Metrics to Track or Toss: Craig Stoss’s 5-Step System to Simplify Support Data

🦸 This newsletter is brought to you by Supportman.io

You don’t need a 37-tab spreadsheet and 15 dashboards to understand your support data. You do need this conversation with Craig Stoss.

Craig is the VP of Partner Solutions at Kodif and a 25-year veteran in support and CX. In this episode, he shares:

• Why single-select fields are destroying your reporting.
  

• The only two metrics you need to start capacity planning.
  

• Why migrating all that old data to your shiny new system is a waste of time.
  

• How AI is flipping data reporting on its head (and why that’s great news).

To skip the summary and go straight to the source:  

And don't forget to check out Supportman.io, the sponsor of this podcast, which connects Intercom to Slack and utilizes AI to provide agents with real-time feedback and surface issues.

The Problem:  

Support leaders are buried in messy, outdated data. We try to migrate every old ticket, build overly complex dashboards, and track dozens of metrics, only to end up with unclean reports no one trusts. It’s confusing for agents, unhelpful for leaders, and impossible to act on.

Worse, we spend so much time trying to make it all look clean that we forget to ask: is this even useful?

The Answer:  

Craig Stoss, VP of Partner Solutions at Kodif, shares a 7-step process to simplify support data, without sacrificing strategy. From what to toss (email threads, rigid dropdowns) to what to track (volume and handle time), Craig shows how to build a smarter, leaner system.

1️⃣  Stop overcomplicating your data structure

• Avoid rigid dropdowns and single-select fields

• Let agents add values dynamically if possible.

• Prioritize flexibility over forced categorization.

2️⃣ Don’t waste time on historical data migrations

• Ask yourself: Will this old data actually change my future business decisions?

• Keep only high-level trends like volume and handle time.

• Forget detailed ticket threads and historical categories—they don’t matter.

3️⃣ Choose a simple and consistent capacity model - 80% capacity tends to make sense

• For high volume, try the Erlang calculator (Craig’s is at stoss.ca).

• Otherwise, a basic formula works:
  (IncomingTickets×AverageHandleTime)÷AvailableWorkHours=RequiredStaff(Incoming Tickets × Average Handle Time) ÷ Available Work Hours = Required Staff

• Don’t adjust for every variable—just use averages and correct as you go.

4️⃣ Factor in real-world time for your team—not just ideal capacity

• Target ~80-85% occupancy to avoid burnout.
  

• Factor in internal shrinkage (meetings, training) and external shrinkage (vacations, sick days) using simple averages.
  

• Don’t try to predict every sick day—keep it high level.

5️⃣ Only track data that drives meaningful business decisions.

• If it doesn’t directly impact hiring, customer experience, or financial decisions, stop tracking it.
  

• Prioritize metrics like bug frequency, refund requests, and volume trends that influence real business outcomes.

The Impact: Craig’s framework has helped support teams reclaim time, reduce reporting stress, and finally build systems that reflect the reality of the work.

The result? Better forecasting, smarter decisions, and more time spent actually supporting customers.

If you’ve ever stared at your metrics and thought, “Why are we even tracking this?”—this is the episode you’ve been waiting for.

We’d love to hear what stood out most. Reply to this email or leave a comment.

🎙 Catch the full conversation on Live Chat with Jen Weaver →

See you next time!

Jen