Some Real Life Results From A Facebook Promoted Post
Facebook promoted posts for business pages give page owners a way to artificially, temporarily increase the edgerank of a post and get it seen my more of their fans.
While some have shown that converting-actions are higher on sponsored stories, sometimes promoted posts make more logical sense for a page.
While I previously included some of my original test-drive stats… now that Facebook has extended the run-time from 3 to 5 days and the reach from “fans” to “fans + friends of fans”… I wanted to take another go at it!
For this run, I needed to put some exposure on an affiliate link for a sale that was only running another 1.75 days.
I wanted something with much stronger impression rate than a traditional ad was going to get in that time span.
Related Resources from B2C
» Free Webcast: Build Better Products by Identifying and Validating Your Riskiest Assumptions
The results… I must say… surprised me… (In an odd sort of way)
As you can see, the content was very promotional. However, it was buffered by a LOT of content in the last two weeks that was non promotional. It also was “enhanced” by an image post, earlier in the day that had gained a lot of likes… thus naturally improving the page’s overall EdgeRank (visibility to fans).
For the purpose of this test… I wanted to drive as much visibility as possible in a day and a half. I selected the $300 option (knowing that I would not spend all of it) which, at the time, had an estimate of 44k – 64k. (The estimate has gone up because the EdgeRank of the page is up at the moment.)
In total, I spent $60.54 cents before stopping the promotion. More on this below.
What did $60.54 buy me?
Well… that’s an interesting situation…
432 post likes and 23 post comments… okay… that’s kinda sexy! (That alone will be a good boost for my page’s overall EdgeRank for about the next 3 days).
108 link clicks? I don’t think so… And don’t even ask me where they found 2 video plays!
So what do we know about the real clicks on that link?
Unfortunately, in the Ad Center, “clicks” mean something very different and are not useful for sussing this out.
According to Pretty Link Pro… which is set to aggressive data collection while filtering for known spam bots. There was only 5 unique clicks generated. (Click on the image to zoom in).
However… while the norton user agents (17 clicks) might actually be uniques… the 56 clicks from one individual come from a known spambot.
So, optimistically, there may be 20 unique clicks… and pessimistically there may be as few as 4 unique clicks by humans. Ouch!
Well… how do we sort it out?
Here’s the real answer => to the best of my knowledge… based on time-tracking and based on knowing who some of my purchasers were… at max I generated a single sale (At $24, so $12 commission) and most likely I truly generated no sales off of this campaign.
But that’s not the odd part.
(Really?? Yeah, really!… watch this!)
The Thing That Made Me Do A Double Take
The odd part comes in the comments that were attained… (Apologies if any of these are foul… as I do not know what half of them say!)
While I can not understand many of the comments… a quick look at some profiles reveals that most of my comments came from Pakistan!
I don’t have a huge (or really even statistically significant) Pakistani following on my page… so this puzzles me.
Where were the comments (or proof of visibility to) my USA, Australia, and UK communities?
Did I not get visibility to my primary communities…. or were my Pakistani friends just overwhelmingly generous with their comments?
While not all of the comments are Pakistani… some are Indian & Malaysian… and some I have no way to identify (but most appear to be of that region)… the demographic is so skewed in percentages as to feel very odd.
I wish Facebook’s translation tool were kicking in as I’d love to be able to know what some of my commentors have said too!!! (It’s really honestly disconcerting to have comments you can’t read.)
It’s very hard to judge whether the demographic curve would have sorted itself out if allowed to run for a full five days. Perhaps we’ll find that out in another test some day.
While the total sales percentage appears abysmal… it actually is typical when aiming marketing intended for the USA/UK/Aussie market at the South East and South Central Asian market. The market that received the message is not the market that the message was designed for. (Message to market mismatch.)
The generous commenting by the South East and South Central Asian community was really cool even if I didn’t know what some of them were saying! (Did this detract from the USA/UK/Aussie markets commenting… or were they not exposed to the promoted post?)
Generally I find that my South East and South Central Asian communities are very willing to share content, comment, engage and make great friends… but are not prone to be buyers for me (partially due to the differences in economic dispositions).
I wonder if the new “Target by Location” feature available to Page posts could have been used to have excluded the unintended markets. However, the reach estimates in the promotion tool do not reflect a decrease in reach when targeting is used… implying that they are not yet correctly connected.
Well… it’s been a learning experience!
I hope you were able to gleam some insight into the Facebook Promote Posts tool from these real world results!
It’s not often that Facebook gives me a reason to have a completely Scooby Doo expression… but this sure did it!
Have you tried out Facebook Promoted Posts yet?