But, though the basics may be easy, getting truly excellent yields from your ads (ie more clicks for less money) requires a mixture of art and science and presupposes both careful planning and sustained attention.
No ad to show here.
Here are five battle-proven tips for squeezing more value out of your Facebook campaigns.
1. Make the most of social context
Facebook adverts are far more effective – as much as twice or three times more in my experience – when they are used to advertise a page that is native to the Facebook platform.
In other words if you send clickers to your brand’s Facebook page, application or a related event you will get 50 – 100% more clicks than if you send them straight to your site.
“But doesn’t this defeat the point of advertising?” you might ask. Not at all. Facebook lets you form a long(er) term bond with a potential customer. It gives you a way to win their trust, to interact with them on a personal level and to gradually nudge them into becoming a reader / customer / volunteer / donor.
What’s more, Facebook ads will often catalyse new fans (or event attendees) into visiting your site after connecting with you on the platform. I’ve found that as many as 25% of the clicks to Facebook nodes, like pages and events, result in immediate “bleed through” traffic to the related brand sites.
2. Use great images, not your logo
Like it or not, Facebook ads are not a brand awareness exercise. They may have more bells and whistles than Google Adwords, but in essence they are about doing the same thing – grabbing attention and funnelling it elsewhere. They are outcome and need focused – not awareness and experience focused.
We all love our brands, and few marketers can resist the opportunity to use their logo whenever and wherever possible. But ask yourself – how often do you notice logos when you use Facebook? When you do notice an advert, isn’t it because of a striking image or some clever wording?
You may think your logo is the most beautiful piece of art ever created, but the chances are that the right piece of stock photography will perform far better in the field. I’ve seen this time and again in campaigns I’ve run – the logo flops and, when it’s swapped for an appropriate photo, the campaign suddenly picks up.
So what sort of images work well? Bear in mind the image will be resized and cropped to 110×80 pixels, so try to pick images with the following characteristics:
- Simple, clear shapes (any fine details will be lost)
- Bold colours
- Landscape orientation (portrait pics won’t survive cropping at this size)
- Close-ups rather than panoramas (again, all perspective will be lost at this size)
- Symbolic value rather than aesthetic value (subtlety is wasted here)
You might even consider, as many advertisers do, using text inside your image. This is a good way to get your point across clearly and boldly, but it (usually) comes at the expense of an eye catching image. Be careful not to trade impact for clarity – you need both equally.
3. Don’t target too narrowly
Facebook’s targeting system is any online marketer’s wet dream. You can aim an advert directly at your intended audience – even if it’s Afrikaans lesbians in their 20s who like Grey’s Anatomy (that’s a whopping 20 Facebook users last time I checked).
The prospect of zero wastage may seem too good to be true, and that’s because it is. Over-targeting is easy and can make your campaign under-perform or fail completely.
The reason? The people in the above example are guaranteed to be targeted by many other brands as part of broader categories. Anyone willing to bid generously on women in their 20s will probably crowd you out. That means you need to bid higher and higher just to be seen – essentially defeating the point of your micro-targeting.
Facebook also relies on its users to tell it about themselves. If they leave important data out – as many people do – then you cannot target them based on that data. Right now there are 2,957,020 self-confessed South Africans on Facebook – 1,412,920 men, 1,467,240 women and 76,860 people whose gender is unknown. If you target either men or women, you will never reach those 76,000 odd people.
This applies even more cogently to our earlier example. I suspect there are tens of thousands of gay South African women on Facebook who are fond of Grey’s Anatomy. They just haven’t all made those aspects of themselves public.
4. Play the long game
Unless your campaign is significantly time bound (ie less than 60 days) there’s no need to blow your budget by bidding the maximum per click (or page view). If you have the time, try to bid as low as possible for each click. I’ve seen campaigns with tens of thousands of clicks at less than $0.10 per click.
As long as you’re doing regular checkups and tune ups (see below), you’ll be able to find a good equilibrium between price and effectiveness.
Another factor to consider is whether to go for clicks or impressions. The latter seem attractive because they are typically a third of the price of the former. But, as I already pointed out, Facebook ads are not a branding exercise. Rather pay more and get guaranteed results (ie clicks) than penny pinch and get only the vague possibility of results.
5. Daily checks, weekly tweaks
Socrates famously remarked that “the unexamined life is not worth living.” And while he was speaking of matters far loftier than Facebook, the spirit of his words hold true: there’s no point in running a Facebook campaign if you’re not going to learn and adjust along the way.
Facebook’s stats are both easy to understand and easy to use – there’s really no excuse to spray and pray, or to let the same advert run for months without adjusting it. The ad platform lets you adjust every aspect of an advert, virtually on the fly. Don’t think that image is working? Try another one and watch the results. The same goes for the wording, or the demographic targeting.
As a rule of thumb: check your ads daily but update them only weekly. You want to be able to accurately track what effect the changes you are making have on performance. Adjusting campaigns daily will make your changes hard to distinguish from natural cycles (like weekends) and background noise.
Also, if you’re going to adjust an advert, try to do it one or two variables at a time – or you won’t know which change made the difference. So change the image, but not the blurb or targeting (or vice versa). If you’ve got the budget and the time you might even consider running slightly different ads simultaneously (aka A/B testing).
Ready? Set, go!
What are you waiting for? Get out there and start experimenting. Give it a few months and you’ll be able to teach me a few tricks. I look forward to hearing how things turn out. (If you’re still not convinced that Facebook ads are worth using — check out my piece from last week.)
