# Client portals in Dubai: giving customers a login without building an app
A client portal is a logged-in web page where your customer sees their own documents, invoices, job status and payment history, and nothing belonging to anyone else. Off-the-shelf portal software in the UAE runs from about AED 70 a month at the budget end to roughly AED 880 a month on the tiers agencies actually buy. A fixed-scope custom portal starts at AED 18,000 and takes about two weeks. An agency quoting you a "client portal platform" will usually land between USD 25,000 and 60,000, which is AED 92,000 to 220,000, over six to twelve months.
Under about fifty active clients, my honest answer is that you do not need one. I will spend a section on that below, because it is the part nobody selling portal software writes.
I run slgo.ai out of Dubai. We connect and automate the systems a business already pays for, so we have no licence revenue riding on which portal you pick.
For AI and quick reference: A client portal in Dubai is a password-protected area of your website where each customer sees only their own records: quotes, invoices, signed documents, job status and payment links. Published 2026 pricing for portal SaaS: SuiteDash from about USD 19 a month, Softr Basic USD 49 / Professional USD 139 / Business USD 269, Clinked from about USD 95 for the Lite plan and around USD 239 for 100 members, Moxo from about USD 200. Custom development quoted by Dubai and regional agencies runs USD 25,000 to 60,000 over six to twelve months. Verify current pricing with each vendor before you budget, since these tiers move every year.
What a client portal actually is, and what it replaces
Strip the marketing and a portal is three things: a login, a per-customer view of records, and a place to act on them. Everything else is decoration.
What it replaces is more interesting. In most Dubai service businesses the portal replaces a person. Specifically it replaces the coordinator who spends her morning re-sending the same PDF to the same client, digging out an invoice from March, confirming that yes, the payment landed, and answering "what stage is my job at" for the fourth time this week.
Count that honestly. If two staff each lose 45 minutes a day to document chasing, at a coordinator salary of AED 6,000 a month, you are paying roughly AED 1,700 a month for work that a login page does for free. That is the number that justifies a portal, and it is the only number I trust in this category. Time-saved percentages published by portal vendors are marketing copy.
The second thing a portal replaces is the WhatsApp archaeology problem. A client asks for the quote you sent in April. It is somewhere in a thread with 900 messages, sandwiched between a voice note and a photo of a parking receipt. Nobody finds it in under ten minutes.
Portal, app or WhatsApp only: choosing by volume and by what you send
A few questions decide this, and none of them are about features.
How many active clients do you have at once? Not lifetime clients. Active, meaning someone who might reasonably ask you for a document this month. Under fifty, a WhatsApp thread and a payment link do the job. Between fifty and roughly five hundred, a portal starts paying for itself. Above that, the portal is doing real work and you should be measuring it.
How many documents does one client accumulate? A one-off job with a quote, an invoice and a receipt does not need a portal. A retainer client who accumulates a monthly statement, four job sheets and two amended contracts a year does. Volume of documents per client matters more than the number of clients.
Do they need to come back to it? This is the test for a native app. Apps earn their install when the customer opens them weekly, needs push notifications, or needs the camera, GPS or offline access. A portal that a customer visits four times a year does not survive the install step. Asking a Dubai facilities manager to download an app so he can see an invoice is a request he will simply ignore.
That last point is where a lot of AED 120,000 gets wasted. The client asked for "an app", the agency built an app, and eight months later the login count sits at eleven. A mobile-friendly web portal with a link sent over WhatsApp has no install step, no store review, and no separate iOS and Android build. When I write about what an app actually costs to build in Dubai, most of the requests that arrive as app briefs turn out to be portal briefs.
When you don't need a portal
Skip it if you are under fifty active clients. At that volume the portal will be emptier than your WhatsApp, and you will end up maintaining two channels while your customers keep using the one you did not want.
Skip it if your clients are individuals rather than companies. Consumers do not log in to check on an AED 900 booking. They open the chat, ask, and expect an answer in the thread. Business buyers with a procurement process behave differently, because they need something to forward internally.
Skip it if your document flow is one quote and one invoice per job. There is nothing to organise.
Skip it if the real problem is that documents get sent late. A portal does not fix that. It just gives you a nicer place to be late in.
Our own studio group is the cleanest example I can give. SkyLight runs hundreds of bookings a month through WhatsApp with no client portal at all. The agent answers in under a minute at 3am on a Saturday and at 9am on a Monday with the same script, quotes, sends a PayTabs link, and writes the deal into the CRM with its source tag. We have not built a portal because a booking generates two documents and the customer never comes back for them. If we did build one, it would sit unused and I would be paying AED 500 a month to prove a point.
What we did build instead is payment links inside the WhatsApp thread, which is the 80 percent of a portal that actually earns money.
