Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.

Skip to 0 minutes and 1 secondWe looked at conceptually how we projected the test results from Arizona to the national level. Now let's put some numbers here. We start with the control state, California. So the number of stores in California that were the control stores was about 10% of the chain stores in California. So from there, we project that in that chain, about 1,420 units were sold in California. Next is to know that California represented about 12% of the national sales. So to do that next, what we need to do is take,

Skip to 0 minutes and 49 seconds1420 and divide by 12%. And that's going to give us the total retail sales for the entire nation within that chain, okay? So the next step is then to say this retailer represented 25% of the sales of Betty Spaghetty across the entire nation. So what's the next step? We're going to take 11,833 / 25% and that's going to be the national units sold. So 11,833. Let's go back to the previous slide and see what was. So 11,833 was the national sales in that retailer and this retailer represented 25% of the national sales. So the national sales across all retailers was 11,833 over 25%.

Skip to 1 minute and 54 secondsThe next step is to then say that the test period, the weeks in June and July when the TV ads were run, was 5.5% of the annual sales. So the annual sales expected without the TV ads. Remember, we're doing the control state projection here. So the annual sales is obtained by taking 47,333 by 5.5%.

Skip to 2 minutes and 32 secondsNow, the key thing to remember here is the test was run during June and July. The national campaign is going to run during the holiday season. So what we have to do is take the annual sales number and look at what percent of annual sales are going to happen during the holiday season. Just like I just said, Betty Spaghetty also has a very seasonal pattern in sales. And that's true for most toys. A large percent of the toy sales happens during the holiday season. Now, for Betty Spaghetty that was about 45%. That's a large number, right? So the holiday units sold without TV ads is 387,273. How do we get that? You take 860,606 times 45%, okay?

Skip to 3 minutes and 40 secondsSo this is what we call the base sales. Sales without TV ads. Now from the very experiment. Because we had test and control, we have some level of causal estimate that about 267% lift can be expected, because of the TV ads, because of marketing. So the lift was about 267%. So the number of units that we expect to sell during the holiday season. If we run the national campaign, it's about 1 million. And that, we get from 387,273 x 267%. This is the expected sales of Betty Spaghetty during the holiday season if you run the national campaign. The keyword here is expected and holiday season. Okay, now what is the break even units?

Skip to 5 minutes and 0 secondsSo we have to compare the expected sales with the break even units. So what is that? So we had already calculated that. That was about 538,000. So the expected units is about 1.03 million. So should Ohio Art run the national campaign? How do you know that? So we know that expected sales is higher than breaking even units. But there is always a but, the test was run in June and July. The national campaign is running during the holiday season, right? So think about that. What are the implications? Should Betty do this campaign? Should Ohio Art have a national campaign during the holiday season for Betty Spaghetty? Think about that and we'll be back.

Calculating Projected Lift: Betty Spaghetty

Learn how to calculate projected lift using the Betty Spaghetty experiment results.

Share this video:

This video is from the free online course:

Marketing Analytics

Darden School of Business, University of Virginia

Contact FutureLearn for Support