Why most Facebook marketing doesn't work


This post was also published in ReadWriteWeb.

For almost four years, since the Facebook Platform was launched, I have been involved in delivering Facebook apps for top brands such as CBS, NBC, Lifetime, Universal Music, Visa and more. Here's what we have learned doesn't work, and more importantly, what does work.



Deep Campaigns Don't Work

First, deep campaigns don't work. Digital agencies love deep, expensive campaigns on Facebook, with tons of pages, interaction, and art. It fits in with how agencies build microsites and websites, and justifies the $100,000-plus price tag that they like to charge. Examples include lightweight games, prediction contests, treasure hunts where you include friends, and such. Unfortunately for agencies and the brands that drop a lot of cash, Facebook users decidedly don't like deep campaigns.

They do not like to spend 20 or 30 minutes on a single brand's page, unless they are consuming innovative, funny, or exclusive content. So a travel site looking for a long time spent on a page should not put up a treasure hunt on a world map where you invite your friends and can together find great prizes after exploring cities. Sounds good in a pitch meeting, but it results in abysmally numbers of active users.

Facebook users are very sophisticated, and there is no way a single campaign is going to compete on game mechanics with CityVille. If you want to build CityVille, it might work. But, even Netflix pulled their Facebook app. You're better off putting up a bunch of funny videos from around the world and leave it at that.

Lots of Apps on One Tab Don't Work

It is easy to think of a Facebook tab like a Web page, and throw a bunch of features on it - such as a poll, gifting, and some videos - all on one tab. However, most users do not show up on a Facebook tab like they do on a Web page. They are usually coming in by clicking on a page's newsfeed posting ("What kind of traveller are you? Take the quiz!"), a friend's newsfeed posting ("I'm a cranky traveller! What kind of traveller are you? Take the quiz?"), or a Facebook ad ("Find out what kind of traveller you are!").

Now, if after clicking on one of these links a user is dropped into a Facebook Page tab with eight different things on it, they are not going to see a quiz immediately and move on. There should only be one engagement feature per tab.

Sweepstakes Don't Work

After an initial onslaught of Facebook sweepstakes promotions, marketers are learning that sweepstakes have very low conversion rates and almost no viral uptake. We're also learning that they attract unengaged users who are there for the prize rather than a relationship with the brand.

Facebook users like to click around and look at stuff, and absolutely do not like filling out forms. We have run highly promoted sweeps campaigns for major artists that included things like backstage passes and a limo ride to the show that had abysmal conversion rates. There is absolutely no incentive to make sweepstakes social.

Why would you invite more people to join a sweepstakes? It reduces your own chances. Have you ever seen a "I just entered a sweepstakes and you should to" posting on someone's wall?

One attempt to increase viral spread in sweepstakes is to offer more prizes when there are more entrants, but all that does is confuse users with conflicting agendas. There is a disincentive to invite people since it reduces your chances of winning, but if enough new people join up perhaps you can win something else... "Ah, too confusing, I'm going to watch videos instead."

Photo and Video Contests Rarely Work

A lot of brands like to do photo and video contests, but unfortunately they do not have the user base that likes to submit photos and videos. Travel and photography brands? For sure. Mobile carrier? Beverage brand? Not likely. Even clothing brands can't pull this off.

Uploading a photo or video is a big investment on the part of the user, and they do not expect to do it for the vast majority of businesses. These campaigns also require the labor to moderate the submissions. If you must run a photo or video Facebook campaign, the best way to do it is actually NOT in an app.

Instead, have users upload the photos and videos to the brand's page, and moderate them there. Then have users get their friends to Like the photos or videos. This way, the campaign leverages all of Facebook's viral channels around photos - when the user uploads the photo, when they Like the photo, when their friends like or comment on their photo submission, it is all highly likely to show up in their friends' feeds, drawing traffic. The great thing about this is that it is easy to do for free, since using all of Facebook's photo and video features are free, and users get to use the known Facebook photo and video interface, which increases conversions.

