Back in the 1980's Jan Carlzon was trying to breathe new life into an ailing Scandinavian Air Services.  He was famous for saying "You cannot improve one thing by 1000% but you can improve 1000 little things by 1%".  Wind the clock forward 20 years and "Team GB" scooped a helmet-ful of gold medals in Beijing by following a similar principle.  According to David Brailsford, the British Cycling Performance Director, their success has come by way of the “aggregation of marginal gains”. 

So perhaps digital marketers should adopt the job title of "Performance Director" and we should set out to find, and improve, hundreds of different things to have a gold-medal-winning impact on our business.  The really good news is that, unlike elite cyclists we have loads of big and quick wins within our grasp.  So, rather than committing to a gruelling training regime, why not set out on a journey of digital optimisation?

Let's look at the micro and the macros to see where marginal gains make a big difference.  First up, driving traffic.  My thanks Mike Rogers of Optimize for this case study but it just shows how powerful a series of marginal gains can be in overall performance.  The beauty of Google Adwords is that everybody has just 95 characters to get their message across so those that use their characters wisely will enjoy the rewards.  Let's see how changing a few words drove click-through rates up...

The Aggregation of Marginal Search Gains delivers 15% more traffic

The real beauty of this case study is that Yell.com gets more traffic for less cost...the Google Quality Score in PPC, where you pay less for more successful ads, means that you end up with a lower cost per click and your ad gets shown more often meaning even more traffic.  Gold Medal Mike Rogers!

Having managed to drive more people to the site for less, lets look at the macro end of marginal gains.  Email marketing is still a land of sub-optimised marketing and with a little effort pulling the right levers you can make a massive impact on campaign performance.  Below is a scenario for an email marketing campaign.  Let's pretend we're a BtoB organisation and we want to generate 40 registrations for a seminar.  Our campaign metrics may look like this...

Not bad, but not good.  However, let's get the Performance Director in and work on each element: 

• Hard bounce rate...use email repair software to re-build basic errors (missing "@" sign, invalid Top Level Domains etc).  Moving forward, use validation scripts in your web registration form (see Flybe for "best in breed").
• Soft bounce rate, use Spam checking software, make sure you are working with your broadcast partner to identify possible Spam blocks at ISP or Corporate firewall level.  We will never get 100% deliverability, but we should be aiming for 90%.
• Open Rate.  Test "From Fields" with different names, use industry or region personalisation in the subject line and work on the appearance of the preview pane by getting exciting text in the first 6 lines not a great big fat banner that will be disabled in virtually all email clients.
• Click-through rate.  Through segmentation and cunning use of personalisation give the illusion that this email has been crated just for the recipient.  Similarly, work on your "persuasion architecture" and sell the benefits, provide clear cals to action and introduce pace, excitement, fear, uncertainty and doubt!
• Conversion rate.  Pre-populate the landing page or, better still, register them in one click by taking them to a confirmation page.  Certainly don't take them to an empty form with acres of fields to fill in.  Repeat the reasons to register and keep the pace to make sure people don't get cold feet at the final stage. 

So, we've sent out the same number of messages but we have generated 41 registrations, rather than 7.  All the improvements are well within "average" conversation rates that marketers enjoy (more on the danger of "averages" in the next post),  As a prissy direct marketer there is one critical issue to consider here.