breme.

Case study · Measuring experience · Xero

Xero removed 160,000 idle subscriptions: what counts as a real customer

By BREME · · 4 minute read

In short

In the half to September 2024, Xero removed 160,000 long idle subscriptions: ones undeployed for more than 24 months and not expected to be deployed. Xero said the removal had minimal effect on revenue, and it now reports churn and growth with those subscriptions taken out. For customer success teams, it is a public example of a simple question: does your customer count measure people using the product, or just people paying for it?

Who this is for: Customer success, revenue operations and finance leads at software and subscription companies who report on customers, churn and account health.

What Xero did

Xero, the accounting software company listed on the ASX, announced in its first half of FY24 that it intended to remove long idle subscriptions. Its H1 FY25 market release, dated 14 November 2024, says the program was carried out in that half with 160,000 subscriptions removed.

Rask Media's report on the result quotes Xero's definition: subscriptions "undeployed for more than 24 months and were not expected to be deployed in a reasonable timeframe." Xero described them as low value and largely in its international segment.

Xero's release says these subscriptions "were a small percentage of Xero's overall subscribers and their removal had minimal effect on revenue." In other words, 160,000 entries in the count were not doing much for customers or for the business.

What it did to the numbers

The removal made the headline numbers look worse before it made them more useful. In H1 FY25 Xero reported 26,000 net subscriber additions, down 87% on the same half a year earlier. Excluding the removed subscriptions, it added 186,000. Total subscribers still grew 6% to 4.19 million.

Average revenue per user rose 15% to $43.08 (NZD), and Xero notes this figure includes the impact of the removal. Fewer low value subscriptions in the count means a higher average.

For the full year to 31 March 2025, Xero reported 254,000 net additions including the removal, and 414,000 excluding it. It reported average monthly churn of 1.03% on the basis that excludes the removal, and calculated total lifetime value of subscribers, $17.9 billion, after the removal.

By FY26, reported on 14 May 2026, Xero had moved to the word customers. It reported 4.92 million customers, 506,000 added in the year, and average monthly churn of 1.14%, which it said remained below pre-pandemic trends. A footnote explains that FY25 lifetime value uses churn excluding the removal because it "reflects churn dynamics present in Xero's go-forward customer base."

The customer success lesson: separate paying from using

Xero's public documents do not say how these subscriptions came to be idle, and we will not guess. What the definition tells us is enough: a subscription can sit on the books for two years without being deployed. Every software company has some version of this. Seats bought and never assigned. Workspaces set up in a trial week and never opened again. Accounts paid on a card nobody checks.

When those sit inside the customer count, three things go wrong for a customer success team:

  • Health scores look better than reality, because inactive accounts are not complaining.
  • Churn arrives as a surprise, because the customer stopped using the product long before they stopped paying.
  • Onboarding looks finished when it never started, because the account was created and the box was ticked.

What to measure instead

Xero's word is deployed. Your product will have its own version: the first set of real data connected, the first workflow run by the customer's own team, the first report someone opens without being asked. In our experience the most useful health signal is the one that proves the customer has started getting value, not the one that proves they signed up.

Practically, that means building the customer count in two layers. The first is everyone with an active subscription. The second is everyone who has reached first value and kept using the product in the last period. The gap between the two is where customer success time should go.

  • Define your deployed moment in plain words and agree it with product and finance.
  • Report paying and deployed customers side by side, every month.
  • Give every paying but undeployed account an owner and a date.
  • Decide in advance what happens at 6, 12 and 24 months undeployed, so the conversation is planned, not awkward.

Honest numbers make better renewals

There is a renewal argument here as well. An account that has paid for months without using the product is not a loyal customer; it is a cancellation that has not happened yet, and often a complaint about value when it does. Reaching out early, helping them start or helping them downgrade, costs a little revenue and protects trust.

Xero showed that a large company can take a visible hit to a headline growth number in exchange for a count that reflects real use, and explain it clearly to the market. Most software companies only need to do that inside their own dashboards.

If you want help defining the moment your customers really start, and building onboarding around it, our customer success page sets out how we work: breme.com.au/customer-success/.

Sources

See your journey from the customer's side

We map the end-to-end journey with you, find the moments that matter most, and build a strategy your teams and systems can deliver.

Start a conversation More insights