KPI Construction Zone

Key Performance Indicators (or “KPIs”) are so frequently emphasized in business schools, management books, or training webinars that their importance is accepted as a truth almost without challenge. Indeed, a commonly cited axiom is “you can’t manage what you can’t measure,” inspiring a vision of a tidy portfolio of leading and lagging indicators, assigned at the company, department, and individual level.

In spite of this well-intended guidance, however, I often find myself working with and outside of my own departments to help teams set up their KPI structure and reporting cadences. Even our best managers and department heads can feel unprepared for this assignment, and setting aside time to figure out what really matters and tightening a team’s focus around a handful of metrics can be difficult with an already long list of day-to-day responsibilities.

The good news is that you can design metrics that offer you and your team value within a handful of weeks, if not faster. Within this post I’ll share how following these steps have worked for us as we have built the right set of KPIs from scratch. Feel free to also refer to the Resource Section for a template to support you through these steps.

1. Ask What does Good Look Like

• Step 1. Ask What Does Good Look Like?

• Step 2. Define Your Ideal KPIs

• Step 3. Build What is Practical

• Step 4. Set Targets

Step 1. Ask What Does Good Look Like?

Our CEO regularly uses this expression to talk about ideal performance at every level. From corporate goals discussed at town halls down to individual responsibilities measured during bi-weekly one-on-ones, we always want to know if we had a good day.

My advice: when overhauling your KPIs, don’t start with KPIs at all. Try to unshackle yourself from your previously existing metrics. Don’t put your existing KPIs at the top of your list, and don’t just load “best top KPIs for my industry” into your search bar. Instead, start by asking how you’d know if things were going well. Instead, the day-to-day language of a department will normally set you on the right path to building your KPI portfolio.

Within our People & Organizational Development team (“POD”, the term for our department that includes common Human Resource functions), we had been challenged to support our departments and hiring managers to reduce the number of budgeted, but unfilled, positions.

When discussing this assignment with the Vice President heading this department and his lead recruiter, we agreed that a good day for the department was any time a candidate accepted a position. Matching the problem (unfilled positions) with this intuitive measure of success (accepted offers) led the team to start using the term “time to hire.” “Good” was then more specifically not just as filling positions, but as filling them quickly.

Step 2. Define your Ideal KPIs

Only after you’ve described a story of success should you start thinking about the numbers again. Imagine a world in which you could have any piece of information about your business that you want, easily, immediately, and accurately. If that were the case, what would you measure that would inform you that you were achieving “good” as defined in Step one?

This is a brainstorming exercise best done with a few people, including your department head or team lead, and ideally some others from her or his team. Try to get to a list of at least five or six, but as many as ten or twelve, potential KPIs. Don’t worry about how we’d get various data points into the same file or database, or even to define the targets yet. Success at this step is aligning on what you want to track.

Steps one and two shouldn’t require a massive time commitment. I’ve often found they can be combined into a single session of an hour or so, and you should be able to create this list of potential KPIs without any substantial research beyond your and your team’s existing understanding of their responsibilities.

Within the POD department, we agreed that the best metric around time to hire would be to measure the gap between a job posting date and start date. Good would be achieved if hiring managers were identifying and extending offers to new Team Members within a defined target threshold. While we hadn’t started with any KPI in mind, this qualitative description of success naturally pushed us towards a specific metric. Ideally we would also be able to calculate the average time to hire by department, location, or role, grouped into cohorts by date.

Step 3. Build What is Practical

At this point, it’s time to confront the practical reality that your KPIs need to be anchored to the reality of the tools at your disposal. Unfortunately, your ideal metrics may require data that you just don’t have. So, if you’re playing “Business Cliche Bingo” at home, it’s time to check off the “Don’t Let Perfect Be the Enemy of Good” box. Just because you can’t get this KPI to reflect exactly what you imagined doesn’t mean that an imperfect version won’t still convey valuable information.

There’s a formal process you can employ in which you map your list from step 2 on a two by two matrix. Segment your KPIs on the horizontal axis by “need to haves” versus “nice to haves.” This isn’t going to be an exact science - what should be in your head is “how important is this piece of information in my assessment of if we’re doing a good job?” The most important (it shouldn’t be more than four or five) go on the right side. The rest go on the left.

Use the vertical axis to identify how closely you can easily match the metric you want with the metric you can calculate, with “easier” being at the top and “harder” being at the bottom. Sometimes data simply isn’t available. Other times you can assemble what you need, but doing so would require meaningful manual work each measurement period. Again, use these rules to sort your list from Step two.

