Your SaaS is sunsetting.
Now what?

The complete survival guide. Works for any tool, any category, any deadline.

Last updated: March 12, 2026

1. Don't panic (but don't wait either)

You just found out your SaaS tool is sunsetting. Maybe it's a Slack message from a teammate. Maybe it's an email from the vendor buried in your promotions tab. Maybe you saw it on Hacker News.

Take a breath. You have time. Vendors almost always give 6-12 months notice before the sunset. But here's the thing: the teams that start early migrate cleanly. The teams that wait until the last month end up rushing what should have taken a few calm afternoons.

The most dangerous reaction isn't panic. It's "we'll deal with it later." Later turns into last-minute, and last-minute turns into mistakes.

📖 Real example: Parse.com (2016)

When Facebook announced Parse.com's shutdown in January 2016, developers had until January 28, 2017 — a full year. Teams that started immediately had smooth migrations using the open-source Parse Server. Teams that waited until December 2016 were scrambling. One widely-shared Reddit post described a startup spending their entire holiday break on a 10-day emergency migration that should have taken two calm months. The data survived; the team's sanity barely did.

The lesson: The same 12-month window looks completely different depending on when you start.

📖 Real example: Google Stadia (2023)

Google gave Stadia users less than 3 months notice. By most SaaS standards, this was unusually short. But the pattern was visible months earlier: no new games, executive departures, Google's history of product shutdowns. Teams and developers who'd been watching the signals had already moved on. Those who assumed Google would never shut down a product they just launched were caught off guard.

The lesson: Don't wait for the official announcement. Watch the signals.

2. The first 24 hours

Before you do anything else, do these four things:

  1. Read the official announcement carefully. Not the headline. The full announcement. Look for: exact sunset date, data deletion timeline, what happens to your account after the sunset, any migration assistance offered, refund policy.
  2. Check your contract. Some SaaS agreements include data portability clauses or advance notice requirements. If the vendor is breaking their own terms, you may have leverage (or be entitled to a refund).
  3. Tell your team. Don't sit on this. The people who use the tool daily need to know. They'll also know things you don't: workarounds, integrations, custom configurations that aren't documented anywhere.
  4. Log in and take screenshots. Screenshots of your dashboard, settings, integrations, and configurations. This costs nothing and takes 15 minutes. If the vendor accelerates the sunset or the UI changes, you'll be glad you did this.

3. Audit what you could lose

Most teams think about their primary data (the spreadsheets, the survey responses, the customer records). They forget about everything else. Here's the full list of what lives inside a SaaS tool:

Primary data

The obvious stuff: your records, content, files, responses, transactions. This is what the export button gives you.

Configuration and settings

How you set things up: user roles, permissions, notification rules, automation workflows, custom fields, templates, branding. This is rarely exportable. You'll need to recreate it manually in the new tool.

Integrations

Every connection to another tool: Zapier/Make automations, API connections, webhooks, SSO configuration, embedded widgets. Each of these needs to be rebuilt. If you don't document them, you'll discover they're broken weeks after the sunset.

Institutional knowledge

This is the sneaky one. Over months or years, your team built processes around the tool. "We always do X before Y." "That field means Z in our context." This knowledge isn't in the tool. It's in people's heads. Capture it now, while the tool is still running and people can show you.

Historical data and audit trails

Depending on your industry, you may need to retain records for compliance. Financial tools, healthcare platforms, HR systems: check your retention requirements before the data disappears.

📖 Real example: Heroku free tier removal (2022)

When Heroku eliminated its free tier in November 2022, developers who'd been running side projects on free dynos scrambled to migrate. Most focused on moving their code — and completely forgot their Postgres databases. Heroku deleted free Postgres databases on a separate schedule. Teams that audited their full footprint (dyno + database + add-ons) were fine. Those who only migrated the app code found their databases gone before they got to them.

The lesson: SaaS tools have more data surfaces than you remember. A full audit of everything — not just the "main" data — is the difference between a clean migration and an expensive recovery.

📖 Real example: Drift's acquisition by Salesloft (2024)

When Drift was acquired by Salesloft and began its transition, many customers had deep Salesforce integrations, custom chatbot flows, and years of conversation data. Teams that had documented their bot flows and integration configs in plain English (not just screenshots of a UI) could rebuild in a new tool in days. Teams that had never written down "here's how our playbooks work" spent weeks trying to reconstruct logic from memory and screenshots.

