Friday, March 21, 2008

Online and offline campaign tracking

This is an excellent case story about online and offline campaign tracking with one of our customers. My second post in history was about SEA LIFE Helsinki and how they tracked their visitors. Now we went little deeper in web analytics, or should I say, optimization of marketing. Sea Life Helsinki opened the winter show over a month ago and we pushed measuring into this project. The campaign included two basic ideas: a pike fishing story competition and 2 in 1 special offer coupon to bring customers at their slowest time, wednesday evenings.

SEA LIFE Helsinki decided to use both offline and online advertising to promote the winter show in aquarium. They had on outdoor advertisement in place for one week and in addition, a magazine article and a press release. They also send a newsletter and we ran a small Google Adwords campaign just to test how online ads worked versus offline advertising. So, we used IndexTools web analytics system for measuring traffic from these different sources. And actually we measured much more than just clicks (click the picture below for getting a better view of the table).

We counted or evaluated impressions, and by doing that, we got either exact (newsletter) or approximate (article based on circulation) click through rate (CTR). We counted also cost of each channel, so we got also a KPI called cost per click (CPC). We measured actions, which in this case were amount of coupon prints, and we got another KPI called cost per action (CPA). And finally, we counted conversion rate, actions divided by clicks (visitors). Unfortunately, if adult entrance fee is 13,50 €, the cost per action is too high in general.

But, we have to remember one important thing: we didn't use campaign specific domain name in outdoor advertising, just sub address It's not absolute truth that outdoor advertising didn't work at all because people are lazy or just don't remember the whole address and they type just . Anyway, the table above includes very much valuable information for marketing and especially next campaign planning.

SEA LIFE has also other website goals what we measure. For example at the same timeframe they got 10 request of offers and one of them came from campaign newsletter. You should propably count this in, especially if it turned out to sales. We also found that newsletters were the most effective way in this particular marketing mix. But at the same time total newsletter subsciption conversion rate was only 0,04 %. Based on this data, one development idea is that we have to get more newsletter subscriptions and we have to bring the sign up form more visible. This is clear suggestion for action and what the web analytics is all about.

This is not all. As you can see, there's a bar code in the coupon. SEA LIFE actually counted customers and sales from the campaign. When customers arrived and gave the coupon, SEA LIFE's cash system was able to read in the bar code. After a month there was 89 returned coupons generating approximately 1.400 € turnover. We have to wait until the end of April, when campaign ends, to get final results. If you look the return on ad spent (ROAS, total revenue divided by total costs) at the moment - it's pretty poor, only 34 %.

The best thing is that we can learn great things from this project, such as which are the most effective channels and what is the best marketing mix in general. SEA LIFE also has to compare their whole turnover e.g. last year's or last month's turnover because every printed coupon is not returned but the visitors are still converting to customers. There are so many variables and this makes web analytics and marketing optimization so challenging. It's not easy and that's why I like it a lot.

1 comment:

  1. At the end of the campaign, total number of returned coupuns was 231, which generated turnover of 3.118,50 €. Return on ad spent was approximately 76 %, giving a negative emphasis to whole campaign. If we would have dropped the outdoor advertising from the campaign, the ROAS would be 1058 %, which means that money spent on advertising would come back more than ten times. This probably would have given a positive return on investment to the campaign too.