Why I Had to Stop Using Zendesk QA (and what I use instead)
Zendesk QA (formerly Klaus) is a fantastic tool with 3 fatal flaws if your goal is Sustainable Support.
It’s a multitool that tries to do everything, which is understandable (the founder’s goal was at least in part “get acquired by a big player” which… worked) but in trying to do everything, it misses a few key features specifically around creating actionable change on your support team.
Note: this article is specifically focused on the Zendesk QA internal interaction review functionality. We call them “Customer Response Reviews” or CRRs.
Flaw 1: Granular Data is Impossible
To be able to move the culture of a support team from “OK” to “world-class” a support leader needs extremely good data, broken down as granularly as possible.
If an agent is technically crushing it (included in the Expertise metric), but struggling to consistently make actionable changes to documentation (included in the Proactiveness metric), I need to be able to see that from a report. Zendesk QA allows leaders to see only the overall score of an agent over time, unless I dig in and manually extract the data around a specific metric (which is unnecessarily hard… ask me how I know).
An agent who is consistently overachieving in Expertise but consistently underachieving in Proactiveness will be able to float through if you’re looking at a Zendesk QA report where the two metrics are combined into one score. You can weigh one metric more than the other in the overall score, but you have to choose in this specific example whether you want the agent to be technically excellent or procedurally excellent. You can’t have both.
Flaw 2: Agent Scores are Demotivating
When I used ZenDesk QA for reviewing support agents, I tried (and failed) to shoehorn in my grading scale to their system. In the Sustainable Support model, there are three simple scores:
1: Keep scoring a 1, and you are headed toward uncomfortable conversations with me and HR.
2: Keep scoring a 2, and you are fine. No complaints.
3: Keep scoring a 3, and you are headed toward a promotion.
Zendesk QA forces me (via a non-dismissable percentage badge) to say “Your score is 60%” for an agent that otherwise I am more than happy with.
By measuring internal reviews like we measure CSAT (what a flawed metric!) we have no way to say “You’re doing just fine” or “we’re going to have to promote you if you keep doing this well.”
Only low performers read “70%” and are satisfied.
Zendesk QA is a great tool for making mediocre performers feel better about their mediocrity.
Flaw 3: Everything is Private
Nobody likes being critiqued in public. So at first glance, having agent scores be between them and their supervisor alone (complete with conversation threads for private chats and pushback) seems like the logical choice.
But Sustainable Support rests in part on the principle that we need to constantly be able to learn from each other. “Safely public ignorance” is critical. As long as a team member is not knowingly doing something wrong, having a spot where the whole team (certainly not the general public or even wider teams) can see and learn from coaching is paramount.
The way Zendesk QA encourages coaching would be akin to a basketball coach taking their point guard into a private office in the middle of practice to talk about how to change up their entry pass to the post, and then separately taking the power forward into the same office for “catching the entry pass” instruction.
When I used Zendesk QA for my teams, the best agents started doing something counterintuitive: they started to share my critical feedback with the whole support team. They wanted everyone to win, and for the entire team to level up their responses in the same way I was helping them to.
Zendesk QA makes this difficult. Heck, I had trouble even finding my own conversations with agents, or finding my supervisors’ CRRs for their agents. Having a place where everyone can see all the reviews for the whole team was nearly impossible.
Importantly for supervisors: knowing that your CRR is visible to everyone also helps to mitigate a lot of the biases that are prone to one-to-one conversations. Everyone can tell when a supervisor is giving preferential treatment and lenient scores to agents based on whether or not they get along better with them.
Find a Better Tool
Zendesk QA was a great tool to get me started on measuring and talking about Sustainable Support principles with my team. Ultimately I ended up making my own tool, so that I can (granularly) track the actual metrics that move the team toward excellence.
My tool is not a Zendesk QA killer (yet). It’s a proof-of-concept that aims for just one tiny piece of the overall QA process for customer service interactions.
What it does well today is force supervisors and leaders to start tracking the right data, in a way that actually motivates the whole team to excel in Sustainable Support.
