On this page
Two metrics with the same name
In an outbound or collections call center, contact rate is the share of calls that reach a live person.
In customer support it is the opposite, how many customers reach out to the team. This entry is about the second.
Why it matters
It is the only support metric that predicts the future.
If volume per customer holds steady, support will grow at the pace of the business and you will keep hiring. If it falls, support stops being a proportional cost. If it rises, there is a product problem no support tool is going to fix.
How to calculate contact rate
Divide the contacts in the period by the base that makes sense for the business.
A contact is each conversation a customer opens, on any channel, not each message. The figures in these examples are illustrative.
- Per customer. Contacts in the month over active customers. 300 contacts and 1,500 customers give 0.2 contacts per customer, or 20 per hundred.
- Per order, in a store. Contacts over orders in the period. 450 contacts on 9,000 orders is a 5% contact rate.
- Per active user, in an app. Contacts over active users, useful when whoever pays is not whoever uses.
The absolute number says nothing
Four hundred tickets a month is a figure without meaning.
With forty customers it is an emergency and with four thousand it is an exceptionally clear product. Only relative to the customer count is it useful for deciding, and only the trend is useful for forecasting.
What drives it up
New features (each one multiplies the possible questions), interface changes with no warning, stale documentation, and customers from a new segment who use the product differently.
All four are avoidable, and all four tend to surprise.
Segment it by plan
It is the most revealing cut.
If the small plan generates three times more questions per customer than the large one and contributes half the revenue, you have a pricing problem dressed up as a support problem.
The relationship with cost
Contact rate times customer count times cost per ticket gives total support cost.
It is the sum you should be able to redo at any moment, and the one that turns an argument about impressions into a decision.
How to reduce it
It comes down by removing the reason the customer writes, not by making it harder to write.
- Document the most repeated questions where customers look for them.
- Show in the product the data people ask for today, such as usage, an order status or an invoice.
- Warn before a change to the interface, the price or the terms.
- Review the fastest-growing topics every month and take them to product.
What is a good contact rate
There is no figure that is good for everyone.
It depends on the sector, the price and how complex the product is, and published averages mix businesses that look nothing alike. The useful comparison is with yourself, month on month and by segment.
An AI agent does not lower it
The customer still wrote in, even if an AI agent answered.
What falls is the share that reaches the team. Measure the two separately, total contact rate and what the AI resolves without a person, because mixing them hides whether the product is creating more questions or fewer.
