Indirect Contributions: The Impact of Mobile Display Ad Impressions on Conversions

To measure the indirect effect of mobile display ad impressions on conversions, Japanese company Leverages used Google Tag Manager to separate visitors into two remarketing lists. As a result, Leverages confirmed that mobile display ads help boost conversion rate.


  • Measure the indirect conversion effect of mobile display ads
  • Determine an ideal media investment mix based on results


  • Used Google Tag Manager to separate visitors into two homogenous remarketing lists, one with normal display banner ads and the other with dummy ones to measure indirect effect
  • Didn’t count users clicking on either ad type; used third-party ad servers to learn the value of an ad’s impression itself


Indirect effect confirmed; the conversion rate for users seeing the normal ad was higher than the dummy one:

  • Organic search conversion rate grew 127%
  • Other route conversion rate grew 34%


As the population ages in Japan, demand for medical services continues to increase. Leverages Co., Ltd., the operator of prominent career website “Kango-no-Shigoto” for nursing professionals, solves staff shortages in hospitals, clinics, and nursing homes by connecting employers and employees through its services. The company produces various marketing initiatives, including video, display, and search ads, to encourage nurses to sign up on its career websites. Pursuing conversion efficiency, Leverages sought to understand contribution of each medium as its next step. Having invested in display ads, Leverages wanted to determine the indirect effect of mobile display ads on the increasing number of nurses using smartphones. The company’s goal was to understand not only how it could improve direct conversions from mobile display ads but also how mobile display ads contribute to conversions indirectly, ultimately enabling it to determine an ideal investment mix.


To determine the indirect effect of mobile display ads, Leverages had to create two homogenous target user groups, so the company divided its website visitors into two Remarketing lists through Google Tag Manager. The company also needed to determine the indirect effects of the mobile display ads from all the conversions, as some users converted naturally without being influenced by ads. To do so, the company put its normal “Kango-no-Shigoto” display ads into one Remarketing list while putting unrelated (dummy) ads into the other Remarketing list to extract indirect effects. In this way, the company compared conversion rates of users who encountered display ads (either normal or dummy) but didn’t click on them. The conversion rate was logged for each of the routes used to arrive at the website (Figure 1) so that the company could determine the indirect effect of display ads on conversions. Users who clicked on the display ads were not counted.

(Figure 1) Conceptual Diagram


The results confirmed the indirect effect of mobile display ads. Compared to the users seeing the dummy ads, the users seeing the normal ads had a 127% higher conversion rate. In addition, users coming via another pathway also had a 34% higher conversion rate (Figure 2). Though Leverages previously evaluated Remarketing ads by cost-per-conversion on a last-click basis, this study showed that display ads have a positive impact on the conversion rate of other inflow pathways as well. When combining the effect of direct and indirect display ads based on expected sales value as calculated through this study, the return on advertising spend (ROAS) was 236% (Figure 3). Based on the results, Leverages significantly increased its investment in display ads.

(Figure 2) Study results
(Figure 3) Return on Advertising Spend Including Indirect Effects

Future prospects

“We used to use KPIs that ignored indirect effects. However, we can now visualize the indirect effect of display ads on increasing conversions thanks to this study. This has significantly changed our investment strategy. In the future, we’ll take the indirect effect into account for new areas such as video ads to further optimize our ad strategy,” says Kosuke Masui, manager, media systems department.

“We have been evaluating each medium’s return on advertising spend based on last-click conversions. This type of evaluation tends to portray pull-advertising, such as search ads, as more efficient than push-style display ads, but we now see that continuing down this road will eventually stop our growth. This study has shown us that even short-term banner ad exposure is effective in getting users to recall our services,” says Mitsuru Tanaka, marketing manager.

“We were concerned that ads with many impressions were taking the credit for conversions that actually originated with other ads, and we assumed that the post-impression effect would show up in the conversion rate. This study confirmed that assumption—the numbers showed that display ads influenced users. I’m confident that these figures will allow us to expand our budget for display ads, enabling us to start new initiatives,” says Hisamitsu Tanahashi, manager, display advertising operations.

We concerned that ads with many impressions were taking the credit for conversions that other ads actually made ... This study confirmed the assumption—the numbers showed that display ads influenced users. —Hisamitsu Tanahashi, Manager, Display Advertising Operations

Think w/Google APAC on LinkedIn

Join a community of digital thought leaders and be inspired.