Last week Adobe and IBM both came out with studies about how little impact Facebook had on referrals to sales on Black Friday’s online shoppers. Facebook then countered with their own evidence showing how they impacted traffic for retailers. Yes, there is surely an attribution modeling issue at hand here. But that is not the end of the discussion.
Facebook has been working on a new conversion tracking program that will allow advertisers to track view-though conversions.
View-through conversion rate is the percentage of users who view an ad and neglect to click on it, but within a certain period of time go to the advertiser’s associated conversion page and undertake the desired action. In other words, Facebook wants to claim that just because and ad showed up at some point on your page, weather you saw it or not, and eventually made it to an advertiser’s page.
So what’s my issue?
I actually saw an ad this morning and it caught my attention; there is a first time for everything. I started to read it and then it disappeared completely. Nearly a hundred+ refreshes and it still never showed up again. I’ll never make it to the advertisers page.
However, I did see an ad from one of my friend’s company (btw I love Hyperwear‘s products):
While searching for the other ad, I discovered that when you click on the “see all” link by all the Facebook ads…
… it will take you to a page with “all the ads you are eligible for” and then you can click to see the ones Facebook believes you recently viewed:
Do you see the ad from above from Hyperwear? No question that I viewed that one. There are a few here that I recognize but most of them believe that I have never seen. Do you think that perhaps there might be a problem in how they attribute things?
Go ahead and try it yourself. Log into Facebook first.
1. Take notice of the ads on screen, or grab a screenshot.
2. Click to view all the ads.
3. Then click on “Recently viewed.”
How did Facebook do?
Are you willing to allow them to validate your advertising spend?
P.S. To be clear, I am NOT saying Facebook advertising doesn’t work, they may have some impact. What I am saying is that they seem to be getting desperate to prove their value at the expense of their advertisers.
Facebook didn’t give me the “See all’ option. Instead it asked me to “Create an Ad” https://www.evernote.com/shard/s2/sh/4e11ecfd-6ddf-4747-918d-b365162e79dc/5346412b77095a77ca539710cb50a737
@techmirth Yes, some people see that and some people “see all”. Just click on the link in the post to view all the ads.
@TheGrok Amazing, or not. I only remember seeing two ads on the Recently viewed, and one ad in the See All. Between this and the huge drop in users who see your page’s feed if you stop paying to promote your feed it leaves a guy pretty disappointed and frustrated. Like tasting used, cold coffee grounds when you were expecting fresh, hot Kona.
We did a poll to validate the IBM findings and saw the Social Media had a solid branding effect, leading to about 25% of respondents buying a product after seeing a Social Media post or ad. (By comparison, about 35% of respondents bought after viewing an email)
http://www.searchinfluence.com/2012/11/black-friday-cyber-monday-sales-data/
I’m not sure if the “flat display” Marketplace ads are the main driving factor here — Organic and Sponsored Stories (which often show in the News Feed) are most likely the highest converting, as they are affected by the filter bubble and will usually have personalized social proof along with them. Twitter is also included here, as opposed to just Facebook, but I think view-through conversions do occur, especially for brand-aware users — in that way, it’s like a TV ad.
Is it a clean attribution? Not likely, but that’s what we get paid for, I guess.
@ferkungamaboobo”If you torture the data long enough, it will confess.” – Ronald Coase
It seems to me that the answer is somewhere in the middle – the IBM model is only considering referrals to sales, and Facebook is taking credit for everything. We know from thousands of years of branding research that exposure over time = purchases.
IMHO, part of the issue is that we are trying to measure something for Facebook that we don’t even measure for other channels. How many referrals did the commercials bring in? Or the billboard? Or the newspaper ad?
We know that it is a combination of touchpoints that leads to purchase, and any attempt to directly put a number beside the impact that one channel has in isolation will always involve many assumptions and ignore other factors that drive purchases.
@KristaNeher Agreed. But what is Facebook defining as “exposure” is at question. They are not going to solve this issue by doing a land grab for credit on the marketing spend. I wish they invested their brain power and programming skills to re-think the nature of social advertising. Not pushing advertising in the face of their user experience and frustrating them as well. This becomes the sounds of the bell ringing in their death march not the bells of success from wall street.
I have seen mixed reviews and have been trying my own hand at targeted ads on FB – I did pull in more sales on Cyber Monday, but I can’t validate that it was because of my ad – I had friends sharing my sweet deal after all *grin*
Katrina I am glad you are testing it on your own. Just don’t follow the crowd blindly.