Saturday, 4 July 2009

A few cool things to try with Facebook ads

I've been playing around quite a bit with Facebook ads over the last few weeks and have some interesting theories to test. I have to say I like the system a lot for its simplicity, targetting possibilities and the ad format with picture included (one negative to mention is having to wait for manual review of your ad text after any, even very minor change, such as capitalisation).

I first played around with Slovak and English ads for The Slovak ads featuring a toilet bowl (from a guest photo taken at a hotel in Bratislava) did generate some interest, but I have no way of gauging their actual effectiveness (their objective was to let Slovaks know is a good place to get help finding accommodation for their guests). The English language ads performed poorly - the two campaign had a total of 42 clicks, since there was no reasonable way of targetting them to a qualified crowd.

A few months ago I ran a campaign to sell my car. I was trying to sell my Alfa Romeo 147 in the middle of the Slovak car-scrapping craze in May. I advertised on Autovia, a leading Slovak car classifieds site, but was not receiving too many calls.

I then put and ad on Facebook with a photo of the car and managed to get almost 1,200 people in Slovakia over the age of 18 to visit my Autovia ad for $125. I did end up finding a buyer, as well.

For the past 10 days or so, I've been running a (massively successful) campaign for designDOT, a new Slovak school of interior design for hobby designers (the 'client', due to personal ties, gave me almost unlimited freedom in experimenting with the campaign and let me have input into landing page design as well).

I've had a chance to play around with targetting (separate ads for women and men, breaking down ads by age groups), budgetting, landing pages, etc.

Two things really intrigue me about the Facebook ads system:
1. Due to the wealth of info Facebook has on its users, you can play around with crazy targetting. You can target ads by very narrow age bands (e.g. you can advertise specifically to 26-year olds, 30-year olds, etc.), by employer, sexual orientation. This lends itself to creating ads that mention the user's age or exploit the other detailed info.
2. This is a hypothesis, which have to test a lot more, but it sounds reasonable. Since advertisers set maximum daily budgets and (at least based on my experience) sometimes use them up in full before the end of the day, a given demographic should be cheaper to reach towards the end of the day in its timezone. There is no "day parting" at the moment, so this can only be done manually. But for some advertisers this could be worthwhile - turn on your ads in the evening and let them end at the end of the day.

I am going to try to entice my colleagues at Pizza SEO to take part in (and bankroll) some more experimentation because I think there is a lot of value in Facebook Ads at the moment.

Were you looking for information on cool things to put on Facebook?


jano said...

first interesting article in this blog! thanks andrej! you're slowly losing the "most boring blog in JG's bookmarks" plaque.

Andrej said...

Dude, you need a favour from me, right?