I’m a massive data nerd. I love spreadsheets. I love experimenting. MASSIVE GEEK.
One of the aspects of digital that I enjoy is being able to see what is working (or not) and tweak things.
These are some of my quick and dirty ways to measure how things are going. Some are specific to particular tools, others apply across a variety of things.
1. Use different collectors in SurveyMonkey
We often use SurveyMonkey for surveys and I try to set up different collectors for different channels, e.g. social media, email to all students, our newsletter, website homepage, url on banner.
This means I can see which promotional channels are working well (so we can focus on those) and which are not (so we can abandon those or try to boost them).
2. Use custom campaigns in Google Analytics
Google Analytics allows you to add information to the end of urls so you can track them via Google Analytics. These are custom campaigns.
I’ve talked before about how I use information to help make decisions including examples of using custom campaigns to track our social media activity.
This works offline too. You can create a long link with the tracking information in it then use a link shortener like bit.ly or your own tools to create shorter, more attractive links to put on printed materials.
3. Track links using tools like bit.ly
Bit.ly and other tools like Hootsuite’s ow.ly shorten your links so they aren’t so long and rambly. They also allow you to track how many people have clicked them.
This is useful as it is, you can see how popular what you’re sharing is.
You can step it up a notch though and use it as a form of split-testing on social media. Try posting tweets on the same topic at different times of the day, with and without images, long and short text or any other combination of factors. If your aim is to get clickthroughs you can use bit.ly to see which format(s) work best.
4. Make use of split-testing where available
If your tools allow for split-testing DO IT! I’ve used it a few times for our email newsletter to test sign-up forms, email bodies and subject lines.
Changing the subject line of our ‘vote now’ elections email resulted in twice as many votes at one point (though over time the gap between the two fell a little). You can see the tests I ran towards the end of my ‘data driven decision-making’ slide deck.
What else have I forgotten? What have you tried? Please let me know. Like I said, I’m a massive nerd for this sort of stuff