Monday, 25 August 2008

100 million facebook ad impressions and counting...

There are more blogs and articles written about the Facebook advertising platform (social ads) than there are advertisers. Most of which have been written by people who have either never run an advertising campaign on Facebook or have perhaps run a very (very) limited run.

Here then is a blog that has been written by someone who has had their adverts served more than one hundred million times.

The initial excitement about Facebook advertising was the prospect that it would truly provide a contextual advertising engine that would enable advertisers to precisely target consumers by their interests.

This caused a number of strong reactions; those concerned with privacy raised the alarm about the level of personal data being collected and those who thought this was the marketing nirvana quickly went from a hallelujah mentality to a one of being harshly critical.

The reality is that Facebook advertising does not present the contextual power that those, like me, who joined the goldrush first thought possible. In short facebook do not know that much about us and what they do is reliant upon what we decide to say about ourselves.

With the exception of a few categories such as educational or relationship status (married, single, engaged) which opens the doors to some interesting advertising opportunities for dating sites or wedding specialists the rest of the targeting is based on keywords.

The theory as FB themselves put it is that you can use keywords to target your adverts to stating that 'your advert is more likely to perform better and continue running successfully if it is being displayed to the users who are most likely to be interested in your product or service':

The trouble for any advertiser who wants to reach a fairly broad audience is that the keywords are based on interests, activities, and favorite books, TV shows, movies, or job titles that users list in their Facebook profiles and whenever you look at a selection you normally find that the resulting number of people who have such a criteria are very low indeed.

The following screen shot shows that there are only 8720 males aged 18-25 interested in watching films:


On the upside certain demographic profiling, namely age, sex and location are really easy to implement and very useful. If, for example, you have a product to advertise to an adult male audience you can easily create a series of adverts to target men only and then segment your marketing by creating ads for different ages (18-24, 25-34 etc.) within your target market each with a different advert - should you desire.

Now let's tackle to next biggest cited reason not to advertise on Facebook "People do not visit Facebook to click adverts". Whilst this is true it does not mean that those who visit Facebook do not click on adverts.

Here then is some stats from one of my advertising campaigns that has been served 100 million times.

A CTR (click through rate) of 0.04% may sound staggeringly low but in reality it is akin to banner advertising and let us look at it another way my FB social adverts have delivered over forty thousand visitors to my website. All of whom have seen a small image and enough copy to let them know what to expect when they clicked on the advert - so that's 41,431 people who have, at least a passing interest in what I have to offer and I for one think this is impressive.

Boiling things down to the basics the question simply becomes did I get a worthwhile ROI (return on investment). Well would I have continued to advertise and spending many thousands of great British pounds (actually you are billed in dollars) if it were not bringing a result? No.

In fact this post itself has sparked the kernel for my next article 'In online marketing opinions count for nothing' (might need to work on the title). The point being with tracking in place and the cash flow to pay for the weekly facebook social advert bills you can test the market for relatively a small amount of money and if it works you can attempt to increase your budget.

What none of this article tackles is how to undertake a meaningful test in the first place - and this, as I have found is the real trick. Advertising on Facebook is a roller coaster experience with very little predictability. I know of several people who have simply failed to get out of the blocks and get there adverts served at all.

If you look at the very first screen shot (top of this article) you can see the peaks and troughs that are caused by your adverts serving extremely well on one day and then not at all the next day but perhaps I will leave this for a future article or commissions.