Tuesday morning, between nine and eleven, the phone at a garage rings six times. Four of those calls are the same question: is my car ready yet? Each one costs three minutes, because whoever picks up has to walk to the workshop, find the mechanic and walk back. Twelve minutes a morning. Nearly an hour a week. Nobody writes it down, so nobody notices.
That is the moment someone says the words client portal. And that is usually one step too big.
A portal is just a page that knows who you are
A client portal is a private part of your website where a customer logs in and sees only their own things: quotes, invoices, the status of their job, documents. It sounds useful, and it is. But the full version is large: accounts, passwords, invitations, permissions, "I've lost my login", plus an admin screen for you. Weeks of work, and then somebody has to maintain it.
The question to answer first is simpler: how many things does a customer actually want to see? For a garage, a print shop or a clinic, it is usually exactly one. Where does my job stand right now.
One page per job, with a link nobody guesses
Instead of accounts, give every job its own page at a long, random address — the same trick parcel carriers use for track-and-trace. You send that link once, by text or email, when the job comes in. The customer keeps the message.
No logins, no forgotten passwords, no help desk. Practical rule: put the same information on it that you would happily say over the phone. That keeps the whole thing simple.
What belongs on that page
- Where it stands now, in your customer's words. "Waiting for a part", not "status 3".
- When it was last updated, with date and time. An old timestamp is honest. A page that looks fresh but isn't costs you trust.
- What happens next, and roughly when. "Part expected Thursday, then half a day's work."
- One thing the customer can do: reply, call, approve a quote.
- Your phone number, plainly visible. The page exists to reduce calls, not to forbid them.
The real cost is not building it, it is updating it
If nobody changes the status, the page says something untrue. And a page that lies is worse than no page at all. So decide before you build: who moves the status, and at what moment?
The best answer is that it happens as a side effect of something you already do. The mechanic already ticks "part ordered" in the workshop system — the page simply reads that. If your team has to open a second programme to keep the customer informed, the page will be out of date by Thursday.
Our rule of thumb: an update should cost at most two taps, on a phone, with dirty hands.
When logins really are worth it
There are situations where a proper portal earns its keep. When a customer has several jobs running at once and wants to look back through their history. When more than one person on the customer's side needs access. When documents go back and forth for approval or signature. And when money is involved: invoices, payment status, outstanding amounts.
But you will build a much better portal after six months of status links. By then you know exactly which four screens people use — and which ten you can skip.
What to do this week
- Count the calls for two days with a tally on paper. How many are "how's it going?"
- Write down the statuses you genuinely use. If you cannot name them in five minutes, you do not yet have a process to automate.
- Do it by hand for a week: at five o'clock, send a short text to everyone with an open job. If the calls drop, it is worth building.
If you would like to think out loud about whether status links are enough for your business or a real portal makes sense, that is exactly the kind of conversation we enjoy at ITSysPro. Half an hour on the phone costs nothing.