One of the most popular queries I get is, “How many CSMs do I need?”. The second most popular is probably the long running, “How many support engineers do I need?” — but that one creates less anxiety, as support managers are quite comfortable with using cases as a unit of work, to the point that they quickly estimate the number of cases that can be solved per person per day based on the complexity of the product.
But for CSMs, there’s no equivalent unit of work. A customer may be large or small (cases can be complex or easy, but they average out; customers, on the other hand, are usually assigned based on size). And the array of services delivered to individual customers is astonishingly large. So we need to work a little harder to define CSM utilization–but it’s really not that hard. Here are the two approaches I use, and I like to use them both.
Idea #1: top down
The first approach is to create a model based on the offerings you are planning, as follows:
- Inventory the onboarding and retention programs.
- Estimate the time required to deliver each touch. For instance, a training session may require an hour, plus 30 minutes of pre time. A QBR may require 3 hours to prepare, etc.
- Add it all up based on your forecast of new and retained customers.
- Calculate active time for CSMs by subtracting vacations, sick time, and routine administrative tasks.
- Divide the total by active time and voila!
(After you complete your calculation, you will likely want to alter the offerings so make them less time-hungry. Revise and repeat.).
Idea #2: time-and-motion study
The second approach is to capture what the CSMs are actually doing. You’ll need to go this route if you don’t have formally-defined offerings (you really should!) but it’s useful to do in any case, as a sanity check.
- Recruit a handful of willing CSMs. Recruit a representative sample, not just top performers.
- For a week, or at least a couple of days, ask them to log their time by customer and activity. They can use an app (try Toggl, for instance) or simply fill out a spreadsheet showing the client they are working with and the type of activity they are engaged in.
- Compile and classify the data.
Time-and-motion studies require effort from the CSMs, will only reflect the behaviors of the test subjects (and only for those days you choose to conduct the study), and they also influence the behaviors. Still, they can be useful if you select your test subjects carefully.Debrief the CSMs afterwards to see what they would have wanted to do, but did not have time for, and where they thought they can be more efficient.
If you use both approaches, contrast the outcomes. You may be surprised to see that your top-down calculations omit important items (such as escalation management, or prep time for various activities) or that time-and-motion outcomes show that activities are routinely avoided (customers may not want weekly calls). Spend time on the reconciliation.
And with that, you will know how many customers a CSM can handle, delivering a specific array of services.
How do you build a staffing model for CSMs? Please share.