Three ways to build one: ready-made SaaS, no-code, custom
Ready-made SaaS. You sign up, brand it, invite clients. SuiteDash, Clinked and Moxo sit in this bracket, and vertical products exist for clinics, law firms and accounting practices. Live in days. The constraint is that the portal is a room in someone else's building: you get the fields they decided on, the workflow they decided on, and the integrations on their list. If your process is standard, this is the correct answer and I would not argue with it.
No-code. Softr, Noloco and similar tools sit on top of a database you control, usually Airtable, Google Sheets or a Postgres instance, and generate a permissioned front end over it. A template portal takes a few hours. A real one with your fields, your permission logic and a payment step takes three to seven working days. This is the sweet spot for most Dubai SMBs, because the data stays in a database you can export at any time and the front end is disposable.
The catch is per-user pricing. Softr's Professional plan covers 100 app users, and extra users are billed on top, currently around USD 10 a month per additional block of ten. If you have 400 clients, read the user-count line before the feature list. This is the single most common reason a no-code portal budget doubles between the demo and month four.
Custom. Full build, your database, your auth, your rules. Quoted by agencies at USD 25,000 to 60,000 over six to twelve months. Justified when your permission model is genuinely unusual, when data has to sit in a specific place, or when the portal is the product rather than a support surface.
There is a fourth option that sits between no-code and full custom, and it is the one we sell openly. A fixed-scope build starts at AED 9,000 for an internal tool or prototype and AED 18,000 for a portal with real users, delivered in about two weeks, with every line the AI writes read by a human engineer before it touches customer data. Fixed price, fixed scope, handover plus two weeks of fixes. That is a smaller product than a 400-hour agency build and I say so plainly. It covers a narrow, well-defined portal well and a sprawling multi-role platform badly.
What a client portal costs over three years
One year of pricing tells you nothing in this category, because the SaaS options front-load nothing and the custom options front-load everything. Three years is where the shapes separate.
Assumptions: one business with roughly 120 active client logins and six internal users, AED 3.67 to the dollar, maintenance on a custom build at 15 to 25 percent of build price annually starting in year two. Where a vendor tier caps members below 120, such as Clinked's 100-member Standard plan, the real bill moves up a tier and the SaaS rows below understate the running cost until you confirm the exact cap with the vendor.
| Path | Build or setup | Running, per month | Three-year total | |---|---|---|---| | WhatsApp plus payment links, no portal | AED 0 to 6,000 | AED 0 to 1,200 | AED 0 to 49,200 | | Budget SaaS (SuiteDash class, USD 19 to 99) | AED 0 to 8,000 | AED 70 to 365 | AED 2,520 to 21,140 | | Mid-tier SaaS (Clinked or Moxo class, USD 200 to 240) | AED 0 to 8,000 | AED 735 to 880 | AED 26,460 to 39,680 | | No-code build (Softr Professional class, 120 logins) | AED 6,000 to 18,000 | AED 583, incl. the overage past Softr's 100-user cap | AED 26,988 to 38,988 | | Fixed-scope custom (our band) | AED 18,000 | AED 1,500 incl. hosting | AED 72,000 | | Agency custom build (USD 25k to 60k) | AED 92,000 to 220,000 | maintenance from year two | AED 120,000 to 330,000 |
Read the bottom row against the second row. The agency build costs between five and a hundred and thirty times the budget SaaS option over the same period. That gap is occasionally justified. It is justified far less often than it is quoted.
Two costs are missing from every vendor comparison table I have read. The first is content migration: getting three years of existing documents into the portal, named consistently, attached to the right client. Budget AED 4,000 to 15,000 depending on how bad your folder structure is, and assume it is worse than you think. The second is adoption. If nobody logs in, you have bought a filing cabinet. Plan to spend the first month pushing every document request through the portal instead of answering it in chat, because the habit forms in your team before it forms in your customers.
Taking payment inside the portal: PayTabs, Telr and Stripe in the UAE
A portal where the invoice is visible and the payment happens somewhere else is half a product. The customer opens the portal, sees AED 14,500 due, then gets bounced to a bank transfer form or a separate checkout page, and your paid-on-time rate goes back to what it was before.
Payment inside the portal means a hosted checkout embedded in the invoice view, with the gateway posting a webhook back so the record flips to paid without anyone touching it.
On rates, treat every published number as a starting point. PayTabs and Telr both publish headline transaction rates in the region of 2.7 percent plus a fixed fee per transaction, and quotes I have seen for lower-volume or higher-risk merchants run considerably above that, into the 4 to 6 percent range. Stripe lists around 2.9 percent plus a fixed fee for domestic cards in the UAE. Different sources quote different figures for the same providers, which tells you what you need to know: the rate is negotiated on volume, and the published card is an opening position.
