Backdoor vs. Frontdoor Products
On content platforms that have successfully become destinations for content consumption
We have millions of people go to our front page and tens of millions of people go to our “B” pages — Jonah Peretti, Buzzfeed CEO
Typically, as online content platforms scale they’re heavily “backdoor” products— that is, non-homepage landing page visits heavily outweigh homepage. Think early days of Youtube, Tumblr, Ustream, where the site has no brand and you’re trying to quickly build the supply side of the market (UGC content), so the “market” is so thin that there isn’t even enough compelling content to merchandise a homepage, much less fill a personalized homefeed. Instead, traffic is pouring in onto content pages via a combination of social + SEO. This is the beauty of unbundled content.
But as the # of good content producers grows, the goal becomes capturing all this traffic they bring in and turning them into repeat users. This traffic is primarily users arriving on a particular piece of content with an intriguing headline/subtitle/thumbnail, i.e., stumbling upon or channel surfing. If you’re lucky, they consume the entire piece of content and share; if you’re extra lucky, they click through to a related piece of content; and if you’re extremely lucky, they’ll register and now you can re-engage them. This is where the business tries to re-bundle the content. A combination of smart product tactics, strong growth on content acquisition, and scaling brand/buzz can augment that luck to help the site become the go-to destination for the user.
Observations from the world of Online Video
I was at Ustream in 2009 when it was thinking through this exact challenge. Our traffic was growing in large step functions for every hot stream as they exploded on twitter (e.g., puppycam, jonas bros, MJ funeral) and we actually had okay cross-promotion (avg visit resulted in ~2 video views), but consistently only ~1% of traffic resulted in registrations. By and large, visitors weren’t coming to Ustream to FIND live video to watch, they were led to Ustream to watch something they already had intent to watch — we were a backdoor product. While I don’t know how these metrics have scaled since, the current homepage probably implies that this is still true.
Why? I don’t think consumers ever differentiated live video (ustream) from recorded video (youtube) the way Silicon Valley did — it’s all just video, and a “frontdoor” go-to online video destination needed to be the one-stop-shop for all good video content, whether live or recorded.
Interestingly enough, sometimes it’s powerful to be the one-stop-shop in “niche” verticals. One competitor of Ustream’s, Justin.tv, saw the popularity of live video in the gaming vertical and acted on it, boldly separating it out as a separate site, Twitch.tv. It’s become the ESPN of e-sports (as opposed to the Yahoo Sports), the destination for live online gaming video. And that one vertical has exploded above the rest of all non-gaming live video added up (see chart).
In the broader online video space, Youtube is the clear winner, but it wasn’t always THE destination for online video. Remember Crackle, Revver, Dailymotion, Metacafe? Early in the days of online video, a friend would AIM/ICQ you a funny video, and you’d go to whatever site it happened to be hosted on (even if it was ebaumsworld). Over time, though, Youtube grew its content producer base, built a brand, and became synonymous with UGC video. The Youtube search bar has become known as a magic portal to all things video across the Internet.
Meanwhile, Maker Studios is trying to become the “if you want to watch awesome videos from the best Youtube channels”. While they currently support an unbundled cobweb of youtube talent (generating 4.5B+ monthly views), they appear to be trying to revive Blip.tv to be the go-to destination (i.e., bundle) for the creme de la creme of online video not locked up by Hulu/Netflix/Vevo.
Why does it matter if a product is “backdoor” vs. “frontdoor”?
- Frontdoor = repeat traffic. Backdoor products can definitely be good businesses. Ustream’s traffic continued to grow, and traffic = ad impressions = revenue. But being a frontdoor go-to destination helps you build out your PV/MAU and your DAU/MAU — how deep users go per session, and how often they come back—which really helps build up those ad impressions. I think you’d be hardpressed to find a content platform that didn’t want these two metrics to grow. With frontdoor products, you develop a relationship with your user. They belong to something. With backdoor products, there’s generally no relationship.
- In the mobile app world, backdoor products are harder to draw app distribution for than frontdoor products. Why? By nature, native mobile apps are frontdoor products. They’re bundled. Users land on an app store page and must be convinced by the app’s value proposition in order to download. In a “pre-pagerank” phase of mobile, search doesn’t give any links into “B pages” — only the app as a whole. So backdoor products that live on unbundling on the Web die by bundling on mobile.
- And mobile apps do matter. Not having a mobile app = foregone growth opportunities. For one, not having a mobile app means you give up app store SEO (e.g., the off-chance that a user DOES search for Upworthy and it’s non-existent). You also give up the possibility of app store promotion, which is somewhat BD-dependent, so backdoor businesses aren’t at a huge disadvantage here. Not having a mobile app is also a foregone power user opportunity. Although Upworthy is a backdoor business, they still do have SOME segment of loyalists (even if 1%) that would download the app and by definition use it frequently, share, evangelize, leave app store reviews, etc.
So how does a backdoor product shift its users to enter through the frontdoor?
You have a ton of traffic coming in on B-pages to consume a piece of content. Here are some thoughts on how to drive those users to eventually think of your site first to consume content:
- Capturing registrations on B-pages (to email re-engage them). Pinterest, for example, is quite aggressive about capturing registrations on logged-out visitors entering on B-pages (try this in an incognito window). So is Quora (incognito).
- Cross-promote user immediately onto more compelling content. Don’t let the user leave without exploring your site. Imagine a TV viewer is channel-surfing and lands on a great show. The cablenet is generally working hard to show you commercials and overlays of other great shows you’d find there. Quickly drive recommended content right after the user finishes the content at hand. (This point is so obvious that you’d rarely find content products not doing this)
- Give the user plenty of jumping off points. Give the visitor plenty of compelling links to explore more nooks and crannies of your product. For example, instead of just showing related content, drive users to user profile links. This gives the visitor a quick tour of your site and what she can find there — it moves the user along her comfort curve. At thredUP, though we’re an e-commerce site, we find a surprisingly high % of item-page visitors clicking a small text link below the fold called “other items from this family»”. It’s a great way to explore.
- Take advantage of related features that are obviously “frontdoors”. For example, while few users would check-in daily to Ustream to see what’s streaming live currently, Ustream’s mobile app is a clear tool for anyone wishing to broadcast live. That’s the hook to get the download, then a gated app registration gets you either email or push notifications to re-engage the user to consuming your feed. Quora’s mobile app is aimed primarily at the ability to ask a question on-the-go.
- Aggressively scale content by driving new visitors consuming content to be content producers (or curators). More content = faster to one-stop-shop. Depending on the product, it may go hand-in-hand with driving registrations.
- Fake content you don’t have (legally), where needed. I can’t remember which ones, but I’ve seen online video sites basically embedding content from Youtube (and qualifying it as such, obviously) to seed the “marketplace.” For publishers that Flipboard doesn’t partner with to build custom story views, it shows the mobile web view instead of blocking it from the app. Several apps with location check-ins augment their own place data with Foursquare places where it’s missing from their own. Upworthy. Etc.
- Aggressively scale content in niche verticals. For broad platforms like Quora (the world’s Q&A), scaling all sorts of Q&A to the point where it’s a one-stop-shop for the world’s hardest-to-find knowledge is kinda hard. Whether they did it purposely or not, Quora has scaled content in certain niches (e.g., #tech, #VC, #quora). In an extreme example, there obviously is no better place in the world to find ideas and knowledge on how Quora works…than on Quora. That would be your go-to. For some tech friends of mine, I know that Quora is actually an app they check multiple times a day to see what’s new in their feeds. This is only true for them because Quora’s gained some critical mass here. Obviously, Twitch is another example of this.
- Aggressively scale content via partnerships, etc.