Ask hosting providers why customers leave and the answers are consistent. In the WebPros/CloudLinux Web Hosting Trends Report 2026, 56% of surveyed providers named price as the main reason customers leave, and 41% said customers move to SaaS website platforms.
The standard reading is churn: the customer walked.
Domain-level infrastructure data suggests a different, and more useful, interpretation. In most cases the customer didn't walk. The wallet fragmented.
A business's online presence used to be one bundle at one provider: website, email, DNS, security, tools. That bundle has visibly come apart:
Looking across 35M+ confirmed-active business domains (domains with verified operating email, live content or platform provisioning; parking, placeholders and dead pages excluded — a domain classification, not a company count):
| observation | share | reading |
|---|---|---|
| Business domains carrying operating email | ~29M of 35M+ | email remains the near-universal trust anchor |
| … whose email runs on a third-party workspace suite (not the hosting provider) | ~22% | a fifth of the email wallet already left the bundle |
| Business domains whose website runs on a SaaS builder platform | 4.2M+ | the front door increasingly lives outside traditional hosting |
Shares are of confirmed-active business domains, European-deep and globally observed, as of the snapshot date. Method and denominators: see how a number is built.
The provider often still holds the domain, the DNS, sometimes the mailbox. What it lost is not the customer — it's the surrounding software spend: analytics, marketing and social tools, security services, collaboration apps. The infrastructure relationship survived. The wallet went sideways.
Front doors migrate quickly. Trust migrates much more slowly.
Customers experiment constantly with what visitors see: website builders, landing-page tools, CDNs, marketing stacks. Moving the underlying trust layer — email, data, backups, the hosting relationship itself — happens far less often.
The clearest evidence sits in the domains whose front door did move. Among the 4.2M+ confirmed-active business domains served on a SaaS website builder:
The website moved. The email relationship — the layer businesses genuinely fear breaking — either went to a workspace suite or stayed exactly where it was. The builder captured the front door and almost none of the trust stack. That asymmetry is why "41% move to SaaS platforms" overstates what was actually lost: often it's one layer of the stack, not the customer.
If the diagnosis were churn, you'd expect hosting groups to respond with retention pricing. What they're actually doing is more telling: buying the surrounding stack. In HostingBrain's validated, source-linked acquisition ledger, adjacent deals — SaaS, cloud, website tools — were near-zero through 2023, then overtook core hosting and domain acquisitions in 2025 (see the year-by-year chart, refreshed weekly). The industry's own capital allocation says fragmented wallet share, not churn.
For operators: the strategic question is not "how do we stop customers leaving" but "how do we capture more of the surrounding software stack while we still own the trust relationship." The trust layer is the distribution advantage — and it erodes slowly, which is an opportunity with a time limit.
For investors and advisors: raw churn metrics undervalue books where the trust layer is intact, and overvalue "sticky" books where the wallet already fragmented. The split between infrastructure retention and wallet retention is measurable — and they diverge.
Reproduce this analysis: the stack_profile tool returns this exact cross-layer cut — globally, per market, or per operator — and it is available on the free tier.
Ask the follow-up yourself. HostingBrain answers questions like this — with the date, denominator and caveats attached — inside Claude and any MCP-compatible assistant.