The Non-Obvious Insights Blog. Non-Obvious Insights
The Non-Obvious Insights Blog.

Dedicated To Helping Readers
Be More Interesting
Since 2004.

As Featured In:

Accidental Measurement And Why We Love Useless Metrics

I'm not a measurement geek by any stretch of the imagination. As much as I love a good statistic, I'd put myself in the same category as most marketers … we realize that measurement is important, but we wish it were easier to grasp and involved less guesswork and questionable assumptions. There are many smart people talking about measurement, and we have several on our team working on a much more sophisticated model for things that have traditionally been difficult to measure such as word of mouth impact and social media engagement.

The real challenge is that measurement online has become an exercise in silliness for many marketing teams. They report on things that don't matter about behaviours that tell you little to nothing about what is actually happening online. Why? There are three core reasons your metrics might suck: because you're just measuring what you measured last year, you're focusing on just finding "one big number" to report or because you just don't know how to measure better.

Luckily, the solution for any of these reasons is increasing your knowledge about what you should measure, and what you should avoid. To help get you started, here is a list of five useless metrics that many marketers use:

  • Accidental Impressions – this may be the most favorite measure that marketers like – counting impressions for ads or sites no matter how long someone stays or if they click it or not. Counting impressions without any context is like the amazing color changing card trick – you're watching the wrong thing without realizing it.
  • Accidental Time Spent – time that a user spends on your site looking for what they want to click on, but unable to find it. This is not a sign of engagement (as commonly assumed) but rather a sign that your design or layout is not usable enough.
  • Accidental Email Open Rate – if you use Outlook, you know that many emails are automatically opened as soon as your cursor hits them. Most email marketers count this as an "open" even though you may have opened the email by accident. If a large number of people getting your email are on Outlook, for example, that high open rate may not actually be an indication that they found the content of your message appealing.
  • Accidental Clicks – who hasn't experienced that annoying banner that pops up at the least convenient time? Sometimes finding the button to close it is so difficult that you end up clicking on the ad accidentally. If you are running a campaign where you are counting these desperate clicks to make you disappear as clicks and conversions, you're getting a skewed view of effectiveness.
  • Accidental "Add to Cart" – a favorite tactic of many online sites is to have every uncertain click on a product automatically add it to a cart. Other sites make you add a product to your cart before seeing the price. If you are tracking your shopping carts and conversion, make sure you're not skewing the numbers with people adding a product they were never thinking about buying.

So what should you be measuring? Unfortunately, there is no single answer as it depends on your goals, but to get smarter about metrics online, one site you should definitely check out is the blog Occam's Razor from Avinash Kaushik. He also has an equally brilliant book called Web Analytics: An Hour A Day that I highly recommend you pick up.

I've been working my way through it since meeting him at a presentation and hearing him speak. His acronym H.I.T.S (How Idiots Track Success) remains one of the more inspired acronyms and one-liners I've heard in a presentation …

5 thoughts on “Accidental Measurement And Why We Love Useless Metrics”

  1. While you do bring up some interesting (and unique) points, I have to disagree with you….

    Some of these “accidental” examples could be remedied by combining data metrics. For instance, impressions that don’t matter do happen (a lot…especially if your stuff isn’t solving a problem or optimized for what the person is looking for) but combining it with entry and exit data, bounce rate and time spent on site, you can get a fairly broad picture of what the visitor was doing on your website.

    Time spent could mean that the visitor was truly interested in your stuff…it could also mean that they went to the bathroom after your page loaded too. Once again, by combining metrics, you can get a better idea of what is going on. I write long content so naturally I would expect longer time spent on my page. To state that someone is looking for a place to click on when they could just as easily hit the “back” button would be more common.

    I am sure “accidents” happen but not enough for it to skew data unrenderable.

    Reply
  2. Thanks for the recommended reads – and your list is a good indication of the un-intended consequences of measuring something simply because we can.

    Since ‘happy accident’ is still considered a rare occurrence in the real world, can we assume that measurements of accidents are really indicators of how well we have irritated our prospects? And, as you mention, when we measure & applaud an accident, then our designers and programmers will work towards ‘improving’ the statistic – further reducing usability. Overall a pretty negative outcome, even if we can prove somebody laid eyeballs on us.

    Reply
  3. Great post! I would add one of my favorite accidental metrics: extended time on site in a back window or back tab. In today’s multi-tasking crazy world, I want to see activity on site, not just time. Time is a presence. We would never measure “time of exposure” for an out of home ad. Doing so from a web perspective is nearly as obscure.

    Reply

Leave a Comment

The Non-Obvious Insights Newsletter. Non-Obvious Insights
Layer 97
The Non-Obvious Insights Newsletter
Layer 118

Skip the obvious and anticipate the future with our weekly newsletter. Join over 25,000 subscribers and start receiving the stories (and insights) you’ve been missing.

All Books

#1 WSJ & USAToday Bestselling Author

Rohit is the author of 8 books on trends, the future of business, building a more human brand with storytelling and how to create a more diverse and inclusive world.

Vector Smart Object

Contact

Have a Question or Inquiry?

Just fill out this form, and we’ll get back to you within 24 hours!

About You

What Are You Contacting Us About*:

Your Message