Case study: how Workable’s TechOps optimized its processes

Imagine leading a team that’s meeting its targets, and yet, you know there’s potential for more. The catch? Identifying those areas for improvement when everything seems to be functioning well.

This is the intriguing challenge that George Zikos, Senior Director of Workable’s Technical Operations team (TechOps), confronted.

With his team’s KPI targets plateauing and expectations remaining high, George embarked on a journey of introspection and process optimization. Let’s see how George and the team uncovered and addressed hidden challenges to not only boost their performance but also to ensure that this well-performing team remained engaged and motivated.

The problems

There were many issues that George and his team recognized, which needed to be resolved:

1. KPIs have plateaued

TechOps is a well established team that has been performing well. Their KPI targets had steadily improved over the years but had recently plateaued. Targets were still being hit which masked a hidden challenge: they were no longer improving.

2. The impediment of overlapping metrics

One issue was the blurred accountability caused by sharing a Jira board with other teams. It made isolating TechOp’s specific impact difficult, particularly for metrics like ‘average resolution days,’ which also involved other teams like Support.

3. Increasing expectations

With a tenure of 2.5 to four years for the team and six years for George himself, expectations of continued performance improvement were high, even factoring in a recent reduction in team size when a departing employee wasn’t replaced.

4. The need for challenging work

With experience comes the risk of stagnation and complacency, especially when the goals at hand are no longer as challenging as before. A highly skilled team needs motivating work to remain engaged.

The goals

Goals are crucial for success – you can’t build solutions without knowing what you’re aiming to do in the end. George’s goals were as follows:

1. Improve performance

Identify processes which can be optimized to improve the team’s performance against its top level KPIs.

2. Motivate the team

Change the team’s perspective, have them focus on something new and challenging to reinvigorate and motivate them.

The method

With clarity around the problems, George was able to get to work on building solutions. These included:

1. Unearth hidden inefficiencies through process audit

George took a detailed look at his team’s work, analyzing hundreds of Jira cards, reading all the comments and tracking the flow of tickets from one team to another.

This very manual, laborious work to understand the work of his team and the processes involved was the key to unlocking improvements in performance.

His deep dive unveiled a hidden inefficiency: frequent back-and-forths between his team and others were causing delays in ticket resolutions.

2. Identifying new, more granular metrics

To quantify this inefficiency, George identified two new metrics that seemed apt: “Tickets Resolved on First Response” (TRFR) and “Median Time to Acknowledge” (MTA) that a ticket was being worked on.

“Having identified where there was an area for improvement, I supplemented my own thoughts with some online research, reading many articles on the metrics other Ops and DevOps teams use to drive performance,” George explains.

3. Tracking the new metrics

George collaborated with IT to set up additional reporting in Jira to track these new metrics, creating a new benchmarking system.

He adds: “Without this reporting, I wouldn’t be able to see that the changes we made were really valuable at all. Jira didn’t have what I needed out of the box so having IT to help was essential.”

4. Bringing the team onboard

George shared these two new metrics with the team, which interestingly resulted in a quick rise in TRFR performance, from 58% to 62%. George put this increase in TRFR performance down to the team simply being aware of this metric, saying;

“We’ve been a team for a while now,” says George. “The team understands that when a metric is being tracked it has some importance, even though I said it was not the priority right now. Knowing that, I believe they ‘self-adjusted’ to optimize it and we saw results straight away.”

MTA did not see an immediate improvement in this way, so more work would be required to make an impact.

5. Setting goals based on new insights

George set a new target of 4 hours for MTA, down from the 5-hour median which was observed during the benchmarking period.

Four hours was chosen as it aligned to how the team splits their day half between Ops tickets and the other half on Labb tickets.
TRFR, despite its improving trend, was only set to be monitored for the rest of 2023 with a goal of maintaining the current level until the 2024 scorecards.

6. Implementation and results

To aid the team in meeting the new MTA target, George knew he would need to focus the team on certain cards instead of simply expecting the team to magically improve speed everywhere.

To that end, George implemented a Zapier automation to send Slack notifications only for new High-Priority and Approvals tickets, excluding the less time-sensitive tickets. This avoided creating too much notification noise for the team but meant the most important tickets could be addressed as quickly as possible. Previously, the team would have to check Jira periodically for any new tickets.

This brought impressive improvements in MTA, dropping from 5 hours to 3.62 hours, with June even hitting an impressive 2.93 hours.

Outcome

TechOps have not only seen improvements in the new metrics but have validated that improving those metrics drives improvements in top level metrics.

Those top level KPIs, which had previously plateaued and seemed impossible to move, improved by 5% in H1 2023.
What makes that improvement even more impressive is how it was accomplished by a smaller team (6 down from 7).

Lessons learned and future goals

George puts it very simply: “Never settle, there is always room for improvement.”

TechOp’s journey underlines the power of detailed analysis and continuous monitoring, even when performance seems satisfactory. The key lesson here is the value of curiosity and persistence; even when performance is good, less apparent problems can still be unearthed and addressed to drive significant improvements.

The team’s next step is to ensure the team maintains the new MTA targets, keeps TRFR consistent, while also working to improve a newly identified metric – the Velocity of Labb Stories.