How do you measure close rate on inbound calls?

Close rate is the share of calls that ended in a sale. It is one of the most watched numbers on a call desk, and one of the easiest to distort by accident, because the definition of a close, and the definition of the calls it is measured against, both have to be nailed down before the number means anything.

The calculation

Close rate = (closed sales / calls handled) x 100

As an example, say a desk handled 50 calls in a week and closed 6 of them. That is 6 divided by 50, times 100, an example close rate of 12 percent for that week. This is a hypothetical, round example, not a figure you should expect from any particular campaign.

Trap one: what counts as a call handled

If your denominator is every call delivered, including ones nobody answered, your close rate will look lower than a desk that only counts calls an agent actually spoke on. Neither choice is wrong, but they are not comparable to each other, and blending them across different weeks or different reports will make your own trend look like it is moving when it is not.

Trap two: when a close actually counts

Insurance sales sometimes finalize days after the call, once paperwork or underwriting clears. If you count a close on the day the call happened, fast closers look better and slow closers look artificially low, even if the same number of calls eventually convert. If you count a close on the day it finalizes, you need to trace it back to the original call to attribute it correctly. Decide which one you are doing and say so on every report.

• Define calls handled once, in writing, and use the same definition every time you report the number.
• Define a close once: application submitted, policy issued, or payment collected are three different lines, and they are not interchangeable.
• Attribute a delayed close back to the original call date if you are measuring by call date, not by close date.

Trap three: mixing campaigns

A close rate blended across campaign types, for example combining calls with very different products or different screening upstream, tells you less than the same number broken out by campaign. Two campaigns that average to a respectable blended close rate can hide one that is performing well and one that is not worth continuing.

A cleaner way to track it

Track close rate per campaign, on a consistent definition of both calls handled and what counts as a close, over a period long enough to smooth out a single unusual day. That is the version of the number worth making a decision on.

Give the number enough time before you judge it

A single day's close rate is close to meaningless on its own. A desk that closes 2 out of 8 calls one day and 1 out of 9 the next has not necessarily gotten worse, it has just handled a small enough sample that ordinary variation looks like a trend. Wait for at least a week, and ideally a few weeks, of consistent volume before treating a close rate change as real rather than noise.

It also helps to separate a new campaign's early close rate from its settled close rate. The first days of any new campaign often carry extra variance while your team is still learning the product, the caller profile, and where the common objections tend to land. Judge a campaign on its close rate once your team has had a real chance to get comfortable with it, not on its very first week.

If more than one agent handles a campaign, it is also worth looking at close rate per agent, not only as a blended desk average. A wide spread between your strongest and weakest closer on the same calls, screened the same way, is often a coaching opportunity that a single desk wide number would hide entirely.

Pair a per agent close rate with a review of that agent's actual calls before drawing a conclusion. A lower close rate can reflect a coaching gap, or it can reflect that agent simply getting a harder run of calls in a given week, and a recording review is usually the fastest way to tell which one you are looking at.

Common questions