What actually predicts renewal

Hi Everyone,

A customer who doesn't renew usually made that decision early in the relationship. The cancellation happened months later, but the answer was already formed.

This matters for what you measure. A lot of customer health dashboards only show risk late in the year, when there is not much time to act on it.

Today, we're covering four things worth tracking early, and three that tell you less than you think.

Why early matters

The first 90 days of a customer relationship have the biggest impact on renewal. Poor onboarding, slow time to value, and weak early engagement are the usual causes.

So what you measure in that 90 day window is what predicts renewal. Anything measured later will confirm what already "happened", but won't change it.

Four things to track

1. Did they get value quickly?

How quickly does the customer get a concrete result from what you sold them? For software, this is the first successful use case. For services, it is the first delivered piece of work that the customer can act on.

This matters more than almost any other measure. Customers who reach a first real outcome quickly stay much longer. Measure the outcome itself. An attended kickoff call is not progress if the work has not started.

2. How deeply are they using what they bought?

Customers who engage with only part of what they bought leave more often than those who use more of it. For a product, this means the number of features used and how heavily each one is used. For a service, it means how many of the agreed workstreams are active.

Score heavy use separately from light use. Real work completed inside your product matters more than time spent opening it.

3. Is the champion still there?

When the person inside the customer's company who bought your product or service leaves that company, about half of those accounts don't renew for another term. When the person leaving is a senior sponsor, the number is closer to two-thirds.

Track this actively. A bounced email or a LinkedIn title change is often the first indication someone has moved on. Confirm the departure quickly and work to meet their replacement before the renewal date.

4. How much of what they bought are they actually using?

If you sold 50 seats and only 12 people log in, or if your retainer covers services they never request, the customer is already questioning the value.

When the customer is paying for things they do not use, renewal will be a difficult conversation.

Three that tell you less than you think

NPS: A score asking how likely someone is to recommend you tells you how the person felt on the day they filled it in. It is a weak predictor of whether they will actually renew.

Use it as a sentiment measure if you want, but not for renewal forecasting.

Raw activity counts: How often someone logs in, or how many meetings they attend, tells you the account is still active. It does not tell you whether the customer is getting value from that time.

Total usage: A single usage number across a whole customer account does not give you enough detail to act on. Different users in the same account can be in very different places. Break usage down by user or by workstream.

When your team sees what the dashboard doesn't

Some customers who look fine by every measure still leave. If someone on your customer team is worried about an account, take the concern seriously even when the data doesn't back it up. Customer-facing people often notice things that a scoring system will not pick up for weeks.

Try this today

Pick a customer you think is healthy. Check the four things above:

• Did they experience a quick outcome in the first few weeks?
  
• How deeply are they using what they bought?
  
• Is the person who signed the deal still in that role?
  
• How much of what they are paying for is actually in use?
  

If the answer to any of those is weak, the account might be at more risk than your dashboard shows. Add it to this week's review.

Go deeper

👉 Amplitude: The 7% Retention Rule – why customers who get value in the first week stay much longer than those who don't

👉 ChurnZero: Your customer champion just left. Now do this – a five-step plan for when an internal champion leaves

👉 Gartner: Unveiling the Customer Effort Score – why a simpler measure predicts loyalty better than NPS

👉 The CS Café: RYG in Customer Success – Hakan Ozturk on why most health scoring systems don't predict churn, and how to fix them

Coming up tomorrow

Tomorrow, we'll cover a question from Basecamp's decision-making guide that separates decisions that remove work from the ones that add more.

That's it for today! Have a great week.

P.S. Forward this to your head of customer success or account management and ask them which "healthy" customer on the current list they are actually worried about.