Skip to main content

If you are like many businesses, Facebook is the social platform you believe you need to be on. You have spent years (and dollars) boosting posts, releasing Facebook ads, and racking up the likes on your business’s page. To determine the impact of your spending, you’ve observed reach, impressions, clicks… shall I go on? Well, Zuckerberg and crew say you are doing it all wrong. Facebook’s latest push is to convince advertisers that these metrics don’t matter. Conversion Lift and its randomized controlled trials do.

State of the Relationship: Facebook and Advertisers

By no means are advertisers fleeing Facebook’s advertising platform. Financially, this is represented by its stock increasing nearly thirty points over the past year. Revenue grew by 49% in the most recent quarter. Nevertheless, Facebook has experienced several public blunders over the past year, revealing flaws in the metrics it shares with advertisers, and losing trust in the process. In September, Facebook admitted that for years it inflated the average time viewers spent viewing video advertisements. Following this incident, the company found five more flaws in the metrics it reports.

With the rapid pace of innovation in the digital marketing arena, it’s hard enough to orchestrate campaigns and select advertising goals. You need to be confident that the metrics you are receiving are accurate. Metrics like how long a Facebook user watches your video can be an indicator of the success or failure of your campaign. Facebook’s response to this thinking is that these metrics don’t illustrate whether it’s worth advertising at all.

What is Conversion Lift?

In response to the above missteps and to convey its efforts of transparency, yesterday (May 24, 2017), Facebook held its very first “State of the (Measurement) Union”… (Maybe leave out the political puns, Zuck?). At its event, Facebook pushed the use of Lift.

Originally launched as a platform on Facebook in 2015, Conversion Lift is your typical stats experiment using randomized controlled trials (think elementary school Science Fair project). Lift breaks users into a test and a control group. In the test group Facebook sometimes shows users the ad, sometimes not. In the control group, the ad will never appear. After the test is complete, Facebook calculates the “lift”; the additional business you captured, by comparing conversions in the group that saw your ads and the one that didn’t.

Randomized Controlled Trials

Facebook believes the power of Lift over regular metrics is its ability to demonstrate how much additional business you capture due to your advertisement. Facebook hopes advertisers will use Lift for value added tactics like choosing between video or link ads, making informed budgeting decisions, and optimizing Facebook campaigns to be more effective. Lift also serves to fix the over-reliance on last-clicks. Lift’s appeal increases when you consider that experts agree randomized controlled trials are the best approach in measuring effectiveness. Importantly, few advertisers are currently using them.

At this time, only advertisers who work directly with Facebook teams will be able to set up conversion lift studies with their account representative. You also need to have Facebook Pixel set up to share your business’s conversion data with Facebook. Facebook is looking to expand its conversion lift offerings to cover more use-cases and more complex studies.

Lift in Action

To tout the benefits of the platform, Facebook published a Lift case study with General Motors. Last year, they tested to see whether Facebook ads were assisting in the sale of its OnStar data plans (WiFi in your car!!!). They launched a Lift study. The results? The Facebook ads had a 2.3% overall sales lift for the OnStar 4G LTE Wi-Fi data plans.

The Dilemma for Advertisers

Advertisers who were upset by Facebook’s previous misreporting of metrics have to “pick their poison”. Lift is clearly superior in measuring a clear return on investment, but transparency is still a problem. Only Facebook can run such experiments on Facebook, because only Facebook has a full view into and full control over its platform.


Business aren’t leaving Facebook any time soon, and they continue to spend advertising dollars on the platform. This is not in question. The problem is that advertisers that are still feeling stymied by Facebook are once again being asked to put their faith and trust in Facebook to “serve their food and ask how you like it”. Facebook believes it has the answer in its Conversion Lift platform to finally offer advertisers the return on investment data they have been craving since the beginning. Another slip up in its reporting, and Facebook and advertisers may have to change their relationship status to “it’s complicated”.