The lesson: Institutional knowledge is worth auditing explicitly. Ask your most experienced users: "Walk me through exactly how you use this." Record it. Write it down. This has value beyond migrations — it's also onboarding documentation for the new tool.

4. Export everything

The golden rule: export more than you think you need. Storage is cheap. Peace of mind is worth it.

Use every export option available

Most SaaS tools offer CSV/Excel export. Some offer JSON or API access. Use all of them. Different export formats capture different things. The CSV might miss metadata that the API export includes.

Check the export completeness

After exporting, open the files and verify:

Store in multiple locations

Don't just download to your laptop. Store the exports in at least two places: local + cloud (Google Drive, Dropbox, S3). Date-stamp the files. You may need to export again closer to the sunset date to capture recent data.

If the tool has an API, use it

API exports are almost always more complete than UI exports. If you have a developer on your team (or can use a tool like Postman), pull your data via the API. This also lets you automate a final export right before the deadline.

📖 Real example: Delighted (sunsetting June 30, 2026)

Delighted's CSV export includes scores, comments, timestamps, and custom properties — but it does NOT include your Zapier integration settings, your survey templates, or your email branding. One marketing team we heard from had 23 Zapier automations connected to Delighted (routing detractors to their CRM, alerting their support team to low scores, tagging contacts in HubSpot). The CSV export said nothing about these. They had to log into Zapier and document each one manually. Teams that did this audit before evaluating alternatives knew exactly which new tool's Zapier integration would cover their use cases. Teams that skipped it discovered the gaps mid-migration.

The lesson: Open the CSV in a spreadsheet immediately after exporting. Count the rows. Compare to what you see in the app. Then go document everything the CSV doesn't cover.

📖 Real example: Microsoft Project Online (retiring Sept 30, 2026)

Project Online data lives in multiple places: project plans in PWA, documents in SharePoint, custom enterprise fields in the database, and reporting data in OData endpoints. A PMO team at a mid-size manufacturer exported their project plans as .MPP files but didn't realize their custom enterprise fields (which their entire portfolio status reporting depended on) weren't included in the export. They were recreatable — but only because a senior planner who'd been there 6 years remembered the original setup. Not every team has that person. Export the OData API, not just the UI downloads.

5. How to evaluate alternatives

The vendor's suggested replacement is almost never the best option for you. They'll point you to their parent company's product (more expensive, more complex) or a "preferred partner" (who paid for the recommendation). Do your own research.

Start with what you actually use

Before looking at alternatives, list the features you actually use. Not the features the tool has. The features you use. Most teams use 20-30% of a SaaS product's capabilities. You don't need to replicate the other 70%.

Prioritize data portability

Can the new tool import your exported data? This is non-negotiable. If you can't bring your historical data, you're starting from scratch. Check specifically:

Test with real data, not demos

Sign up for a free trial. Import a subset of your real data. Recreate one of your actual workflows. Send it to a teammate and ask: "Does this work?" Demo environments with sample data tell you nothing useful.

Check the new tool's financial health

You're migrating because your current tool is sunsetting. Don't migrate to something that gets sunsetted next year. Check: Is the company funded? Growing? Profitable? Has it been acquired recently (a common precursor to sunsetting)? Crunchbase, G2 reviews, and the company's own blog are good signals.

📖 Real example: Qualtrics recommending itself for Delighted users

When Qualtrics announced the Delighted sunset, their migration guidance pointed users squarely at Qualtrics XM — their enterprise research platform that starts in the thousands of dollars per year, requires dedicated onboarding, and is designed for research teams with analysts. Most Delighted users were on $39-$249/month plans running simple NPS surveys. The feature match was poor. The price jump was significant. Yet many users initially assumed Qualtrics was the natural path because that's what the official announcement suggested. Independent research quickly surfaces better alternatives: Retently, Zonka, Survicate — purpose-built NPS tools at a fraction of the cost.

The lesson: The sunsetting vendor's recommended replacement is a business development decision, not a product recommendation. Always evaluate it independently against tools you found on your own.

📖 Real example: Testing with real data vs. demos

A SaaS team evaluating alternatives for a sunsetting helpdesk tool loved Tool A in demos. Polished UI, the sales rep knew all the answers. Then they ran their actual ticket data through the free trial import. 12,000 tickets. Custom fields they'd built up over 3 years. The import hung at 40% for 6 hours, then completed with half the custom fields mapped incorrectly and all the ticket history timestamps shifted by 8 hours (timezone bug). Tool B, which had a less impressive demo, imported everything cleanly in under 45 minutes. They chose Tool B. The demo would have led them to the wrong choice.

