Want a simple way to make your policies more effective? Then take a leaf out of designer Dimo Yarovinsky's book. He created a wonderful piece of art called I Agree by printing out the standard terms and conditions from several popular social media platforms onto rolls of coloured paper. The idea was to highlight the overwhelming nature of the terms and that it is extremely unlikely that any user will ever have read them.
Image by Dima Yarovinsky
In case you're wondering, the 'winner' was Instagram, whose terms and conditions stretched to 17,161 words, which Dima estimates would take a whopping 86 minutes to read.
His art gave me an idea. Why not apply the same logic to company policies that employees are expected to read and use the reading time metric as a feature instead of something you keep secret? Just as many blogs — including this one — publish 'estimated reading time' at the top (4 mins) why not integrate it into company policies? After all, unlike social media terms and conditions, policies are — at least, in theory — designed to actually be read by employees.
That idea required policy owners to disclose the estimated reading time — in a large font, or ideally in a highly visible icon — at the top of the policy and in brackets whenever the policy is referred to. In simple terms, turning "67. Policy on Conflicts Of Interest" into "67. Policy on Conflicts Of Interest (15 mins)".
As is often the case, disclosure changes behaviour and the clients that I have helped to implement this have seen several benefits:
It gives users a sense of how long they’ll need to read the policy. Telling us how long something is likely to take helps us feel in control. That's why transport operators tell us how long we’re going to have to wait for the next train — we're more relaxed about waiting for longer if we know when the next train is coming. When it comes to reading policies, letting someone know how long they'll need to dedicate to reading — particularly when presented in digital form — helps them plan when they'll look at it. It’s not only a nice thing to do that makes policies more user friendly, but it also sends a signal that you're respectful of — or at the very least, aware of the demands on — their time.
Policy owners are reminded that the document they've written has a target audience who are going to need to spend time reading it. Time that comes at a cost to the organisation yet is often invisible to the policy owner. The small act of being forced to calculate and disclose the reading time helps make this more salient. This isn’t about specifying an optimal policy length but rather nudging the policy owner to think about whether the time they're asking their readers to spend is appropriate.
Collecting the data across all policies provides valuable insights into the efficacy of the policy framework. It permits the asking of questions like: Are there particular policies that are an unusual length? If so, why? Is there a correlation between policies that are breached and the length of those policies? Are policies that are long not being read? Or are policies that are short, not detailed enough? By adding up the reading time of all policies — or even just the bare minimum ones that employees are expected to have read — gives some sense of how much time, and therefore money, the organisation is spending on the exercise. Whether that's a good investment is then a separate question.
How do I do that?
The first thing to remember is that you don't need to be 100% accurate. After all, not everyone reads at the same speed. That's why it's an 'estimated' reading time. You just need to be consistent in how you approach it. With that said, here are five ways you can obtain that estimate:
Word Count - the simplest way is to count the number of words. Top tip: you don't have to do this manually — your document editing software can do it for you! The average person reads at a speed of 200 - 250 words per minute. So counting the number of words in the policy and dividing by 200 (let's be conservative here) is a fairly safe proxy for reading time. Of course, it takes less time to read the word 'no' than it does to read a word like 'inconsequential'. But, these things tend to average themselves out.
Human reader - the second way you can determine how long it takes to read a policy is to ask someone to read it. The important thing is that this should be done by someone independent, who is not a specialist in the subject the policy covers. Obviously, that requires someone to spend time reading it — but if you're publishing policies that are designed to be read, then arguably that should be happening anyway. And, of course, you'll be using different 'proof readers' across policies, which will lead to some disparity in measurement. But since we're looking for indicative trends, not 100% accurate data, it needn't concern us too much!
Specialist software - if you want something a little more accurate, then there is software available to do the hard work for you. I highly recommend Grammarly, a cloud-based 'writing assistant' that provides not only far more accurate estimates of reading time but also includes additional insights like a 'readability score'. Alternative options are summarised in this review. Do also be aware that Word and Google Docs also have plugins that can do some, but by no means all, of these things.
So there you have it. A very simple and cost-free idea that can help make your policies, handbooks and other documents that much more effective.
If you liked this post, and are interested in learning how to apply other Behavioural Science-led techniques in your Ethics or Compliance program, then do have a look at our training and coaching offerings.