Data-driven attribution gives GÖSAŞ a 10% conversion boost at 5% less cost

March 2018

As the credit card portfolio of Garanti Bank, GÖSAŞ believed that investing in non-brand keywords and mobile ads would drive incremental leads. However, their last-click attribution model suggested that these approaches couldn’t achieve good a good cost per acquisition. To overcome this disconnect, the GÖSAŞ marketinģ team and their agency Hype were looking for a model that would show the effect of each touchpoint on overall conversions.


Garanti Ödeme Sistemleri AŞ

Credit card portfolio of Garanti Bank

Launched: 1999

Headquarters: Istanbul, Turkey


Increasing credit card sales

Accurately quantify effect of investing in non-brand keywords and mobile ads


Implemented data-driven attribution modelling


Drove 10% more conversions at 5% lower cost

Cost per conversion decreased by 13%

Conversion rate increased by 17%

“Both our data and recent studies were signalling that there’s a considerable amount of conversions starting on mobile and ending on desktop”, explains Seckin Caglin, EVP of Marketing & Digital. “However, with the last-click model we were not able to track the exact conversion path for the cross-device activity. This fact was affecting all of our measuring and optimisation process.”

While GÖSAŞ considered time decay, position-based and first-click models, they found these all used predetermined coefficients for the touchpoints, which were likely to vary over the time. On the other hand, they believed that data-driven attribution modelling would suit them better and be much more reliable because it would take their own data into consideration. To activate the model, they simply changed the conversion setting in AdWords.

“With the help of data-driven attribution, we started to manage our investments more accurately.”

– Seckin Caglin, EVP of Marketing & Digital, Garanti Payment Systems

“Thanks to data-driven attribution modelling, we became able to monitor the added value of non-branded keywords and to better understand cross-channel conversions paths”, Seckin says. “We’re now more confident in investing in those keywords and in mobile. With the help of data-driven attribution, we started to manage our investments more accurately.”

Using the new model, GÖSAŞ gained 10% more conversions with a 5% reduction in cost. The cost per conversion decreased by 13% and the conversion rate increased by 17%. Looking to the future, the team is keen to further their efforts in attribution. “With these positive results, we’re now considering Google Attribution 360 products to better understand not only the impact of multiple devices and non-branded keywords, but also the impact of all our marketing channels.”

Inside Google Marketing: How we (finally) proved the value of influencer marketing