Drift Got Switched Off. Now What?

The Email Nobody Wanted To Open

If you were a Drift customer, you already know the one. It landed with a subject line about "an important update to your account" and every single person who read it immediately thought the same thing. Oh no. Here we go.

Drift is done. Clari and Salesloft switched it off on March 6, 2026. Customers got somewhere between 60 and 90 days to pack their things and leave. For a tool that spent years telling everyone that conversational marketing was the future, the ending was pretty quiet.

And look, I am not here to dance on anybody grave. Drift was genuinely good at what it did. The playbook builder was clever. The routing was solid. Plenty of teams booked real meetings off that widget and the product deserved better than a sunset notice.

But here we are. So let us talk about what you actually do now, because most of the advice floating around right now is garbage.

Why Most Of The Advice You Are Reading Is Bad

Go search for Drift alternatives right now. I will wait.

Notice anything? Every single result points you at another enterprise tool. Qualified. Intercom. Some AI SDR platform with a demo request form and no pricing page. The whole first page assumes you were paying Drift enterprise money and want to keep paying somebody enterprise money.

Here is the thing though. By the end, Drift minimum contracts were sitting around 30 thousand dollars a year. Most of the people scrambling right now were not on those plans. They were on old grandfathered pricing from years ago, the cheap tiers that got quietly deprioritised, the plans that nobody at Drift wanted to talk about anymore.

Those people are the ones getting ignored. A three person startup that used Drift to catch questions on a pricing page does not need a platform with an implementation consultant. They need a chat widget that works.

Be Honest About What You Actually Used

Before you go shopping, do this. Open your old Drift account while you still can and look at what actually ran.

I have done this exercise with a few teams now and the result is almost always the same. Somebody built fourteen playbooks over two years. Two of them fired regularly. One of them was the plain "chat with us" bubble. The other was a message that popped up on the pricing page after about thirty seconds.

That is it. That is what 30 grand a year was buying.

The other twelve playbooks were built during an enthusiastic Tuesday afternoon, tested twice, and then forgotten. Some of them had not fired in eight months. One of them was targeting a campaign that ended in 2024.

So write down the two or three things that genuinely worked. Those are your actual requirements. Everything else is nostalgia.

Do The Export Before You Do Anything Else

Seriously, stop reading and go do this if you have not already.

When a product sunsets, the data access goes away faster than anyone expects. Export your conversation transcripts. Export your contacts. Export anything with an email address attached to it. Save it somewhere boring like a Google Drive folder called drift-archive.

And set your expectations correctly here. No chat tool on earth is going to import your Drift conversation history in a nice clean way. That is not a thing. What you are saving is a reference archive so that when a customer says "I talked to someone about this in January" you can actually go look it up.

Treat it like a shoebox of old receipts. You are not going to reorganise it. You just want to still have it.

The Real Question Is Where Replies Land

Okay so now the fun part. What do you replace it with?

Most people frame this as a feature comparison. Which tool has playbooks, which has routing, which has an AI thing. I think that is the wrong question, and I think it is why so many teams end up with a second dead widget eighteen months later.

The right question is much dumber. When a visitor types a message on your site at 4pm on a Thursday, where does that message physically show up, and will a human being see it?

Because the honest failure mode of every chat tool ever built is not missing features. It is a dashboard nobody opens. You install the thing, you are excited for a week, and then the tab gets closed during a laptop restart and never comes back. Two months later you find eleven unread conversations and one of them was somebody trying to give you money.

I have done that. It is a genuinely awful feeling.

Why Some Teams Are Routing Chat Into Telegram

So here is the approach that has been quietly picking up steam, and yes, I build a tool that does this, so weigh that accordingly.

Instead of putting messages in a dashboard, you put them in an app you already check compulsively. For a lot of small teams that app is Telegram. Visitor types on your website, the message lands in a Telegram group, whoever is around replies in the thread, and the reply shows up in the widget on the site.

No new tab. No new login. No new habit to build. Your phone buzzes the same way it buzzes when your friend sends you a meme, and you answer while standing in a queue at the post office.

Is it right for everyone? No. If you have a twenty person support team with shift schedules and SLA reporting, you want a real helpdesk with a real queue. I am not going to pretend otherwise. But if you are between one and eight people and your actual problem is "messages get missed," moving the inbox to a place you already live solves the problem better than any feature list.

Rebuilding The Two Playbooks That Mattered

Remember those two playbooks you identified earlier? Good news. They are usually easy to rebuild.

The classic Drift setup that actually drove pipeline was some version of this: if a visitor is on a high intent page, and they have been there long enough to be reading rather than bouncing, say something.

That is a proactive trigger and most modern widgets have some version of it. You set a condition like "URL contains /pricing" plus "time on page is more than 30 seconds" and you attach a message. Something human. Not "Hi there! I noticed you were browsing our pricing page!" because everyone knows that is a robot and it makes people close the tab.

