Drift Is Being Sunset. Here's Your 30-Minute Exit Plan

TL;DR: In March 2026, Salesloft announced the gradual sunset of Drift, the product that invented conversational marketing, and began referring its customers to a third-party successor most of them had never evaluated. If you're running Drift, nothing you did caused this, but the clock is now running on your website's most important conversion channel. The good news: migrating to a modern AI sales agent is no longer a quarter-long project. This guide covers what the sunset actually means, what to salvage from your Drift setup, what to demand from whatever you move to, and how Aijent gets you from "sunset notice" to "live AI agent booking real meetings" in about half an hour, from A$199 a month with no contract.

On 5 March 2026, the company that owns Drift announced its "gradual sunset" and began referring existing Drift customers to a successor product from a different company entirely.

Sit with that sentence for a moment, because if you're a Drift customer it contains everything you need to know about your position. The product that invented conversational marketing, the one that taught the entire B2B world that websites could talk, is being wound down by its owner, and the recommended path forward is a vendor you never chose, never evaluated, and never signed with.

You did nothing wrong. You bought the category creator, which in 2021 was the smartest move on the board. Then the board changed: Drift was acquired, the acquirer merged again, and somewhere in a strategy meeting you weren't invited to, your website's front door became a legacy asset.

Here's the thing about sunsets, though. They're predictable, they're visible, and if you move early, they're an upgrade opportunity wearing a crisis costume. This is your exit plan.

What "gradual sunset" actually means for you

Let's decode the language, because "gradual sunset" is doing a lot of soft work.

There is no published end-of-life date. The public product page still markets the product. Meanwhile the official announcement refers existing clients elsewhere. That combination tells you three things.

First, investment in the product has functionally ended. Nobody builds meaningful new capability into a product they've announced the sunset of. Every week you stay, the gap between your chat experience and the state of the art widens.

Second, the timeline is theirs, not yours. "Gradual" means "until it isn't." Companies in wind-down accelerate when the maintenance cost outweighs the remaining revenue, and you will get less notice of the final date than you'd like.

Third, and most importantly: your migration is happening either way. The only question is whether it happens on your schedule, calmly, with time to test, or on theirs.

The audit: what you're actually losing, and what you're not

Before you migrate anything, take an honest inventory of what your Drift deployment actually does today, because for most customers it's less than the invoice suggests.

Drift was built in the playbook era. Its conversations are, at their core, rule-based flows: predefined paths, button choices, routing logic, with a layer of generative AI added later on top of the older architecture. If your deployment is like most, you have a greeting playbook, a pricing playbook, a demo-request playbook, and a calendar integration that shows visitors a rep's availability.

Here's the liberating part: none of that is precious. The playbooks encode two things, your qualifying questions and your routing rules, and both of those live in your head and your CRM, not in the tool. What you're actually migrating is about a page of institutional knowledge: who you sell to, what you ask them, who gets a meeting, and where the data goes.

Write that page down. It's the only artefact from your Drift era worth keeping, and it will take you twenty minutes.

What to demand from whatever comes next

This is where the crisis becomes the upgrade, because the market has moved a long way since your Drift contract started, and the sunset is your permission slip to skip a generation. Whatever you evaluate next, hold it to these five standards.

One: real conversation, not playbook theatre. The next agent should hold an open conversation powered by an actual language model, answering from your real website content, not walking visitors through a flowchart you have to build and maintain. If the setup process involves designing conversation paths, you're buying the past again.

Two: it books the meeting itself. The single biggest conversion leak in the old model was the handoff: the moment the visitor is ready and the tool shows them a calendar widget or shunts them to a scheduling link. The modern standard is booking inside the conversation: a real Zoom meeting created, a calendar invite sent, confirmation delivered in the same chat window. Demand to see this happen live before you sign anything.

Three: your CRM is the system of record, automatically. Contact created, deal created, the full conversation transcript attached, on every booking, without a human touching it. Not "integration available." Automatic.

Four: you set the standards for who gets through. The old world captured everyone and let your sales team sort the pile. The new standard is a definable gate: you specify who qualifies for a meeting, who should get a callback from your team, and who gets a polite no. Your calendar is a finite resource. The next tool should treat it like one.

Five: pricing you can read. After what just happened to the category, buy nothing demo-gated, nothing with an annual lock-in, and nothing whose costs float with usage meters. You've just lived through what happens when your vendor's business model has more moving parts than your deployment. Flat, published, monthly. Accept nothing less.

The 30-minute migration

Here's what moving to Aijent actually involves, timed honestly.

Minutes one to five: sign up and let it read. Start the free trial. During signup, Aijent automatically crawls your website and starts learning your business: your products, your pricing, your language. This is not a setup task. It happens while you're choosing your agent's name.

Minutes five to fifteen: transfer the page of knowledge. Remember the page you wrote in the audit? This is where it goes. Tell your agent who you sell to and how you want it to behave. Connect your calendar (Google or Microsoft), your Zoom, and your CRM (HubSpot, Salesforce, Pipedrive or Zoho) with one-click OAuth connections. No tickets, no consultant.

Minutes fifteen to twenty: swap the tag. Remove Drift's snippet from your site. Paste Aijent's single script tag in its place. If you deployed Drift through a tag manager, this step is a two-minute change.

