Monday Marketing Minute
Sometimes it tempting to think that a brand spanking new marketing strategy is going to transform sales, and in some cases this happens. Absolutely.
Businesses occasionally place big bets on a new approach. The difficulty is when it doesn't work and you've spent your entire marketing budget (or more) on it. Then you're in a tough spot - do you stop all marketing, or go back to what you were doing before?
A better approach might be to test things at a relatively low level. As long as you're getting your marketing out to a big enough sample for the results to be statistically significant, this can work well. Things then either work or they don't, and you have the ability to ramp up the things that seem to be working and cancel the things that don't.
The key is to test and let your audience and customers tell you what’s working, not go in with preconceived ideas of what will and will not work.
You can test a lot of stuff. Digital marketing is reasonably easy to test, in fact both Google Ads and Facebook ads are set up so you can test all sorts of messages against each other (Google has in fact built in some fairly sophisticated testing options which were covered briefly here).
But there are plenty of ways to test other - offline - marketing. Readers Digest used to send test sales letters out alongside their main letter (called 'the control') and got ever increasing response rates and sales (read more here).
There are also online tools which allow you to serve different web pages up to visitors to see which version gets better results. And if you want people to contact you by phone or email, you can insert a special offer code, or give a different phone number or email address to measure results.
A word of caution - if you want meaningful results, it's hard to test more than one variable at a time. If you change three or four different things at once, you'll never know which one thing made the crucial difference. Testing needs to be a rigorous process, but you'll be rewarded for doing it right.