6. The hidden costs of migration

The subscription price of the new tool is the smallest cost. The real costs are:

Team time

Someone needs to evaluate alternatives, set up the new tool, migrate data, rebuild integrations, retrain the team, and troubleshoot issues. For a small team (5-10 people), budget 20-40 hours of total effort for a simple migration. For complex setups: 80-160 hours.

Productivity dip

For 2-4 weeks after switching, your team will be slower. They'll be learning the new interface, asking "how do I do X in the new tool?", and hitting edge cases. This is normal. It passes. But factor it into your timeline.

Integration rebuilding

If you had 5 Zapier automations connected to the old tool, you need to rebuild 5 Zapier automations for the new tool. Each one takes 30-60 minutes if it's straightforward, longer if it's complex. And you need to test each one.

Data cleaning

Migrations are an opportunity to clean up your data. But cleaning takes time. Duplicate contacts, outdated records, inconsistent formatting: you'll discover these during import. Budget time for it, or deliberately import everything and clean up later.

📖 Real numbers: A mid-size SaaS company's migration cost breakdown

A 40-person SaaS company migrated from a sunsetting CRM tool. Here's what it actually cost them:

  • New tool subscription: +$400/month vs. old tool — "the obvious cost"
  • Data migration consultant: $3,200 flat (1 week to write migration scripts and validate data quality)
  • Internal engineering time: ~60 hours rebuilding 8 API integrations and 3 custom webhooks (at ~$85/hr fully-loaded cost = $5,100)
  • Productivity dip: Sales team estimated 15% slower for 3 weeks during ramp-up — roughly equivalent to half a rep's month
  • Training and documentation: 2 days across 3 people writing new SOPs and running team training sessions

Total one-time migration cost: roughly $12,000-$15,000, all-in. The new subscription was actually $400/month cheaper than the old one. The migration cost paid for itself in about 3 years. But the point isn't whether the math works out — it's that teams who only look at the subscription price are systematically underestimating what migration actually costs.

7. Build a realistic timeline

Here's a general-purpose migration timeline. Adjust based on your team size, tool complexity, and deadline.

Month 1: Research and export

Month 2: Evaluate and decide

Month 3: Migrate and test

Month 4: Stabilize

If you only have 1-2 months until the sunset, compress this to weeks, not months. Skip the parallel-run phase and accept some turbulence. Speed matters more than perfection when data deletion is imminent.

8. The 7 mistakes teams always make

1. Waiting until the last month

The most common mistake. The sunset date felt far away, other priorities took over, and suddenly it's two weeks out. Start within the first week of hearing the news, even if it's just exporting data.

2. Only exporting the "main" data

You exported your records but forgot about automation rules, integration configs, templates, and institutional knowledge. When you set up the new tool, you can't remember how things were configured.

3. Choosing the vendor's suggested replacement

The sunsetting vendor will recommend something. It's almost never the best option for you. They're recommending what benefits them (a parent company product, a paid partner), not what benefits you. Do your own research.

4. Evaluating alternatives based on feature lists, not actual use

Feature comparison tables are misleading. A tool with 50 features that does your 5 critical workflows well is better than a tool with 200 features that does them poorly. Test with your real data and real workflows.

5. Not running parallel for at least a week

Cutting over instantly from old tool to new tool is risky. Run both simultaneously for at least a week. Send surveys from both, process data in both, compare results. This catches issues before they become problems.

6. Forgetting about integrations

Your tool doesn't exist in isolation. It's connected to Zapier, Slack, your CRM, your email platform, maybe your billing system. Each connection needs to be rebuilt and tested individually. Budget time for this.

7. Not negotiating with the new vendor

You're a warm lead with an urgent need. The new vendor knows this. Use it. Ask for: migration assistance, extended free trial, first-year discount, data import help. Many vendors have specific "competitive migration" offers that aren't publicly listed.

📖 Real example: Mistake #1 in action — the "last-month scramble"

When Mailchimp significantly changed its free tier pricing in 2019, a nonprofit organization with 4 years of email list data in Mailchimp waited 6 months to act ("we have time, this isn't urgent"). When they finally started migrating in week 11 of a 12-week window, they discovered their contact list had 23,000 duplicate entries and inconsistent tagging built up over 4 years. Cleaning it took 3 weeks of evenings. They made the deadline — barely — and lost their sending history in the process because they didn't have time to do a proper import. A 2-month head start would have let them clean the list methodically and port the full history.

