Experimental Sales

You may not be aware of it but most of the big websites such as Amazon regularly run experiments on you.

You (and some others) may see a web page which is different to the majority. What's happening is that you are part of an experiment where they test your behaviour compared to the majority.  It's kind of like evolution. If the experiment is successful (I guess their normal measure is you buy more) then the experiment is adopted across the site.  One of Amazon's most successful experiments was "Customers who bought this also bought this".

I really like the concept of running low risk experiments - it drives innovation.  Running these experiments on a web site with millions of visitors is pretty straight forward since it's possible to determine what is normal behaviour for a group and the control group is large enough to be statistically relevant and show how the change deviates from normal behaviour. 

In  sales, decision makers are getting tougher to reach and sales practices which worked 10 years ago may not be effective today. Without evolution, sales will surely die.  I love the quotation "Doing the same thing and expecting a different result is the definition of insanity". If the sales approach isnt delivering results, why continue to do it the same way? Enter experimentation.

Market Footprint sells complex high value technology to business customers. It is not a high volume sales model, so how can we apply the experimentation mindset in this context?

If you have ever worked with six sigma, one of the sampling methods is to take a small sample of 100 things. Given 100 is not a statistically significant sample size, I am always fascinated to find that even in a small sample it is possible to gain insights in behaviour.

When selling high tech, a sample of 100,000 is extremely unlikely whereas a sample of 100 is perfectly feasible.

One area where I experiment is email.  Finding mobile numbers for people you want to speak with is tricky whereas email addresses are pretty easy to find. Sending introductory emails is therefore a routine task.

And here the lab experiment begins. I regularly tweak the wording to see if I get a better response rate. I might send the same email to 10 targets and a different email to another 10 targets and see what response rate is for each group. One email might be very clear about why I want to contact them whereas the other email might be very vague and talk solely about benefits. 

I find even with small samples there is a difference is response rates. Just like Thomas Edison, I like to continue experimenting - as a result  "I know hundreds of selling techniques that dont work".

I would be very interested to hear what other sales experiments people run - feel free to comment.