Your telemarketing project seems to have gotten off to a slow start. The expected conversion rate was 5% and you’re only getting 3%. Your boss (or client) looks like he’s getting ready to hit the panic button. Heck, you’ve already hit the panic button yourself.

OK, now what?

You do the first thing that comes to mind: you re-write the telemarketing script.

Your agents are back on the phones, and, what do you know? They’re getting a 5% conversion rate. The new script must be much better. You saved the day… or did you?

The last thing your boss or client wants to hear when things aren’t going well is that you want to keep doing what you’re doing. After all, isn’t the definition of insanity “always doing the same thing over and over again and expecting different results”[Einstein]?

Well, yes and no.

If we’re talking about banging your head against the wall, then even just once is insanity.

In the world of marketing, though, head-banging is how we make progress. In fact, early success is often the worst thing that could happen for, “mediocrity is the enemy of excellence.”

Your first mistake was not running several telemarketing scripts against each other in an A/B split test. But, that’s another subject.

You’ve compounded the problem by not taking into consideration variance and statistical significance. And, given how you’ve handled the situation up until now, you’re most likely going to make the further error of putting your project on autopilot now that you’ve hit the 5% conversion rate you were looking for.

Statistical Significance

Don’t worry, I’m not going to bore you with math. Most introductory texts on statistics will teach you what you need to know.

Suffice it to say that knowing in advance the conversion rate you expect (want) to achieve, with a little Excel wizardry you can calculate how many conversations your telemarketing agents need to complete before you can say with confidence that your telemarketing script is going to work or needs a re-write.

Let’s say, in this scenario, that had you continued to test the first script until reaching statistical significance, you would have found that it actually generated an 8% conversion rate for you.

What you need to realize is that that’s a 60% improvement in relative performance. Sure, an absolute 3% improvement doesn’t seem like much. But, if you look at the bottom line, this could mean a jump in ROI for you from 30% to 90% or more.

Now that you’ve happily settled for 5%, you’ll never realize how much more profitable your campaign could be.

So, here’s the takeaway from this post:

1. Calculate in advance how many conversations need to be completed to test your telemarketing scripts
2. Start by running an A/B split test
3. Always continue to challenge your winning telemarketing scripts with new split test ideas
4. Don’t panic!

If you would like me to help you with this calculation, email me or reply to this post.

-Gene