Try something closer to how a shop assistant talks. "Any questions on the plans? Happy to just tell you which one fits." Short. Unbothered. Easy to ignore, which is exactly why people answer it.

Here is a rough mapping of the Drift concepts to what you should look for:

What Drift called it	What you are looking for now
Playbook	Proactive trigger with URL and time conditions
Routing rules	A shared group or thread per conversation
Drift Intel	Visitor info panel showing page, referrer, country
Chatbot qualification	Pre chat form or an AI agent with your docs
Meeting booking	A calendar link pasted in the reply, honestly

That last one annoys people but I stand by it. Native calendar booking inside chat is nice. It is not thirty thousand dollars nice.

The Budget Reality Check

Let us do some quick maths because this is where the decision usually gets made anyway.

Drift enterprise minimum was in the region of 30 thousand a year. Intercom will run you 29 dollars per seat per month on the entry plan, and then Fin AI charges around 99 cents for every resolved conversation on top. Five seats plus a thousand AI answers a month puts you well past 1,100 dollars monthly before anyone has done anything clever.

Meanwhile a flat rate widget with unlimited seats and AI included sits somewhere between 12 and 30 dollars a month. Same visitor. Same question. Wildly different invoice.

I am not saying cheap always wins. Sometimes you genuinely need the expensive thing. But you should at least know the size of the gap before you sign, because the gap here is roughly the cost of a part time employee.

The Lesson Worth Keeping From All This

Here is what I actually took away from watching Drift go dark, and it has very little to do with chat widgets.

Your customer conversations should not live somewhere that can be switched off by a company you have never met. Drift got acquired, then acquired again, then sunset. Meta killed the Messenger chat plugin in 2024 and stranded something like half a million businesses overnight. This keeps happening and it will keep happening.

So when you pick the replacement, ask one extra question. If this vendor disappeared tomorrow, what would I still have?

If the answer is "a CSV export and a phone number," you are in reasonable shape. If the answer is "nothing, because the entire relationship lived inside their platform," maybe pick differently.

Telegram groups are nice here for a boring reason. The conversation history sits in a chat app that you and your team already own, on an account nobody can revoke, that has survived a lot longer than most SaaS products. Even if you stopped using my tool tomorrow, the messages are still in your Telegram.

That is not a sales pitch so much as a suggestion about where to keep things.

What To Do This Week

Short list. No fluff.

• Export everything from Drift today if you have not. Access windows close early.
• Write down the two playbooks that actually fired. Ignore the rest.
• Pick a replacement based on where replies land, not on the feature grid.
• Rebuild only those two triggers. You can always add more later.
• Put a real human notification somewhere you cannot ignore. Phone, group chat, whatever works.
• Set a calendar reminder for six weeks out to check whether anyone is actually replying.

What The First Two Weeks Actually Feel Like

Nobody writes about this part and it is the part that decides whether the migration sticks.

Week one is fine. You are paying attention, the novelty is there, you reply to everything within a minute and feel enormously competent. This is not real. Week one is always fine, on every tool, including the one you are leaving.

Week two is where you find the gaps. Somebody messages while you are in a meeting and you discover your notification settings are wrong. A colleague replies to a conversation you were already handling and the visitor gets two different answers. Someone asks a question that needs an account lookup and you realise you have no way to see who they are.

None of these are disasters, but they are the things to actively watch for rather than assume away. Write down each one as it happens. At the end of the fortnight you will have a short list of genuine gaps, which is far more useful than any feature comparison you could have read beforehand.

And be prepared for volume to look different. Teams switching from a dashboard tool to a phone based one often see message volume go up, not because more people are visiting but because you are finally answering fast enough that conversations continue instead of dying. That looks like extra work. It is actually the tool doing its job.

Resist Rebuilding Everything On Day One

The temptation during a migration is to recreate your old setup exactly, feature for feature, before you go live. Fight it.

You just established that most of your Drift playbooks never fired. If you rebuild all fourteen in the new tool, you have carried the clutter across and learned nothing. Start with the bubble and one trigger. Live with that for a fortnight. Add things only when you feel their absence.

This sounds like laziness dressed up as strategy but there is a real reason for it. Every automation you add is something that can misfire, annoy a visitor, or quietly break without telling you. A setup with two moving parts that you understand completely beats a setup with twenty that you configured once and never audited.

The teams I have seen do this well ended up running a much simpler operation than they had on Drift, and were slightly embarrassed about how little they missed.

That last one matters more than the tool choice. Every chat widget works on day one. The question is whether it still works in October.

Drift is gone. That genuinely stinks for the people who liked it. But if the replacement ends up being simpler, cheaper, and lands in a place you actually look at, this might be one of those rare migrations that leaves you better off.