Case Study: SAZKAmobil
Banner ads don't work... Or do they?
Banner ads don't work... Or do they?
Situation
SAZKAmobil is the largest Czech virtual mobile carrier. We were approached with a brief geared towards communicating a 50% discount on their pre-paid plans.
Initially, we'd thought it was going to be a piece of cake - crank up spend in search, communicate the huge discount, and that's it. Even a monkey can do that, right? Well, not really.
To make the starting point even more fun, we knew that PPC channels were approaching their limit of efficiency. Thus, any new sign-ups would be disproportionately more expensive, making further investment into search unprofitable in the long term.
Objectives
There was only one objective - to sell more pre-paid plans in the next four months than in the last four months.
Activities
Instead of trying to squeeze more out of a potentially unprofitable channel or optimising others close to conversion, we went the other direction. Hello, banner ads!
At this point, you must be thinking we'd gone mad. Purely sales-based objective, other channels available for optimisation, highly competitive market. Who in their right mind would want to invest money into ads that show ostensibly low conversion rates and high CPA in Google Analytics?
There's method to our madness. We borrowed the theory of mass marketing from Ehrenberg Bass Institute and overlaid it with just a tiny bit of segmentation. Our assumption was that if we show our ads to enough people, we will pre-qualify them to buy and maybe spark interest in the SAZKAmobil brand.
To make the most out of the media mix, we complemented directly purchased media space with RTB. And we went for it - DMP data from ad exchanges, private deals, multiple retargeting scenarios, high-visibility placements, native advertising - the whole megillah.
Results
So, one day we open our reporting dashboard and find out that RTB campaigns' CPA on the product that costs 2.76 EUR is over 330 EUR. Not a great score, right? Every rational marketer would've had shut down all the campaigns at this point. We stayed the course.
The CPA went down considerably by the end of the campaign, but it still seemed like a fiasco. That is until we looked at other channels.
Sales from direct visits increased by 213.76%. That orange curve is the time period before the campaign launched.
Organic traffic from Google was no slouch either. It delivered a 122.75% increase in sales compared to the pre-campaign period. At the time, there were no SEO outreach-based activities running - just the campaign.
Considering the overall campaign performance, there was an undeniable lift in Conversion Rate, Goal Completions and Conversion Value. Sales went up by 470.77%. More users had been visiting the site before the campaign started, but they didn't buy anything. When the campaign ended, we had a huge stack of orders.
You can see that some of the data is missing from the screen shots. That's because we don't like to be sued for breach of confidentiality.
Based on these results, we have verified that only performance marketing doesn't save the day. The short cannot thrive without the long.
Did the campaign make money? Not in the short term, but the Customer Lifetime Value metric is looking mighty fine. And yes, SAZKAmobil is one of those clients that actually knows what CLV is and is not afraid to use it.
"It took a lot of hard work to step out of the performance marketing mindset. I'm glad we did it, though. On-line branding initiatives are now firmly planted in our marketing plan alongside TV and POS outreach."
"And as for ROI... He who laughs last, laughs best; and I think that our C-suite is already beginning to smile."
Radim Voňavka, Marketing Specialist, SAZKAmobil