Live and Direct (Almost)
April 15th, 2017


I work on Market St. in San Francisco. It’s like the Broadway of NYC or the Congress Ave of Austin (I used to live in NYC and Austin). On my walks home from work I would see these similar-looking advertisements on bus stops, billboards and even buses, but I never fully understood what was going on. They were black and covered in text so I never really got to understand what they were about. One day I finally decided to take some time and read the ads, and to my consternation (that’s a new word I learned), I found that the ads were instructions on how to use Facebook’s Live feature!

In my limited field of view, I perceive Facebook Live’s popularity to be growing. However, I can’t necessarily attribute this perceived traction to good product-market fit or exorbitant marketing spend. The Facebook live ad campaign, created by their internal agency, The Factory, has made headlines in the advertising community for its bold, multi-platform strategy. I think the Facebook Live feature seems awesome, the video advertising efforts are phenomenal and the intent behind this campaign is great, but I feel that their static (bus stops, billboards, buses) ad strategy could have been greatly improved. Specifically, I claim that consumers cannot derive the intent of the static Facebook Live ads upon a casual glance. As such, in this analysis I sought out to collect data that would prove or disprove this claim. Note: I wanted this research to be rigorous so I could come to some categorical conclusions, but I felt there were many shortcomings in my approach, which I will discuss more in depth. I want to add that my analysis is largely a learning experience for me, and that I intend for this piece to be received as constructive criticism, not a castigation (another new word I learned, which means severe/harmful criticism).

Static ads from the Facebook Live campaign (source)


My goal was to test how well people would be able to discern the intent of the static Facebook Live ads from a quick glance. I understood the intent of these ads to be “how to use Facebook live.” The approach I converged on to test this was to create a form that would present the ad for 3 seconds, then ask the viewer what they understood the ad to be about. However, before presenting the respondent with the ad, there was a message that informed them that they’d be seeing an ad shortly. At first, the question the viewer received after seeing the ad was: “What did you understand the ad to be about?” Due to reasons I detail in the “Shortcomings” section, midway through the experiment I changed the question to: “What did you think the takeaway of that ad was?”

I was not sure of any solution that would allow me to build this type of form, and I wanted to code a bit, so I built an extremely basic web page using HTML/jQuery, deployed it to heroku and used the google-spreadsheet npm package to dump the results into a Google Sheet. When I deployed the app, I sent it to some friends and had them submit their findings. Once I felt that I had burned a sufficient amount of social capital spamming my friends with my “research project”, I combed through the results and very subjectively placed them in the following categories:

  1. Not enough time
  2. How to use Facebook Live
  3. Live, but no FB
  4. Facebook Live
  5. Other

My “app” (wait for it to show the image):


In this section I go over all the reasons my research should not be trusted. There’s a lot of them. ¯\_(ツ)_/¯

  1. 3 seconds
    1. I very-unscientifically tested myself reading other static ads for 3 seconds, and it seemed like a sufficient amount of time to grok an ad. This article states “Six seconds has been touted as the industry average for reading a billboard. So, around six words is all you should use to get the message across.” In hindsight, I probably should’ve added some extra time to the timer.
  2. The question
    1. When I first started circulating the form, the follow-up question was “What did you understand the ad to be about?” I realized shortly thereafter that I was getting some responses saying “it was about Facebook live.” This led me to believe my question was more oriented around what the topic of the ad was, as opposed to the intent. As such, I reformatted the question to be “What did you think the takeaway of that ad was?” Again this was because I understand the actual takeaway of the ad was to inform people as to how to use Facebook live. The data would be better if I had this question from the beginning or if I had a better question altogether.
  3. Categorization
    1. I definitely could have messed this up!
  4. Sample size/bias
    1. In total, I received 52 responses from friends and family. There are many potential points of failure here. First, just the size of the sample set is not that large, and more substantial observations could be collected from a larger sample size. Second, the sample was predominately comprised of Millenials that share similar interests and preferences to me, for the most part. Third, respondents may have saw this as an opportunity to flex their humor muscles, and may not have given this fully serious consideration.


Raw data (Some repsonses are not the most professional :p)

Category Count Percentage of total (52)
Not enough time 9 17.31%
How to use Facebook Live 13 25.00%
Live, but no FB 9 17.31%
Facebook Live 7 13.46%
Other 14 26.92%
  1. (20/52) 38.46% of responses had some indication of Facebook Live
    • Derived by combining counts of “How to use Facebook Live” and “Facebook Live”
  2. (29/52) 55.77% responses had any relation to Facebook Live or just livestreaming in general
    • Derived by combining counts of “How to use Facebook Live,” “Live, but no FB” and “Facebook Live”
  3. A lot of the folks that fell into “Other” thought it was something related to animals


Again, I didn’t take the most scientific approach, but I think the data is still somewhat meaningful. I think the data lightly suggest that people are not able to digest that much text in passing and there exist simpler, more powerful ways to deliver the same message. I think the approach Facebook took with the Live video ads was much better and could have been translated to their static ads to achieve a greater impact on consumers.


As I mentioned in the introduction, I want this to be a constructive critique. As such, I wanted to discuss what I would have done if tasked with this campaign. I plan to write more in depth about the ease of criticism as compared to the difficulties of creation, but until then I’ll leave this as an artifact of my views on criticism.

Simply put, I’d just take a still out of one of their brilliant video ads and overlay the three steps required to go live. Imagine the following:

I claim that this conveys what the product is and how to use it, with more visual direct visual cues. This mockup is obviously not production-quality, as I literally screen grabbed multiple stills from Facebook’s “Hidden Talent” video ad to capture their contorted gentleman, fonts, colors and emojis. Nonetheless, I hope that these suggestions can help someone improve the efficacy of a similar awareness campaign in the future, or the quality of a study on the efficacy of an awareness campaign. I sure learned a lot!