“What if the Vodafone pay monthly price plan is like jam?” I asked in my interview at Vodafone.
Before my interview, I’d read the brilliant Neuro Web Design by Susan Weinschenk who explains if you offer too much choice, people freeze’.
She cites an experiment where 2 stalls selling jam get different results due to the number of choices – the stall with 6 jars on display, sold to 30% of visitors but the one with 24 jars on display only sold to 3% of visitors.
The most important page on Vodafone’s website had a lot of choice. Vodafone’s core business and online journey is selling a phone with a pay monthly plan. In 2011 the Vodafone plan page showed 15 or more plans over 2 to 3 pages, with a grid of 6 columns, 126 elements and other visual noise.
After starting at Vodafone as a user experience manager, I got the sales data for every phone and every plan they sold and was struck by the pattern in the spreadsheet. Consistently ONE plan comprised ~70% of sales and 3-5 plans comprised ~90 – 95% of sales.
I put forward the hypothesis that ‘conversion will go up if we only show 3 to 4 plans comprising 90% of sales’ and proposed a series of AB/n and MVT tests, which went on the backlog.
Several months later, the sales team heard that the CEO Guy Laurence had told commercial product owners to ‘sort out the plans’. Sales, commercial product owners and I worked together to run the tests.
A series of tests June to July 2012 proved that showing less plans increased sales. Commercial product owners had the confidence to halve the number of plans and remove 18 month plans altogether in August 2012. As a result, online conversion increased significantly.