Which one is the king of direct ROI – Google or Facebook? Critical thoughts about Facebook’s view-attribution model

Display channels, such as Google Display Network, are using view-attribution. This means that sales are attributed for display ads even though customers have not clicked ads but only seen them before making a purchase. By default Google AdWords uses click-attribution, which means that sales are attributed for search ads only when people have clicked on these ads before making a purchase. Facebook’s default attribution model includes both of these. It attributes sales for ads when users have clicked on ad or seen an ad (see below image).

Are view-conversions justified? If somebody is browsing on Facebook and then an ad is visible on the screen for three seconds, is it justified to say that this have affected for a purchase decision. This goes deep in human psychology, which is not my area of expertise. However, one thing that I can say is that clicking an ad indicates more interest towards a brand than only seeing an ad. A click is an action whereas a seeing an ad is not concrete action, and we cannot even be sure if a potential customer even noticed that specific ad.

Can more budget be allocated for Facebook ads in a case where majority of conversions are view-conversions? I have been facing this question, and that is truly a interesting topic. Now lets dig into that question.

Facebook beats display ads, but then comes Google search ads…

Display ads were using view-attribution long before Facebook paid ads game in. Display ads are generating a lot of visibility, few clicks, and almost not sales at all – even though view-attribution is used. Display ad campaigns rarely generate direct ROI, at least from my own experience. There are place for display ads, and that is for campaigns building brand awareness, of which goals are increasing sales in a long-term.

I am not saying that when Facebook uses same view-attribution elements than display, it means that Facebook is not up for direct ROI campaigns. In fact, Facebook is effective for generating sales both view-conversions and click-conversions. Ratio between these variates. From my experience it is clear that Facebook ads beat display ads in direct ROI.

With Google AdWords, this comparison is more complicated because both Facebook and Google search ads are good for generating direct sales. The problem is that these channels are using different attribution models. Is it justified to increase Facebook’s budget over Google AdWords’ budget if Facebook is using different attribution model than Google?

One argument is that this comparison need to be made equal by using same attribution window between these channels. This would mean that view-conversions should be excluded on Facebook’s results. That is not the best way because attribution model need to be decided based on channel’s role in the purchasing funnel. Facebook and display networks are visuals channels, and visual ads are affecting people’s minds and leading them towards purchasing decisions. In comparison to that, Google search ads are not visual but they are only text ads. Text ads are not that visual and for that reason only click-attribution is used for Google search ads.

One possibility is to adjust Facebook view-conversions’ weight, for example, only 50 % of view-conversions are allocated as sales for Facebook ads.

It is giving a wrong picture about Facebook’s potential if all view-conversions are excluded. One possibility is to adjust Facebook view-conversions’ weight. This means that only certain percentage of view-conversions are allocated for Facebook ads. This percentage need to be decided based on the value you give to these view-conversions, for example, how often Facebook is part of the purchasing process. Sadly, Facebook is not supporting this features, and this analysis means more manually work when comparing results of Google and Facebook.

I put my trust in Facebook’s view-conversions

When creating Facebook campaigns, different technical choices are affecting how much you can trust in view-conversion.

news feed vs. right column placement

Right column ads are much smaller and they are not located directly in the news feed if compared to news feed ads. For this reason, right column ads are much likely to be affected by banner blindness. This means that potential customer do not notice an ad because too many things are demanding ones attention on the screen. For example, when browsing Facebook’s news feed, you may not remember those small ads on the upper right corner if compared to the larger ads on the news feed. For this reason, I put more value for view-conversions in a case of news feed ads.

retargeting current customers vs. acquiring new customers

Secondly, view-conversion are more biased in the case of retargeting ads. People who have visited your site are more likely to buy without any advertising compare to those who are not your current customers yet. When people see an retargeting ad on Facebook, they might have made this purchase without seeing that Facebook ad.

Still, I strongly recommended for using Facebook retargeting even though it would generate majority of view-through conversions. Here is one way to look at this. Facebook retargeting view-conversions are similar compared to the advertising with Google brand keywords. It is mostly recommended for advertising with brand keywords in Google even though people might have ended up to your site through organic results. Reason for this is that there are so many distractions on the internet. Even though potential customers would Google your brand, they might end up for competitor’s site.

Same logic goes for Facebook retargeting in a case of view-through conversions. When people are browsing on Facebook, they might have already decided to buy your brand’s product. Still, they mind end up for buying competitor’s product because competitor is advertising to your potential customers on Facebook with a nice discount. Facebook retargeting is like closing a hot deal whether it generates click-through conversions or view-through conversions.

If you encounter a situations where your Facebook retargeting campaign only generates view-conversions, then what?

In situations like mentioned in above quote, it is wise to question the value of these view-conversions. You can try to exclude right column and advertise only on news feed, and see is there any difference on results. If excluding right column does not affect the results, you can examine the total revenue numbers of your eCommerce store. You can compare how total revenue has developed compared to your Facebook sales. You can also try to changes your ads’ message to be more clickable. If you end up transferring your Facebook budget to other channels by excluding view-conversions, how does that affect your total sales?


It depends on many variables on which percentage of your Facebook sales are click-conversions or view-conversions. Your product, your target groups, and your creative are affecting how people are reacted for your Facebook ads whether they click on it or go to Google after seeing your Facebook ad to look for more informations before a purchasing decision.

Facebook advertising is effective tool for generating direct sales as well as for brand visibility. With Google search ads you are not building brand with visual elements but making direct sales for your eCommerce. When optimising your budget, it all starts with goals. Are you optimising solely on short-term ROI, building a brand, or leaving room for early stage experiments for figuring out your customers segment.

If optimising only for direct ROI, then it comes down for attribution modelling. Both channels, Google and Facebook, have their place in customers’ purchasing path. If transferring all Facebook budget to Google, because Google is better with absolute click-attribution, it might end-up lowering your overall sales. This is because customers’ need multiple touch points with your brand before making a purchase. For example, Google Analytics offers free tool for analysing customers paths (see image below).

When deciding whether to trust on Facebook’s view-attribution or not, these four points are wise to consider.

  1. right column view-conversions are less valuable compared to news feed ones
  2. view-through conversions are more valuable in a case of new customers compared to retargeting
  3. Facebook’s view-conversions’ weighting can be adjusted, for example, only 50 % of the view-through conversions can be allocated for Facebook ads
  4. due to Facebook’s visual nature, it is justified to use Facebook’s view-attribution along with click-attribution

I would like to know, how do you use Facebook’s view-through conversions when doing digital marketing? Do you agree with my points or do you have different angle to this discussion?


Thank you Joni Salminen for inspiring to write this article. We have had a nice discussions about different attributions models and how to allocate budget between different marketing channels. #viewattributiongate

Article has been published originally on Parcero’s blog.

– Tommi