Two things at once: restore a working route for clients today, and start a paper trail with the vendor that has dates on it.

Do the first one first, because it is entirely within your control and it takes ten minutes. Change your voicemail greeting if you can still reach it. Put a notice and an alternative contact on your website. Forward the line to a phone that works, even if that means temporarily losing the separation you set the line up for. Send a short written message to current clients telling them how to reach you this week. A broken line costs you very little if there is an obvious second door; it costs you a great deal if there is not.

Then, and only then, deal with the vendor.

Support failure is the most consistent complaint attached to every product in this category. Reviewers describe not being able to speak to anyone by phone, urgent flags taking hours to produce a reply, being sent a copied help article written for an interface that no longer exists, and — the one that gets repeated most — many hours across calls and emails ending in "there's nothing we can do." One describes weeks of trying just to establish whether the product would work with their practice setup at all.

Meanwhile client communications stay broken. That is what makes this different from ordinary bad service.

How to make the vendor move

Put it in writing and number the facts. What broke, the exact time it started, what you have tried, how many clients are affected. Vague urgency gets queued; specifics get escalated.

Use one channel and keep it. Opening a second ticket resets your place and splits the history. Reply on the existing thread.

Ask for a named person and a next update time. Not a fix time — an update time. It is a smaller ask and it is much harder to ignore.

Say what you will do if it is not resolved. Not as a threat. As a schedule: if this is not working by Friday I will be starting a transfer of the number. That sentence moves tickets, and you should mean it.

Keep dates. If this ends in a billing dispute or a transfer, the record of what you asked and when is the only thing that helps you.

How to test support before you buy

You cannot inspect this on a pricing page, but you can test it in an afternoon, and almost nobody does.

Email support before you sign up, with a real question. Time the reply. Read whether it answers you or points at a help article. This one test predicts more about your next two years than any feature list.

Ask what happens at 4pm on a Friday. Specifically: how do I reach a person if my line is down and clients cannot reach me. Note whether the answer is a channel or a promise.

Ask who answers. Whether the person replying works for the company, and whether they can escalate to someone who can change something.

Look for the failure stories, not the ratings. Reviews that describe an outage and how it ended are worth more than a hundred five-star entries. Search the vendor's name alongside "support" and read the worst results.

Bigger is not safer here

The intuition that a large company will be more available is exactly backwards in this category. The largest providers of second-line services are also the ones with no phone number, no named contact and a support model built around deflection. The complaints quoted above are attached to well-known products with large user bases.

What actually predicts good support is whether the people running the company are close enough to the product to fix something. That is usually a small company, and the trade you are making is a real one: fewer features, more chance of reaching a human who can act. Whether that trade is worth it depends on how much of your practice runs through the line.

The thing that makes all of this survivable

A second route in, published everywhere the number is published.

If your website, directory listings and email signature all carry an address or a booking link alongside the phone number, then an outage is an inconvenience. If the number is the only way in, an outage is a practice that has gone dark, and you are negotiating with a support queue from a position of complete dependence.

That is a five-minute change and it is worth making today, whether or not anything is currently broken.

About Rivet

Rivet is a Canadian practice line built for therapists — a separate number for calls, texts and voicemail, with video sessions and clinical tools in the same place. Your data stays in Canada, and transcription runs on Rivet's own hardware rather than a third-party AI service.

One plan, $65 CAD a month, everything included. Fourteen-day trial, no card.

See what's included · Start the trial