Like Blocks Rarely Work

Like blocking, where a user has to "Like" a Facebook Page in order to access a feature, typically has a 50% or more drop off rate, even when there is something there that is actually worth liking the page to get, such as exclusive content or a great coupon. Putting a Like block on basic content will almost guarantee a 100% drop off rate.

Be very, very selective about Like blocks and be sure to tell the user that it is worth it to them. A Like is the mailing list opt-in of the Facebook world, so be willing to offer up some goodness and know that most will opt not to Like.

Extended Permissions Rarely Work

A brand on Facebook should be like a casual friend or neighbor and not try to suck people into heavy levels of interaction. What do you do with a friend? Comment on their photos, like their status, vote on their outfit.
Asking the user for a laundry list of access to their profile usually results in a 30% or more drop off rate, and that is for well known brands that they trust. Do you really need to know their relationship status? Generally a brand already knows its demographic - does a youth-oriented clothing brand really need to validate that it is 16 to 25 year-old women that are engaging with the brand?

So while it sounds good to ask for extended permissions, do the math and monitor the drop off rate to ensure that it is worth it to you, otherwise the overall campaign ROI may not turn out the way you want, especially if the campaign is being graded on number of engagements.

Unbranded Apps Don't Work

It's got to look good, and be on brand. In the early days of Facebook, a brand could put up a basic presence with some turnkey apps, and users accepted that. Now that Facebook is all grown up, a brand presence needs to be on par with its website. Facebook users are savvy and will judge your brand in comparison to the best they've seen.

Dedicated Facebook Storefronts Kinda Work Right Now, But Soon Won't Work

Dedicated Facebook storefronts are the rage on Facebook right now, but they are unfortunately not integrated with an e-commerce site's existing payment and inventory systems, and are therefore a logistical nightmare. The best bet right now is to list featured products on a Facebook Page with click-thrus off of Facebook to the e-commerce site.

Now that Facebook is supporting iFrame tabs in pages, an existing e-commerce site can be skinned to fit in a 520-pixel-wide Facebook Fan Page, thereby integrating existing payment and inventory systems into the Facebook Page.

So What Does Work? Promotions and Consistent, Lightweight Engagement

Make sure your fans get something in return for liking your page with promotions likes offers for fans that they can easily redeem. The more lucrative the deals offer, the more sharing with friends with happen. Fans want things like exclusive products/services, drastically discounted prices akin to Groupon type deals, and early notification and registration for upcoming events, ideally exclusive to fans. Promotions should make the fan feel like they are a brand insider, not just a standard consumer.

A big secret of Facebook marketing is that it is easy and cheap to drive promotions using ads targeted only at your fans that link to landing tabs that deliver the offer and encourage fans to share to their newsfeed.

A brand on Facebook should be like a casual friend or neighbor and not try to suck people into heavy levels of interaction. What do you do with a friend? Comment on their photos, like their status, vote on their outfit. These types of interactions take seconds, not minutes, and definitely not hours.

A brand on Facebook should offer their users regularly updated, simple to interact with engagement features. Each of the engagement apps should be fully branded, and run in a separate tab with traffic driven from wall posts, newsfeed and Facebook ad units to increase engagement. Start with a personality quiz. Then two weeks later put up a poll. Then try a trivia app. For special events, put up a gifting app for Valentine's Day, or for the holiday season, a holiday song card.

Some brands, like media properties and well-known consumer brands, get an immediate fan base for this type of lightweight engagement. For the rest, building a fan base on Facebook is no different than building a mailing list in the previous generation of the Internet. It takes consistent engagement, and builds over time.

Methods to accelerate growth include tying Facebook ad campaigns with engagement apps and driving traffic from the homepage. The apps should still be lightweight and fun, with the conversion goal of getting the user to like the brand.

The point is to regularly put up new, fresh engagement features that are easy and fun for users to interact with, that they will want to post to their wall and share with their friends. Then users will interact with your brand just like they interact with their friends on Facebook!