What actually breaks when founders
migrate software.

Software vendors sell migration as a 30-minute task. *"Click 'export', click 'import', you're done."* Founders learn the truth on the other side: the export works, the import works, and then a hundred small things · automation logic, billing webhooks, custom fields, plan-tier features, DNS records, embedded forms, third-party integrations · quietly don't transfer.

This is the first edition of the Switch Index · a public, recurring report on what actually happens when solo and bootstrapped SaaS founders migrate software. We document each migration end-to-end, then look for the patterns that repeat.

The patterns are sharper than we expected.

The five risks that appeared in almost every migration

Across the 17 documented migrations, the same handful of failure modes kept showing up · not just occasionally, but as the norm. Here's the full breakdown:

The takeaway: the bottleneck in software migration isn't the export button. It's everything that lives around the data · the workflows, the integrations, the financial plumbing. That part is your problem, and vendors have no incentive to make it easy.

The cleanest migrations were 7-step, ~1-hour switches

Among the 17, eight were classified as Easy · typically taking under 90 minutes end-to-end with one to three named gotchas. Patterns we noticed:

• Same-category swaps tend to be cleaner than category-jumps. Mailchimp → Beehiiv (newsletter to newsletter) is meaningfully easier than Mailchimp → ConvertKit + Loops (splitting newsletter from automation).
• Modern destinations import better. Tools built in the last five years (Linear, Resend, Beehiiv, Attio) tend to ship dedicated importers for the most common origins. Older tools usually expect you to bring CSVs.
• Founder-led tools are friendlier on the way out. Of the 17, founder-led destinations had the fewest reported gotchas · partially because they actively want to win the switcher's loyalty.

The hardest migrations were the ones that touched billing

Of the 9 medium-difficulty migrations, almost every one involved either:

• Active paid subscribers who needed to be migrated without interrupting their billing (newsletter and SaaS), or
• Custom fields, automations, or pipeline configurations that had been accumulating for 18+ months and lacked documentation.

The most expensive thing in a migration isn't the time you spend on the day. It's the things you don't know you need to migrate until they break.

What this means for picking your next tool

The traditional way to compare software is to look at features, pricing, and reviews. Those questions matter. But they answer the wrong question: "is this tool worth signing up for?". The harder question · the one most founders only learn 18 months in · is "will I be able to leave when I want to?"

That's why every review on fewertools now ships with an ownership badge (founder-led / public / PE-owned / acquired) and we're rolling out a per-tool Switch Risk Score derived from this data. The goal isn't to scare you off any specific tool. It's to make the switching cost visible before you sign, not after.

Pick tools whose data you can actually leave with. Pick tools whose owners have a track record of not raising prices 3× post-acquisition. Pick the tool that fits today, but verify the migration path before you commit. That's the editorial position underneath everything we publish.

The 12 worst-case findings, by category

Here's the per-category snapshot from the dataset. (Categories with only one observation are flagged.)

• Newsletter / Email (3 migrations): billing complications and engagement-history loss most common. Mailchimp → Beehiiv and Mailchimp → ConvertKit both flagged historical data as a primary risk.
• Project management / Issue tracking (3 migrations): automation rebuild was universal. Asana → Linear, Jira → Linear, and Trello → Notion all required rebuilding view filters and custom fields.
• Website / CMS (2 migrations): Webflow → Framer and WordPress → Framer both involve DNS, SSL, and SEO redirect work. SEO impact was the longest-tail risk.
• CRM (1 migration): HubSpot → Attio surfaced custom-property mismatch as the dominant gotcha. Pipeline stages, deal cards, and email-sequence automations all needed rebuilds.
• Hosting / Backend (2 migrations): Heroku → Railway and Firebase → Supabase shared environment-variable, CI/CD, and custom-domain reconfig as the dominant tasks.
• Cloud storage (1 migration): Dropbox → Google Drive · folder permissions and shared-link rewriting were under-documented.
• Analytics (1 migration): Google Analytics → Plausible · historical data is functionally lost; the Plausible team is upfront about it but vendors moving in the other direction usually aren't.
• Automation (1 migration): Zapier → Make · every Zap had to be rebuilt from scratch; trigger semantics differ.

If you want to audit your own stack

The questions in this report are the same questions inside our free Stack Audit at fewertools.com/audit. Paste the tools you use, and you'll get an instant report showing your Stack Archetype, Switch Risk Score, ownership breakdown, named £ in waste, and exactly what to keep, cut, replace, and add.

It takes about five minutes. There's no sign-up to see the result. The audit uses the same methodology we'll publish each year as the Switch Index updates.

Cite as: Feyisitan, C. (2026). The Switch Index 2026: what actually breaks when founders migrate software. fewertools.com/switch-index. Published 2026-04-28.