KPIs and Metrics: The Difference and How To Use Both

What Is a Metric?

A metric is a quantitative measurement used to evaluate the performance of a project. Metrics can cover any aspect or component of project performance. There aren't many bounds placed on what a metric can be. Generally speaking, metrics are derived by comparing measurements taken over time to a predetermined baseline. The idea is that these comparisons tell you whether your project is on track or not. However, not all metrics are created equally. 

Using the car and sedan example above, the number of active assembly lines is a useful metric. It gives you an understanding of your factory. However, it's not vital to the success of your operations. It doesn't tell you whether you're making the right kind of cars, it doesn't tell you if you're meeting your orders. Sure, you can use the metric if you find that you need to produce more cars. But it's not the metric that told you whether you were behind or not. That's what a KPI is for. It's a special type of metric. 

What Is a KPI

As we said above, a KPI, or key performance indicator, is a specific type of metric. A KPI is a quantitative measurement, just like any other metric. Specifically though, a KPI keeps track of whether a project is achieving its core goals or targets. You can think of it as a super-targeted metric. 

It's easy to remember what a KPI is by splitting the constituent words into two parts—key and performance indicator. Key refers to the core goals and targets. Metrics are only KPIs if they relate to something vital to the success of a project — they tell you how effectively you are achieving your strategic and operational goals. Meanwhile, performance indicator refers to the quantifiable nature of KPIs. That is, KPIs should be specific and measurable. This allows you to easily evaluate, compare, and gauge performance. 

Using the citrus example above, the number of citrus trees planted is a metric. It's good to know, but it's not vital to success. Meanwhile, the percentage of lemon orders filled is a KPI. This tells you whether you're meeting core demands of your lemon business. If you find you're behind, you can then look at the number of lemon trees and see how many are flowering or fruiting. This might change your planting, watering, or fertilizing schedule. It all comes from the KPI telling you whether you are achieving your core goals or not. 

Real-World Examples

Okay, so we've explained metrics and KPIs. We've touched on the difference between them. But what does that difference really look like? How do you know whether a metric is important enough to be "key"? 

I'm sorry to say it's different for every business, every project, and every stakeholder. However, there's no need to lose hope! Metrics and KPIs are best understood through examples. 

Metric Examples

Of course, all the KPIs described below are also metrics. But metrics encompass much more than just KPIs. Generic metrics you'll see mentioned include things like amount of work completed (percentage of tasks or work items that have been completed out of all of the required tasks), and effort estimation versus actual effort (number of man-hours it takes to do a job compared to what was estimated). Here are some DevOps related metrics: 

• Number of user stories shows how many user stories there are in the backlog.

• Number of pull requests lists the number of pull requests waiting to be checked or approved.

• Environment utilization shows where releases and features are in the development cycle (production, acceptance testing, staged, under development, etc.) for different environments (mobile app, financial platform, website, etc.).

• Sprint burndown tracks completion of work throughout a sprint.

For more on some of the best Agile metrics to track, check out this post. 

KPI Examples

Commonly, you'll see articles that mention sales KPIs. KPIs like client conversion ratio (the percentage of leads that buy a product), and sales growth (increase or decrease in sales per monthly, quarterly, or yearly period) are easy for all audiences to understand. 

However, as a technology focused website, we thought it best offer examples of some common IT- and DevOps-related KPIs: 

• Deployment frequency records how often features and capabilities are launched. This directly indicates whether your project is on track, ahead, or falling behind.

• Mean feature lead time records how long, on average, it takes for a feature change to occur in the code. This indicates whether your changes are happening fast enough to satisfy client requests.

• Mean deployment time measures how long, on average, it takes to roll out your deployments once they're approved. You'll be able to see if there are bottlenecks holding up the project.

• Deployment failure rate shows how often deployments cause outages or other issues (this is sometimes called mean time to failure). From this you'll know whether you're impacting service level agreements, such as uptime.

For more information on how Plutora displays these kinds of KPIs, check out their business intelligence section. 

Using Both To Your Advantage

Okay, so by now you should have a good idea of the difference between metrics and the subset of metrics that are KPIs. The next important step is to make sure you can use them to your advantage! This involves first defining good metrics and KPIs, and second, using business intelligence to display the values on the right dashboard. 

Creating Metrics and KPIs

First up, metrics and KPIs are different for every business, every project, and every stakeholder. So how do you create good ones for your organization? Luckily, the SMART criteria provide a guideline to follow: 

1. Be specific. There's no use having vague metrics and KPIs the stakeholders can't understand. Define your measures clearly and with detail.

2. Make it measurable. All metrics should be reduced to a number. You want it to be quick and easy to understand.

3. Ensure it's assignable. Who's going to take ownership of the metric to make sure it's meeting the expectations?

4. Be realistic. Sure, you can try to measure number of conversations as a proxy for better communication, but it's not very realistic.

5. Make it time-related. A metric is designed to provide a snapshot. If it's not time-related, you won't be able to compare metrics in week 1 and week 10 of a project.

Finally, if it's a KPI: 

• Ensure the thing you measure is core to success. Does the measure directly indicate whether your project is on track or not? That's the key part of KPI.

Choosing the Right Dashboard or Report

I'll bet you're ready to jump right in and start recording data now that you know how to make good metrics and KPIs. But that's not the whole picture. 

To fully benefit from the data, you need to be able to review, understand, and communicate your findings. 

Enter business intelligence. 

Many of you are familiar with the benefits of visualization and business intelligence. Arguably, those benefits are most clearly seen with metrics and KPIs. They're a perfect match. Metrics are specific, and measurable, while business intelligence is all about turning business data into meaningful insights. 

What Next?

Now it's time to get down to business. 

First, think about your individual organization, projects, and stakeholders. What measures are they interested in? 

Draft some KPI and metric ideas using the SMART guidelines above. Then, evaluate each by asking if it makes sense for your circumstances. 

Next, consider where the data will come from. Can you make each one measurable and specific? 

Then, have a look at how you might present them on Plutora's dashboards. Are you visualizing the right things? 

Finally, put all this together to improve your business performance!