after discovery | 20+ weeks |
Integrations: where the money goes
If you want to know why a quote is high, read the integrations section. Every added system brings discovery, data mapping, failure handling, and testing. These are the details that never appear in the budget table.
ZATCA: not a “send invoice” button
Integrating with Fatoora means issuing the invoice in the required format, stamping it, submitting it, and handling the response: accepted, rejected, or accepted with warnings. It also means credit and debit notes, and retries when the network drops. See the integrations page.
Nafath: a national identity, not a password
Nafath puts verification in the user’s app, not in your form. Integration needs registration per environment, a short acceptance window — the Nafath guide names 60 seconds — and handling for pending, rejected, and expired requests.
Mada and payments: a PCI scope, not a checkout page
Any portal that takes payments needs a licensed provider, handling for two-factor authentication, refunds, and reconciliation. If you store card data inside your own system, you expand the compliance scope significantly. Usually it is better to leave it with the provider.
From reading to shipping
Want this built, not just explained?
Bring us the problem behind the article. We will tell you what it takes, what it costs, and whether we are the right team.
ERP and CRM: the data matters more than the connection
Connecting is easy. The hard part is deciding which system owns the truth, what happens when a record changes in both, and how invoices reconcile with ledger entries. That work is what prevents contradictory reports two months later.
Single sign-on: an organisational decision before a technical one
Does the employee already have an Active Directory or Entra account? Are groups managed centrally? Who approves a new user? If those answers are unclear, the integration will be built twice: once on paper and once in code.
Costs that are missing from the proposal
The proposal usually covers the build. But a portal is a living thing: it is hosted, monitored, patched, and billed by third parties. These lines need to be in your budget from day one.
| Line | Who pays | Note |
|---|---|---|
| Hosting and database | Usually you | Depends on provider, size, and storage location |
| SMS and email | You | Billed per message or per thousand |
| E-invoicing certificate | You | Usually through an approved provider |
| Payment gateway fees | You | Percentage per transaction plus fixed fees |
| Content entry | You, or a separate scope | Hundreds of pages in two languages take time |
| Care | You | Starts after the warranty period |
| Your team’s time | You | Meetings, decisions, acceptance testing |
How to compare two quotes without being fooled
Turn each proposal into the same columns. If you cannot, the two prices are not comparable. The table below is the minimum each proposal should show.
| Question | Why it changes the price |
|---|---|
| How many workflows, not screens? | Logic and exceptions are the real work |
| Which integrations, by name, with what data? | “ERP” is not one line item |
| Who enters the legacy data? | Migration can be a third of the project |
| What is the test scope? | Devices, languages, roles, and failure cases |
| What is the support agreement? | Response and resolution are not the same |
| Who owns the code and data? | Ownership sets the cost of leaving |
| What is excluded? | The missing line returns on the final invoice |
Six ways to lower the cost without breaking the system
Start with one workflow. Ship the path with the most pain, then add the rest once it proves itself.
Use an identity you already have. If you run Microsoft Entra or Active Directory, connect it instead of building a second account system.
Buy the commodity. Payments, messaging, and storage are not places for differentiation. Do not build what you can buy with a better security posture.
Fix the content scope. The number of pages, languages, and attachments changes the price more than most clients expect.
Budget care from the start. A portal that is not maintained becomes a new project a year later.
Do not build a mobile app in release one. A responsive portal that works on a phone may be enough until the need is proven.
When is an off-the-shelf platform cheaper?
If your workflow is standard — a generic client portal, tickets, documents, simple invoices — an off-the-shelf platform may be the smarter call. You pay a subscription and reduce the build. A custom build wins when the workflow itself is the competitive advantage, when regulatory integrations are heavy, or when you need an Arabic experience the platform does not offer.
Do not make the decision ideological. Compare three years of subscription with three years of build and care, then add the exit cost of each option. The comparison page breaks down that decision.
What changes the number after discovery
Even the best estimate remains an estimate. These are the items we re-price after discovery, and we say so before the contract, not after.
The number of systems to sync, not just the number of APIs.
The quality and volume of legacy data.
The number of roles and permissions, and whether record-level access is required.
The need for Nafath, Mada, or an e-invoicing certificate, and each one’s approval path.
Whether the content is ready or needs writing and Arabic review.
The bottom line
A portal is not one cost. It is a build, then integrations, then care. If you want a number you can defend, ask for a written scope before the price, and read what will not be built as carefully as what will.
Sources
- Rasad’s published price bands — the only source for the numbers in this article.
- ZATCA: integration-phase requirements.
- Nafath user guide — the 60-second acceptance window.
- PDPL Implementing Regulation.
