Drive, WhatsApp, or a client portal: what your studio runs on

Every architect who has run a practice for more than three years has had the same fortnight. Two projects hit working drawings at once, a client asks for something you're fairly sure you sent in April, and you spend a Saturday going through an inbox to work out what you actually issued. Somewhere in that Saturday you decide the studio needs a system.

Then Monday arrives and the decision goes nowhere, because every option looks like a month of setup for a benefit you can't quite name. This is an attempt to name it, including the case for changing nothing.

When Drive and WhatsApp are genuinely fine

If you're one or two people with two or three live projects, the free stack isn't a compromise. It's the correct answer, and adopting software instead would be a straightforward mistake.

The reason is that the whole system fits in one person's head. You know what you sent because you sent it. You know which revision is current because you drew it last Tuesday. There's no coordination problem, so a tool that solves coordination problems is solving something you don't have, in exchange for a monthly fee and an afternoon of setup.

Spend that afternoon on the things that scale with you instead: a numbering convention, a title block that carries status properly, and a drawing register in a spreadsheet. Those habits will be worth more at ten people than any software you could buy today, and they cost nothing.

The four setups

There are four realistic arrangements. Compare them on what each is actually good at rather than on features.

Most studios end up in the second row, and it's worth understanding why it disappoints. General project tools are built around tasks: something needs doing, someone is assigned, it gets closed. But architecture practice isn't primarily a task problem. It's a document-and-money problem. The drawing is the unit, the fee is what releases it, and the client is a party who should see their own project and nothing else. You can bend a board of cards into that shape. You'll be maintaining the bend forever.

The checklist

Whatever you're evaluating, including staying where you are, these are the questions that separate tools still working in year three from tools you abandon in month four.

1. Does each client get a space that is only theirs? Not a folder you remember not to share widely, but genuine separation. Ask whether it is enforced below the interface or is simply a filter the software could get wrong. Workspace isolation that lives in the database is a different guarantee from one that lives in a screen.
2. Can a file be held until an invoice is paid? This is the single feature most tools do not have and most practices most need. Without it, your fee schedule is aspirational. With it, issuing on payment is the default rather than an act of willpower.
3. Is there a record of who saw what, and when? You will want this exactly once, and by then it is too late to start collecting it. An audit log is unglamorous until the afternoon it settles a dispute.
4. Where does the money settle? Straight into your own account, or into a platform wallet that pays out later on someone else's schedule? Settling to your own account means the tool never holds your fee, which is a materially different relationship.
5. Does it take your formats without conversion? If drawings have to be flattened or exported to be usable, the tool will be worked around within a month.
6. Can links expire and be revoked? Sharing with a consultant should not mean creating something permanent. See sending drawings without losing control.
7. What happens to your data if you leave? Ask before you join, not after. A tool that cannot export your records cleanly is a tool with a hostage.
8. How long to onboard one client? If it is more than a few minutes, you will stop doing it properly by the fourth client, and a system used inconsistently is worse than none.

Build versus buy

Every second architect has a cousin who writes software, and the thought is reasonable enough: this isn't complicated, it's a folder, a login and a payment link.

It isn't complicated to build badly, which is the trap. The visible part, uploading drawings and showing them to a client, is a weekend. The parts that decide whether the thing is safe to run a practice on are not: tenant isolation that holds against a hand-crafted request, files served through short-lived signed links rather than public URLs, payment status settled by signature-verified webhooks rather than by whatever the browser reports on the way back from checkout, and idempotent handling so a duplicate event never grants access twice. That list is a fair description of what a payment integration has to get right before anyone can trust it with a fee.

Get the webhook handling wrong and a client who abandons a payment can end up with the drawings anyway. Get the isolation wrong and one client can read another's project. Neither failure announces itself. You find out from someone else.

Build if the differentiator is truly yours and you intend to maintain it for years. Buy if what you want is for the thing to work on a Tuesday while you do architecture. Most practices, honestly assessed, are in the second group.

Switching without losing a season

What keeps most studios on a setup they've outgrown isn't loyalty. It's the migration, which looks like a month. Done in the right order it's closer to a week, because you don't move everything.

So which row

Solo with two projects: row one, and spend the saved money on better habits. If the problem is internal coordination across a team and fees aren't an issue, row two fits. Large multi-consultant jobs with formal RFI processes: row three exists for you and earns its price.

Row four is for the studio whose actual problem is the one this article has been circling: drawings going out before the fee comes in, no clean record of what was issued to whom, and a different arrangement for every client. That's a narrow description, and if it doesn't fit you then it doesn't fit you.

It's the description AtelierLab is built against. A private workspace per client holding the drawings, the invoices and the thread. Sets that stay sealed until the payment clears and then release themselves. Payments that settle into your own Razorpay account, never a wallet in between. If the fee problem isn't your problem, one of the other three rows will serve you better, and we'd rather say so.