Minutes twenty to thirty: test like a stranger. Open your site in a private window and behave like your best prospect. Ask a hard product question. Give qualifying answers. Book the meeting. Watch the Zoom invite arrive and the contact, deal and transcript land in your CRM. Then behave like a tyre-kicker and watch the difference.

That's the migration. The thing your last chat platform charged an onboarding fee and a project plan for is now a lunch break.

What you'll stop paying for

The old model's costs came in layers: the platform subscription, the annual commitment, the onboarding engagement, and the dedicated human. Drift famously worked best with a dedicated conversational marketing manager tending the playbooks, and independent analyses consistently budgeted weeks of internal effort for a proper rollout.

The replacement column looks like this: A$199 a month, no contract, no onboarding fee, no playbooks to garden, and a 14-day free trial, no credit card required that runs beside your existing setup so you can compare pipelines before you commit to anything. Your conversational marketing manager, if you have one, gets promoted from flowchart maintenance to strategy: deciding who qualifies, reviewing transcripts, tuning the standards. The machine does the nights and weekends.

The printable migration checklist

For the operators who want this as a checklist rather than prose, here is the complete exit, in order, with owners. Total elapsed time for a standard deployment: one working session.

Before the switch (owner: whoever owns the website funnel)

1. Export your Drift conversation history and contact data while access is guaranteed. Wind-downs are not famous for generous data-export windows late in the process; do this first, this week, regardless of what you migrate to.
2. Write the one-page brief: who you sell to, your three qualifying questions, your routing rules (who gets which meetings), and where the data must land in the CRM.
3. Inventory every page carrying the Drift snippet, and note whether it's hard-coded or in a tag manager.
4. Screenshot your current monthly numbers: conversations started, emails captured, meetings booked. This is your baseline for the side-by-side.

The switch (one afternoon) 5. Start the Aijent trial; let the auto-crawl read your site while you complete the wizard. 6. Paste the one-page brief into your agent's behaviour instructions. On the Growth plan and above, set your SureGate door policy too: who's a GO, who's a REVIEW for callback, who gets the polite no. 7. Connect calendar, Zoom and CRM (one-click authorisations, no tickets). 8. Swap the snippet. If you're cautious, run both for a week on different pages; the tag manager makes this trivial.

After the switch (week one) 9. Test as a stranger: one buyer journey end-to-end (watch the Zoom invite land and the CRM record appear), one tyre-kicker journey (watch the polite no). 10. Read every transcript for the first few days and tighten your door policy sentences. This is the new "playbook maintenance," except it's five minutes and it's in English. 11. Compare week-one numbers against the baseline. Meetings booked is the metric that matters; everything else is commentary. 12. Cancel the sunsetting contract on your own timeline, in writing, with your data already exported. Exits are calm when step one happened first.

What about the successor they're referring you to?

A fair question deserves a fair answer: Salesloft is referring Drift customers to a named successor product, and you should evaluate it if it suits you. We'd just offer three neutral tests to run on any recommended successor, ours included.

First, whose choice was it? A migration path chosen by your outgoing vendor optimises for their transition, not necessarily your outcome. You're already being forced to move; the one silver lining of a forced move is that the whole market is suddenly your shortlist, not just the name in the sunset memo.

Second, does it fix the reason Drift aged, or preserve it? Drift's architecture was playbooks with AI added later. If the successor's setup process still involves designing conversation flows, you're migrating the maintenance burden along with the widget, and you'll do this exercise again in three years.

Third, what does the booking moment look like? Run the test that matters: pose as a qualified buyer and see whether the meeting gets created inside the conversation, with a real video link and a calendar invite, or whether you're handed a scheduling page at the finish line. It's the single sharpest question in the category, and it takes ninety seconds to answer on anyone's product, which is exactly why we keep suggesting it.

The five demands, restated as a scorecard

Take the five standards from earlier and score any candidate out of five, one point each: real conversation trained on your content (not designed flows); books the meeting in-conversation (real video link, invite sent, no handoff); CRM writes automatic on every booking (contact, deal, full transcript); a definable gate for who gets through (your rules, changeable by you, with a human-callback tier); pricing you can read (published, flat, monthly, no contract).

Aijent scores five, and we publish the receipts for each point on our comparison page, alongside the rows where competitors beat us, because credibility is earned by concession. Whatever you choose, don't accept fewer than four, and never accept a zero on the booking question. The booking question is the job.

The bigger lesson from the sunset

One more thing worth taking with you, because it should shape every software decision you make in this category from now on.

Drift didn't fail as a product. It succeeded so thoroughly that it proved a market, got acquired, and got strategically retired by an owner with a different roadmap. The same consolidation has now swallowed nearly every major name in the space: the category leader belongs to Salesforce, and the biggest AI support platform is on its way there too. When you buy from a consolidating category, you're not just evaluating the product. You're evaluating whose strategy deck your renewal lives in.

Aijent is independent, self-serve, and built specifically for the businesses the enterprise roll-up left behind. Our roadmap answers to our customers, because there's nobody else in the building.

Start before the clock does

If you're a Drift customer, you have something most software buyers never get: advance notice. The sunset is announced, the successor is someone else's choice, and the migration is thirty minutes.

Run the trial beside your existing setup. Same traffic, same month, two pipelines. Count the meetings. Then archive the past on your own schedule, not theirs.

Start your free 14-day trial, no credit card required at aijency.ai. Bring the one page that matters. Leave the playbooks.