What happens to your booking links if we shut down

Your booking link is in email signatures you cannot edit, calendar invites already sent, a printed QR code, and other people's CRMs. It is one of the few pieces of infrastructure a small business hands out to strangers and then forgets about — which makes it exactly the wrong thing to have hosted by a company that might quietly stop existing.

This is not hypothetical. People who paid for scheduling tools have watched Probooking announce a shutdown date, MeetFox disappear after an acquisition, and Harmonizely become SimplyMeet with the lifetime deal they had bought quietly reinterpreted. The most quoted line on the subject is about a business paying for a bespoke booking widget: “what he's paying for is not being able to leave.”

What we have committed to

Clause 7 of our covenant, in full, because a paraphrase of a legal commitment is a different commitment:

Three things about that clause are worth pulling out, because the interesting part of any promise is its trigger and its enforcement rather than its sentiment.

• The trigger is broader than bankruptcy. It covers shutting down, being acquired and discontinued, and simply ceasing to maintain the product. The middle one is the common death and the one most promises omit.
• It is in the terms, not in marketing copy. It is written into the contract as a binding commitment — an open-source licence with a trigger — rather than a statement of intent on a page we could quietly reword.
• The covenant is versioned and public. Every change is announced, dated and explained, so the promise you are relying on is a document with a history rather than a claim you have to take on trust today.

The failure this does not cover

Now the part that matters more, because you have almost certainly seen it happen more recently than you have seen a scheduling tool die.

In April 2026 Cal.com took its commercial repository closed-source. The surviving public fork is real but strips the team features — no teams, no round-robin, no routing, no SSO — so for the audience most likely to have chosen it on the strength of “open source”, the escape hatch closed. The reaction named it as a pattern rather than an incident: “They use ‘open source’ to market and build up their product, take investment, get greedy, and then go closed source.”

The honest defence is not a better promise. It is that a promise should be worth less to you than something you can verify today — so the rest of this post is the part that does not depend on our future conduct at all.

What you can verify right now

Your data comes out, in one click, while everything is fine

Export is not a wind-down feature we would build if the worst happened; it is in the product now and you should use it this week, before you need it. It produces your profile and handles, every event type, every booking with its attendees, your calendar connections and your billing record — in an open, documented format, free, including after you cancel.

The test of an export is not whether it exists but whether it is complete, so ours is guarded by a test that fails when a new database column is added without either exporting it or recording, in writing, why it is excluded. That is how the guest-attendee column added this month ended up in the export on the day it shipped rather than a year later. What comes out, in detail.

Your link does not get handed to a stranger

This is the failure people do not think about until it happens. You leave, your handle goes back in the pool, someone else takes it, and every signature and QR code still in the world now points at their booking page.

When an account is deleted, its handle is quarantined for 90days rather than released — long enough for signatures, calendar invites and printed codes in the wild to stop being live traffic to a stranger. The record that enforces it holds a one-way hash of the address rather than the address itself, so it is a “do not silently recreate, do not silently reassign” marker rather than retained personal data. During that window the page can serve a forwarding link you chose.

Leaving is a button, and we publish the way out

Covenant §5 commits to self-serve cancellation with no retention gauntlet, and to publishing migration guides to our competitors. That last part is the one you can check without trusting us: the guide for moving off us exists and is linked from the help centre, written while we would prefer you stayed.

What to actually do about vendor risk

Not just with us. This is the checklist we would want a friend to run on any tool that holds a public URL of theirs.

1. Export now, not later. A company in trouble is the worst time to discover its export is broken or gated.
2. Find out what happens to your URL when you leave. Reassigned immediately, quarantined, or dead? Most vendors do not say, and the answer determines whether leaving breaks things for people who are not you.
3. Check whether the promise is in the contract. A commitment on a marketing page can be edited on a Tuesday. One in the terms cannot, quietly.
4. Prefer a link you control. A redirect on your own domain pointing at whatever scheduler you currently use means changing vendors is a DNS edit rather than a hunt through every signature you have ever written. We do not offer custom domains, so this is advice that routes around us as much as anyone.