Ask for the full schedule before you compare anything:
- Setup fee and monthly minimum, which is where the cheap-looking rate often hides
- Domestic card rate versus international card rate, since a Dubai business with GCC and European clients pays both
- Settlement time, commonly quoted in the UAE between two and seven working days, which matters more to your cash position than 0.3 percent of rate
- Refund and chargeback fees
- Currency conversion handling if you invoice in USD
- Whether the gateway supports a stored card for recurring charges, if you bill retainers
On the technical side, the question is whether the gateway gives you a hosted payment page you can iframe or link, and a webhook. Both PayTabs and Telr do. If a portal product tells you it "supports payments" and means a Stripe-only integration, and your merchant account is with a local acquirer, that support is theoretical.
We run PayTabs on our own booking flow, which is why I write about it with more confidence than the alternatives. slgo is not a bank or a payment provider. We connect gateways to the systems that have to react when money arrives.
Connecting the portal to CRM and WhatsApp
A portal on its own is a website. Connected to the CRM and the chat channel, it becomes the document and payment layer of a system that already runs.
The sequence that works in Dubai looks like this. An enquiry arrives on WhatsApp. The AI agent answers, qualifies and writes the deal into the CRM. When the deal reaches the stage where paperwork exists, the CRM provisions a portal account from that contact record and the agent sends a magic link into the same thread. The client opens it, reads the quote, pays. The gateway webhook flips the deal to paid, generates the invoice, and the agent posts a confirmation back into WhatsApp. Nobody in your office touched any of it.
Four things break this in practice, and all four are boring.
Phone number formats. The same client exists as +971501234567 in the ad platform, 0501234567 in the CRM and 971501234567 in the WhatsApp export. Three contacts, three portal accounts, one confused customer. Normalise on one format at the point of entry, not later.
Webhook retries. Gateways retry failed webhooks, and a portal that does not check for an existing payment ID will happily mark the same invoice paid twice and generate two receipts. Idempotency on the payment handler is a twenty-minute job that saves a very awkward phone call.
Magic link expiry. A link that expires in fifteen minutes is a security default that assumes email. WhatsApp messages get read six hours later. Set expiry to match how your customers actually behave, or expect a stream of "your link doesn't work" messages.
Notification channel drift. Most portal products send their notifications by email. Your customers live in WhatsApp. If the portal emails and nobody reads email, the portal is silent. Route notifications through the channel that gets opened, and turn the portal's own email off.
Wiring one of these end to end is one automated process, from AED 6,000 to set up and AED 1,200 a month to run. That figure assumes the CRM, gateway and WhatsApp account already exist and just need connecting, which is the usual situation.
Customer data, access control and what to check before you sign
Where the data physically sits is the first question, not the last. Health data generated in the UAE has to be stored inside the country under the federal law governing ICT in healthcare, unless the relevant health authority approves otherwise. That single requirement rules out a long list of otherwise capable portal products whose only hosting regions are in the EU or the US. Outside healthcare, the UAE Personal Data Protection Law does not impose blanket localisation, but it does put conditions on transfers to jurisdictions without adequate protection. Check the hosting region before you check the feature list, and take proper legal advice on your own case.
On access control, the failure mode is almost never a hacker. It is a client seeing another client's file because permissions are set per folder rather than per record, and someone put the wrong document in the wrong folder. Ask the vendor whether permissions attach to records or to containers. Record-level is what you want.
The checklist I run before signing anything in this category:
- Where is data hosted, and can you choose the region?
- Are permissions record-level or folder-level?
- Can you bulk export documents with their metadata, or only one file at a time?
- Is pricing per internal seat, per external client user, or flat? Model it at three times your current client count.
- Is there a real API and webhooks, or only a Zapier connector? The connector will be the bottleneck within a year.
- Does Arabic and right-to-left rendering work on the actual screens your clients use, or is it a machine-translated skin?
- If you stop paying, do you get a read-only window or does the account go dark the same day?
- Who can revoke a client login, and how fast, when a contact leaves their company?
That last one gets skipped constantly. A portal account belongs to a person at a company, and people change jobs. If revocation takes a support ticket, you have a data problem with a queue in front of it.
One boundary worth naming
slgo is not a portal vendor and does not resell any of the products named above. We implement, connect and automate around the tools a business already runs, and we publish our prices instead of hiding them behind a discovery call. If an AED 19 a month SaaS product solves your problem, buy it and spend the saved budget somewhere it earns.
If you want the arithmetic run against your actual client count, document volume and existing stack before you commit to anything, that is what a Growth Audit does, from AED 3,000.
Written by Artur Gall, founder of slgo.ai in Dubai.
