What Comes First: Data or KPIs?

by Stacey Barr |

It’s not hard to spend millions on setting up new data management and business intelligence systems. And it can take years to implement, too. Plus there’s the tricky piece of getting people to buy in to collection and capturing all that data. But we need to do it, before we can start measuring what matters… Or do we?

chicken looking at an egg

It’s easy to argue that you can’t have a performance measure until you have the data for it.

But it’s equally logical to argue that you don’t know what data to collect until you know what you need measure.

Don’t you agree?

Sure, not all data is collected just so we can measure performance. We also need data for regulatory and governance purposes, operational decision-making, day to day transactions, research, and exploratory analysis.

These purposes for data are often accommodated in the requirements definition phase of a data management project. But the requirements for KPIs and performance measures are often overlooked.

A big reason for this is the lack of a deliberate process for performance measurement. It’s not often seen as a legitimate management process, like governance, regulation, and decision-making are. But it is.

If we delay our performance measurement process until we have data, we will have data that fails to fit the needs of our likely KPIs. Why? Because:

  • the data management project will make assumptions about the types, formats, granularity and linking that is useful for the data to have – assumptions that often prove wrong
  • the data management project will focus on the data that people have a known need for, and won’t capture data for not-yet-known needs (which performance measures and KPIs often point out)
  • new requirements for data, which performance measures and KPIs often point out, need to be pilot tested and tweaked before they become routine parts of the data management system

And you’ll be forced to fit your KPIs to the data you have. That’s tragic, because one of the most frequent – and powerful – impacts of a good performance measurement process, is that your mind is opened and you see more clearly what you need to know about performance. You see what you hadn’t seen before.

So don’t delay your performance measurement until after you have the data. You won’t have the data, you’ll be a year or two behind in your quest for great KPIs, and you’ll have a load of rework to do on your data management system.

No data should be no excuse to measure the right KPIs. Instead, commit to designing the performance measures you need. Then commit to pilot testing those measures with manual data collection and capture. Then tweak the data, tweak the measures, and use them both as part of the requirements specification for your data management system.


How have you linked your performance measurement process to your data management process? Which one drives the other?

Speak Your Mind

Your email address will not be published. Required fields are marked *

  1. Agree with you! I believe it’s important to define the performance measurement not by considering available data but by considering what is important to the organization/team in place. And often needed data doesn’t exists in IT systems. This doesn’t prevent to start with softer information and decide to collect more factual data if it makes sense (cost/benefit wise). I’d like to quote Jean Rostand: “Waiting to have enough information to act is to condemn ourselves to inaction”.
    And more than performance measurement, it’s all about the decision making process that needs to be put in place. Start from day 1.

  2. George Lowry says:

    It is critical, but a luxury, to identify the measure and data before you define a process necessary to collect said data. We find it more common that many already have a defined process capable of capturing the necessary data only to have their organizations not follow the process. This results in poor data quality that causes visualized information to not be trustworthy. In these cases, we take the approach to get customers looking at their data (anyway) and defining the screening criteria necessary to identify what is bad about it. This typically will result in identification of the underlying process (or change management) deficiencies so the process or organization can be targeted for correction. No sense in throwing the baby out with the bathwater so to speak.

Upcoming KPI Training

>> Africa, In-Person, Dubai UAE, 3-5 May 2023

>> Australia/NZ/Asia Pacific, In-Person, Wellington NZ, 9-11 May 2023

>> UK & Europe, Online Interactive, 22-26 May 2023

>> North America, Online Interactive, 29 May - 2 June 2023

Register for the next PuMP Blueprint Workshop near you

Reprinting Articles

You are welcome to use articles from the Measure Up blog, with these requirements

Connect with Stacey

Haven’t found what you’re looking for? Want more information? Fill out the form below and I’ll get in touch with you as soon as possible.

    *We respect your email privacy.
    Suite 117 Level 14,
    167 Eagle Street,
    Brisbane Qld 4000,
    Stacey Barr Pty Ltd
    ACN: 129953635
    Director: Stacey Barr