20: Measurement Inc.
Measuring performance is tough. As learning professionals, we've long been told that measuring the business impact of our solutions is the holy grail. Kirkpatrick level 4 etc.
But if we're honest, how often have we succeeded?
In my experience, it's notoriously difficult.
In earlier emails (subject: Square pegs and Right tool, wrong shed), I shared an example of an opportunity to improve sales at a tech company by reducing time taken to get new starters onboarded. And whilst that initiative never got off the ground, it was an example in which measuring business impact might have been possible. Sales can be easier to measure because the business already has mechanisms in place to track this.
But we rarely work on situations with such clear outcomes. So, we often resort to measuring engagement with a survey or what's been memorised via an assessment.
And whilst that's better than nothing, it's not an objective way to measure performance.
As employees, we experience how performance is measured first-hand. Whether it's on the job or during a formal review, our performance is evaluated by our managers. And whilst organisations have developed sophisticated ways of doing this, it's fair to say they're largely based on observation and opinion.
In the last few emails, we uncovered the discipline of Workflow Engineering - an approach that leverages an ecosystem of workflows, allowing us to engineer performance.
And with a system like this, we can consider a more objective approach to performance evaluation. By defining the desired workflow outcome, and then tracking whether that outcome was achieved, we can evaluate performance against the result.
Now, this may throw up some objections:
What if the desired outcome is difficult to define?
What if it can't be measured reliably?
What if the result takes months to appear?
All valid objections.
But the beauty of workflows is that we're not only restricted to measuring the outcome - we can also track whether the workflow was used.
Let's say our new salesperson follows the workflow but still doesn't make the sale. How does one evaluate that?
Historically, this would come down to the manager’s judgment. Which is an incredibly unfair responsibility, because they're often reliant on limited exposure and second-hand feedback.
But had the salesperson been using a workflow, the manager could see whether that person had followed the steps. They could see what decisions had been made, what outputs had been produced, and whether those outputs were up to scratch.
And let's be clear - in this scenario, the opinion of the manager is still vital. But it's grounded in visible work and predefined criteria.
Now, let's imagine two salespeople go through this process, both completing the workflow steps. But one makes the sale and the other doesn't. What then?
Well, firstly they can be recognised for using the workflow and following best practice.
But secondly, we might want to review the workflow - does something need fixing? Maybe the guidance wasn't clear. Maybe the steps weren't correctly defined. Maybe an extra review needs adding.
For example, a post mortem may reveal the salesperson followed the workflow correctly and produced high-quality work, but the customer had already committed to a competitor’s product. That helps us rule out the salesperson’s approach as the problem and look for clues elsewhere. Perhaps the lead qualification process needs fixing?
Now, one example might not tell us much. But as the same workflows are used repeatedly, patterns would emerge. We could compare how the work was performed with the outcomes it produced, and see where problems kept happening.
The outcomes tell us what happened - the workflow data helps us understand why.
And because the work is visible, we can improve the part of the workflow most likely to have influenced the result.
Hence, Workflow Engineering. Tweak the workflow to improve the outcome.
And this gives businesses what they've been chasing for decades - a more reliable way to understand, measure and improve how work is performed and the outcomes it delivers.
Yours,
- Ant
