Atlassian's migration pages say where it's going

Where does a company tell you which of its own products it plans to stop selling?

Not on the homepage. Not in the keynote. It tells you on the pages it builds to move customers off that product. Atlassian is the clearest example in B2B SaaS, because it has run the same play in public for years: pick a deployment model to retire, build a migration path out of it, and let the path do the talking.

The deadline came before the announcement felt real

Atlassian announced the end of Server sales in October 2020 and set end of support for February 2024. That is a long runway. What matters for competitive reading is not the date itself. It is the volume of supporting material that appeared around it.

Within months, the public surface filled up with migration guides, cost calculators, comparison tables, "why cloud" pages and partner directories for migration services. Each page was small. Together they said one thing: the company had already decided, and the remaining work was persuasion.

If you sold against Jira or Confluence in that period, those pages were a gift. They listed the objections Atlassian expected from its own customers. Data residency. Custom apps that do not exist in cloud. User tiers that cost more after the move. A competitor could read the FAQ and know exactly which buyers were unhappy and why.

The tooling tells you the investment level

A press release costs nothing. A migration assistant costs engineering quarters.

Atlassian ships dedicated migration assistants for Jira and Confluence, plus documentation for app vendors on how to make their marketplace apps migration-ready. That second part is the signal people skip. When a platform company writes docs for its ecosystem about a transition, it is spending partner goodwill on the move. Companies do not do that for a direction they might reverse.

The same pattern shows up on smaller surfaces:

• Free or extended cloud trials for migrating customers. The company is willing to give revenue away to remove friction.
• Loyalty or dual-licensing offers. It expects customers to hesitate, and it pays them to stop hesitating.
• Partner badges for migration specialists. It needs more hands than its own services team has.

Each of these is a line item on someone's budget. Read them as budget, not as marketing.

Pricing pages moved in the same direction

The pricing history adds the last piece. Self-managed tiers saw repeated list-price increases while cloud tiers gained features and bundles. You do not need an insider to see the slope. You need old snapshots next to new ones, which is the whole method behind Wayback pricing archaeology.

When one deployment model gets more expensive every year and the other gets more generous, the company has told you its preference. The price list is the policy.

What this means if you compete with Atlassian, or with anyone like it

The lesson travels well beyond Atlassian. Any company that runs more than one product line, deployment model or plan generation will eventually retire one. The migration surface appears before the retirement is official, and it is far more specific than any roadmap talk.

Watch for these pages on your competitors:

1. A new "migrate from X to Y" guide, especially inside their own product family.
2. Docs aimed at partners or app developers about a transition.
3. A comparison table between two of their own offerings.
4. FAQ entries that answer objections nobody on your team has heard yet.

This is the same instinct as reading a deprecations page or a changelog: the operational pages leak strategy because they have to be precise.

The hard part is noticing the first page when it appears, not the fiftieth. Seeto watches public surfaces like docs, FAQs and pricing pages continuously and reports each change as a discrete event, so a new migration guide shows up as a diff on the day it ships. It does not read the guide for you. That part is still your job.

A company can keep a strategy out of its keynote. It cannot keep it out of the help center.