Here's the gist of my comments to the support manager about each of the five metrics:

(1) Timeliness of notes: Most trouble ticketing systems can track this, typically by determining the period between the customer's request for help, and the time the support rep communicated with the customer prior to creating or updating notes. When measuring timeliness, it's important to subtract periods when the center is closed or the support rep is not scheduled to work.

(2) Accuracy of notes: Frankly, I think this is difficult to measure within a support center in the absence of a knowledge base. (An internal evaluation of a knowledge base of answers to callers' questions would be helpful, but it doesn't necessarily address support reps' notes.) Instead of accuracy, a more helpful metric is resolution, a metric I address in more detail below.

(3) Sense of ownership: If the trouble ticketing system documents whether and how often support reps hand off requests to others rather than resolving them – and whether these reps are available and have the training to assist with these requests – then support managers should be able to measure individuals' and teams' ownership of issues that customers bring to the support center's attention.

(4) Contact with customers: This metric, as I understood it, seemed to refer to how much time a support rep devotes to communicating with customers.

(5) Resolution: As is true with timeliness of notes, most trouble ticketing systems can track resolution, typically by determining the period between the customer's request and the center's resolution of the request, subtracting hours when the center is closed or ordinarily unavailable.

I shared my belief with the support manager that among all these metrics, resolution is the most important. The best way for a support center to conceive of resolution is not in terms of the support rep, but rather in terms of the customer. Besides indicating the overall time a support center devotes to resolving a customer's request, a measurement of resolution should also indicate all the reps who assist a customer with a particular issue, and how long each rep is communicating with the customer.

After I presented my thoughts to the support manager about the individual metrics, I provided some feedback about the five metrics as a group.

First, the five metrics focused too much on documentation but not enough on outcomes. Second, these metrics emphasized individual efforts rather than the collective ability of support reps to assist customers. Third, and most important, the support center had an alternative to calculating all five metrics, which was to roll up these metrics into one or two that encapsulated some of the others.

Resolution, for instance, contains within its definition support reps' sense of ownership of customers' requests, as well as the reps' contact with customers. And if you define resolution as a workflow where notes are part of the resolution process, then resolution can also serve as an indicator of how timely support reps are with documenting their efforts to assist customers.

The upshot: If the support center were to roll up these five metrics into one or two metrics that treat resolution as a sequence of steps in time that can involve multiple individuals, then the manager would have fewer metrics to report on.

For the sake of evaluating operational efficiency and effectiveness, one metric I would retain among the five is resolution. Another metric I would keep, also for the sake of gauging the support's center efficiency and effectiveness, is sense of ownership.

I've already explained why resolution is the most important metric of the five we started out with. Why is sense of ownership important as well?

Sense of ownership is a clear indicator of how many support reps participate in resolving support requests. For requests where reps' training should enable reps to assist customers without requiring their colleagues' assistance, ownership could be a useful measure of efficiency. But I wouldn't recommend using the notion of ownership as a means to discourage collaboration among colleagues, especially in circumstances when collaboration yields faster resolutions for customers than attempts among reps to resolve issues all by themselves.

In addition to revealing tendencies among support reps to resolve issues on their own, ownership also helps identify which types of support requests lend themselves best to collaboration, or whether there are gaps in reps' training.

By looking at ownership as a measure of an overall customer support operation – in terms of determining training needs for reps and discovering which types of requests are easiest to resolve through collaboration – you have a better sense of how well reps fulfill customers' support needs than if you treat ownership strictly as an indicator of an individual's self-sufficiency.

So there we are. We began with a set of five metrics, and boiled them down to two.

Actually, there is one metric among the five that I proposed the customer support manager eliminate entirely: accuracy. Why? When a support rep seeks to discover the reason for a request for support, the outcome of the diagnosis – especially if a customer decides if a request is resolved – reveals more about the usefulness of the support rep's notes, both to customers and to other support reps, than any attempt to determine the notes' accuracy.

With that said, I would point out to readers that if you have a knowledge base already in place, it is helpful to have people review its entries on a regular basis to ensure that the entries convey accurate answers to customers (if customers can view them) and to support reps who consult the knowledge base on behalf of customers.

The moral of this story is that before you think about ways to calculate a metric, the first and more important step is to determine whether you need that metric at all. The more efficiently you select your metrics, the more information about your call center you can absorb, and convey to others, in the least amount of time.