While many companies have achieved success in the pay per click space, when launching advertising campaigns on Facebook, many companies have failed. . Let’s look at some typical issues and how to deal with them :
• Keywords are not interests : You have keywords on Google versus interests on Facebook. In the former, someone is actively searching for something and is expressing immediate intent. In the latter, you’re targeting WHO someone is, as opposed to WHEN they are going to buy. You’re likely hitting them weeks and months before they search, so your targeting and ad copy must be different. We’ve seen PPC companies attempt to peddle translation tools that convert search keywords into Facebook interests. You might as well make chicken salad out of chicken poop– not possible. In search you know WHEN, but not WHO– in Facebook, you know WHO, but not when.
• Ads take users away from Facebook : Users who are on Facebook don’t appreciate being yanked out of their browsing experience. So don’t send them to your website– send them to your Facebook fan page. But that also requires that you have a custom tab on your Facebook page — a landing page that is just as specific as any PPC landing page, whether sending people to a particular product page, video testimonial, store locator, or whatever. And that does take a bit of engineering effort as there are few app makers that can build FBML apps. WebTrends just bought Transpond for that very reason.
• The Call to Action is too Strong : Imagine you’re having a nice dinner with a friend. Then some loud salesman interrupts your meal to pitch his wares. You’ve never seen this guy before— he’s not a friend, and you aren’t exactly interested in buying his stuff right NOW, thought it’s something you might consider later. That’s what Facebook advertisers do today– they shout over the din of the other shouting advertisers, just as you see in the content networks. With Facebook advertising, you can use microtargeting and geographic targeting to avoid “shouting” at your audience.
• There isn’t multi-step engagement : Because advertisers are trying to go from impression all the way through to the sale in the same visit (yes, it works in PPC because you can target bottom of funnel terms), they fail. Instead, have one set of ads designed only to get fans from the right target audience. Then another set of ads messaging just fans. Then another set of ads for friends of fans. You wouldn’t say the same thing to someone off the street versus a friend you’ve known for a while, now would you? When advertising on Facebook, you need to segment based on the desired user action (“liking” your page, signing up for a newsletter, purchasing a product, etc. And no, this concept is not available in mainstream PPC tools– those software companies are still trying to jam the round peg in the square hole.
• They aren’t refreshing ads daily : In PPC you can make some ads and they can live a long time. We have ads that are years old that continue to build good Quality Scores. We just leave those campaigns as is–set it and forget it. In Facebook, ads burn out in days . The smaller the audience, the more quickly ad copy needs to be changed . Google ads don’t burn out because it’s a different set of users searching on the keyword each day. In Facebook, you’re hitting the same inventory over and over– especially since the average user spends 7 hours a week on Facebook and consumes dozen of pages. With no frequency capping on Facebook, you better keep your ad copy fresh– not just because you want to split test, but because you don’t want to burn out by wasting inventory on the same people over and over.
• Their analytics is sending you the wrong message : If you’re measuring conversions, odds are that it’s the unspoken last click attribution. In other words, the user may have come to your site multiple times via organic, paid search, email, social, or other sources– but only that last click (likely a branded Google click) got 100% of the credit. In paid search, there is the concept of the “assist” and the “view through conversion” to give credit to other touchpoints prior to conversion. In the world of multi-channel marketing, where consumers take in multiple inputs before making a decision, you have to measure how many Facebook visits (or even impressions) resulted in an eventual conversion later. Facebook does have a conversion tracking tool and Ads API– but it’s still too buggy for mainstream users.
• You’re Putting Impressions over ROI : True, when you have a new page, you want to get a lot of fans. If you’re a media buyer, you might even be looking for raw CPMs. But a fan is not a fan. You need to measure what those fans are worth. And there is no one size fits all– you can’t just use the ClickZ figure of $3.65 per fan and multiply by the number of fans you have. You have to measure how many of your fans eventually convert and then calculate back to an average fan value. If 5% of your fans eventually buy something and that something is worth $100, then a fan is worth $5 with full attribution. If you find the overlap is 33% between channels on average (3 visits on average between all channels prior to conversion), then your fan is worth $5 divided by 3–or $1.67.
There are no software packages that will save you from these pitfalls–you or someone in your organization must develop the targeting, ad copy, and landing tabs that reflect your unique selling proposition. In the same way that great traditional PPC has tight linkages between the keyword, ads, and landing page– on Facebook, you must have tight interests, ultra personal ad copy, and many interest-related landing tabs.
Dennis Yu is CEO of BlitzLocal, a Facebook advertising agency that has been serving brands and local resellers for 3 years. Come hear him speak at PPC Summit.