The value of A/B Testing
A true story… Down the hall from our office is the sales & marketing office for a construction company that specializes in steel buildings. I’ve gotten to be friends with the heads of sales there.
He was telling me about a new video he recently placed on his website’s home page. He had hired a professional spokesperson, written a friendly script and produced a very nice little welcome video. Immediately, site activity dropped and bounce rates jumped. Huh?!
Rather than remove the video – as would have been indicated by the numbers – he decided to try something… to replace the static picture on the video screen (that’s the image you see before you click the ‘play’ button). He replaced the image of the spokesperson with an image of one of his steel buildings. Site activity and time on site jumped right back up to normal levels. What?!
Logically, it’s hard to imagine that the image on a website’s video would have such a significant impact on website traffic, but “numbers don’t lie.”
So, what do steel building have to do with our industry? Nothing, of course… but the idea of paying attention to the details and being willing to test aspects of your marketing absolutely does.
And while most of us don’t have the time to focus on all of the details and testing all of the variations… in an effort to improve your marketing outcomes, the idea of doing some A/B testing is worth considering. For example…
- Emails: test subject lines, images, layout, click-thru buttons
- Landing pages (e.g. to download an eBook): headlines, copy, length of copy, fields on the download form, call-to-action buttons
- Advertisements: Headlines, imagery, copy, ad size & location, keywords (for PPC ads)
- Websites: placement & layout of content, call-to-action buttons, imagery
- Your content: look at the number of blog visitors, viewers of your LinkedIn posts, download of eBooks, etc. – understand what topics resonate best with the marketplace
But you can also conduct some non-traditional A/B tests:
- What about the ‘standard’ voice mail message that you leave? Test different ones to improve call-back rate.
- What about the elevator pitch you use at networking events? Try different ones to see which ones lead to the deeper conversations.
This concept isn’t just theory… it has real-world benefits. Imagine if you have an email database of 2,000 names and you’re able to increase the average open rate by just 2% (a very achievable increase). That means that 40 additional people (clients and prospective clients) will now be exposed to your marketing message. Not bad, huh?
How to get started? Easy… pick just one element of one tactic and set up a simple A/B test. Implement, track the results and see which option ‘wins.’ Then, test that winner against another option for the next test. Keep learning and getting better. When you’re ready, pick a second item to test… and go!
And now the rest of the story… my friend down the hall made one more change. He re-shot the video, this time replacing the female spokesperson with a male spokesperson (exact same script). Result? Complete views – those visitors who watched all the way to the end – doubled! Go figure!