Often your favorite KPIs are sitting in that top right box. For those that are this lucky, you can stop here and move to the next step.

Our ideal time to hire calculation fell firmly into the right side of our grid; it seemed to be the most valuable way to measure good for the department. As we refined the metric in this step, we also determined that segmenting new Team Members by department (Technology, Sales, Exec Positions, Other) would provide additional useful information when assessing our success.

Ease of calculation was less clear, though. Our recruiting software was distinct from our human resource software; we tracked when a position was posted and the corresponding start date for a Team Member hired for that position in two different systems. This calculation therefore would require a change in one of the systems or the use of manually updated spreadsheets. In our case, given our size (about 250 Team Members, with around 5-15 new hires per month), we agreed that asking the recruiters on the POD team to manually update a shared spreadsheet was not unusually burdensome. So while not formally in that top right quadrant, this framework revealed that this KPI, while requiring a little brute force and duct tape to calculate each month, was our best option to measure performance.

In contrast, if you find that this exercise left you without clear direction in selecting your KPIs, there may be options available based on your specific challenge.

Challenge	Potential Resolution
​Can you live without that metric?	Move on with the KPIs you identified for now, and revisit in the future, when tooling or available reports may catch up with your needs
Is the data you want unavailable?	Can you live with less precision or broader segmentation of data? In our case, if we weren’t able to easily track time to hire by department, we would have asked if we could still get value by tracking the KPI for the company as a whole?
Is there significant effort required in the calculation?	Consider measuring over a longer period of time, so the effort required for calculations is less frequent. You may then gradually learn to automate or simplify certain aspects.

Step 4. Set Targets

Financial and operational metrics will demonstrate their greatest value when measured against defined targets. Taking time to set these goals will improve the value and significance of the weekly, monthly, or quarterly presentation of your department’s performance. We have found three useful methods for setting the benchmarks for our newly identified KPIs.

Type of Benchmark	Detail	Example
Budgetary	​Intrinsic goal required to reach short or long term Company targets; often defined by Budget	​The target for new customer leads by week will be set as required to reach the customer acquisition budget
Industry	​Based on results of peer group or broader industry participants	Annual customer retention targets set to match results achieved by best-in-breed peers
Historical	Maintain or improve relative to company or department’s historical results	Seek to improve on prior year’s employee voluntary turnover rate

When defining our time to hire goals, we were able to rely on the monthly results from the prior two years. We did have to scrub the historical data for our purpose, segmenting all new hires by department, since the information was not being captured in that manner within the recruiting software. Given our size, that work required to improve the quality of the data used in setting our target was reasonable. From there, we recommended an improvement relative to historical performance and aligned on those KPI targets with the department hiring managers.

Should none of these three benchmark types yield a useful target, continue to move forward with calculating and recording the results of your KPI. As you start to review the results with your team, a natural goal should reveal itself.

Your Turn

As mentioned, this exercise can likely be completed quickly, to great benefits. That said, unless software tools were put into place specifically to calculate certain KPIs, the results of Step 3 will likely demand some manual calculations. Don’t be discouraged and don’t quit there! Let’s all say it again together: Don’t let perfect be the enemy of good! There are a number of ways to push through and end up with some useful metrics for you and your team to discuss on a regular basis despite the need for data manipulation and manual calculations.

• If you or your team don’t have the time to manually calculate five or six KPIs per week, then start with two or three.

• Some organizations have reporting or FP&A teams for which reporting is part of their mandate. Check with your assigned finance team resource or, for smaller organizations, even your CFO, for help. My role (CFO, and now President) includes helping teams get up and running on tracking KPIs - I found that I could often help them consider options to calculate the metrics they had identified with less effort, or even have an FP&A team member absorb the responsibility.

• Ask your Team Members if anyone wants to take the lead. Many departments have an Excel whiz or numbers person hiding in plain sight, and he or she may be eager to step up and show off their skills. It’s also a great way for individual contributors to distinguish themselves through a new responsibility as they consider their career development.

Finally, remember that this won’t be your final draft. Once you have a set of KPIs and begin tracking and reporting against your targets, you’ll decide that something isn’t as valuable as you thought it would be, and some other piece of information is lacking.

As usual, I recommend just getting started. Grab the template in the Resource section, pull together a handful of your team members and start brainstorming on what good looks like, and you’ll find pieces falling together quickly.