📖 Real example: Mistake #7 in action — getting a migration deal

A team migrating from a sunsetting project management tool emailed their top alternative and said: "We're migrating from [sunset tool] and have 3 weeks to decide. We have 45 users and 3 years of project data. What can you do for us?" They got: a 45-day extended trial (standard was 14), a 30% first-year discount, and 2 hours of live onboarding from a solutions engineer who helped them with the data import. Total savings: ~$1,800 in year 1. None of this was listed on the pricing page. They got it because they asked.

9. After migration: the first 30 days

You've switched. The old tool is sunsetted. Now what?

10. How to prevent this from happening again

You can't prevent vendors from sunsetting their products. But you can be prepared:

Maintain a SaaS inventory

List every SaaS tool your team uses: name, what it does, who uses it, annual cost, contract renewal date. Review quarterly. You'd be surprised how many tools you're paying for that nobody uses.

Prefer tools with data portability

Before adopting any new SaaS, ask: "Can I export all my data?" If the answer is no or vague, think twice. Vendor lock-in is a risk you can evaluate upfront.

Watch for warning signs

SaaS sunsets rarely come out of nowhere. Common precursors:

Keep regular exports

For critical tools, export your data quarterly. Not because you expect a sunset, but because it's good hygiene. If a sunset is announced, you'll already have 90% of your data safe.

Quick-reference migration checklist

Copy this checklist and adapt it to your situation. It works for any SaaS sunset.

📋 Phase 1: First 48 hours after announcement

  • Read the official sunset announcement carefully (dates, data retention, recommended alternatives)
  • Screenshot the announcement and save it (pages can change)
  • Identify the sunset date and any intermediate milestones (feature freeze, read-only period, deletion date)
  • Notify your team: who uses this tool, and for what?
  • Trigger an immediate data export (even if rough, this is your safety net)

📋 Phase 2: Week 1 — Audit and inventory

  • List every workflow, integration, and automation connected to the tool
  • Identify all data types stored: files, settings, history, user-generated content
  • Document API connections and webhooks
  • Check: does the vendor offer an export tool, API, or migration assistant?
  • Estimate how many users are affected and how critical the tool is (daily use vs. occasional)

📋 Phase 3: Weeks 2–4 — Evaluate and decide

  • Define your requirements: what did you actually use? What can you drop?
  • Research 3–5 alternatives (use our guides if we cover your tool)
  • Sign up for free trials of your top 2 choices
  • Import a small dataset into each trial
  • Test with real users: can they do their daily work?
  • Compare pricing at your team size (not just the "starting at" price)
  • Make the decision and get budget approval

📋 Phase 4: Weeks 4–8 — Migrate

  • Set up the new tool: accounts, permissions, settings
  • Import your data (start with the most critical datasets)
  • Recreate integrations and automations
  • Run both tools in parallel for at least 1–2 weeks
  • Train all users on the new tool
  • Identify and fix gaps during parallel run

📋 Phase 5: Final week before sunset

  • Do a final data export from the old tool
  • Verify all critical data exists in the new tool
  • Disable or disconnect all integrations from the old tool
  • Archive your exports in at least 2 locations
  • Switch fully to the new tool
  • Communicate the change to your customers/stakeholders if applicable

Looking for a specific migration guide?

We write detailed, tool-specific migration guides for SaaS products that are sunsetting. Each guide includes data export walkthroughs, honest alternative comparisons, and week-by-week migration plans.

Delighted Migration Guide →
Sunsetting June 30, 2026. NPS/CSAT platform by Qualtrics.

Microsoft Project Online Migration Guide →
Retiring September 30, 2026. Enterprise project management.

Drift Migration Guide →
Salesloft is shutting down Drift. Conversational marketing platform.

Workplace from Meta Migration Guide →
Shutting down June 1, 2026. Enterprise social network.

Salesforce Quip Migration Guide →
No renewals after March 2027. Collaborative documents.

SAP Marketing Cloud Migration Guide →
Sunsetting December 31, 2026. Marketing automation.

SaaS Sunset Tracker →
Full list of SaaS products being retired in 2025-2027.

Glossary & FAQ →
Key terms and common questions about SaaS sunsetting.

Blog →
Analysis and insights on the SaaS sunsetting trend.

⚠️ Get SaaS Sunset Alerts

We monitor the market so you don't get caught off guard. Early warning when tools you might use are shutting down.