Are you guilty of using the following methods as your approach to measure selection:

*

brainstorming with your team in a one-hour session during your two-day planning workshop?
*

trawling the internet or other places to find out what others like you measure?
* asking your IT guy or gal what data you have and creating measures from that?
* hoping someone will tell you (maybe a consultant or a stakeholder)?

These aren't approaches to measure selection. They are just ways to gather ideas for what to measure. None of these methods include any kind of overt and deliberate evaluation of which measures are the best measures. And you're probably wondering why your organisation has so many meaningless measures! Plus, on the flip side, these methods may have left you high and dry without any viable options for measuring some of those less tangible results like culture or sustainability or engagement or confidence.

If this is your burden - having no logical, practical way of choosing or designing the measures that can truly convince you that you're making the differences you need to in the world - then here are three tips that might make your life a little easier:

1. measure the result, not the action
2. what you can observe and describe, you can measure
3. don't measure it just because it's easy

measure the result, not the action

"Ensure that people with a disability do not experience discrimination and have their particular needs for services and support acknowledged and met." Such an inspirational and noble goal is so easily cheapened by a measure like "Establishment of an effective Advisory Council on Disability". Such measures track the activity associated with the initiatives hypothesised to produce the results implied by such wonderful goals. They can't let us know how much or how frequently people with a disability experience discrimination.

No doubt you're going to want to monitor activities in your organisation, but what meaning does that have unless you are first monitoring the results those activities exist to produce or influence?

what you can observe and describe, you can measure

Why is it so hard to measure results like "achieving equi-marginal efficiency for trade-exposed industries on a least-cost trajectory within a general equilibrium model"? The answer is not because no-one has thought of the right measure yet! If we go back to basics, measuring is about observing and collecting specific information about something. If you don't know how to recognise when that something is happening, you can't know where and when and how to collect information about it, can you?

So before you can measure "achieving equi-marginal efficiency for trade-exposed industries on a least-cost trajectory within a general equilibrium model", you need to know what equi-marginal efficiency and least-cost trajectories look and feel and sound like when they're happening. Even more, you need to be able to describe it in words that are evocative, words that conjure rich and detailed and shared pictures in the minds of people before they select or design measures.

don't measure it just because it's easy

It's easier to select more viable candidates for measuring a result when you describe it richly and without weasel words. In fact, you can end up with so many candidate measures that you might be tempted to pick those that are the easiest to bring to life. You already have the data, it would be a waste not to use it, no-one's got the time to collect more data. But you'd be falling into one of the deepest traps of organisational performance management: not making sure that your organisation has the data it really needs.

You'll need to think about more than just the feasibility of each potential measure in deciding on the best ones. Measures are meaningful when they have strong relevance to the result you want them to evidence.
use your brain when you design your measures

Three basic steps to better measures: focus on the result and not the activities, articulate clearly what that result looks like, and shortlist your potential measures by balancing feasibility with strength of relevance. Yes it will take a little more time that you have probably been giving to measure selection, but it will save you loads more time than you have probably been wasting managing with the wrong measures.