When we launched StatusPage.io in mid-March of 2013, our conversion rate was pretty terrible. We knew that we'd need to turn traffic in to customers much more efficiently if we were ever going to create a serious business. At the beginning of July, we started investing a serious amount of time into increasing that number and I'm happy to report that we've increased from .136% to .424% -- an increase of 311%!
The most important step in increasing your conversion rate is finding out why people aren't converting in the first place. So we gathered tons of data as to why people weren't converting. Here are some of the things we did:
If you don't have a way to visualize your conversion funnel from visitor all the way through paid user, that should be your first step. For those who don't have this already set up and don't know how to do it, subscribe to get updates at the bottom as I'll be writing another post detailing the tools and methodology. We started digging into the data we've been collecting with various companies like Google Analytics and Mixpanel (use Segment.io to manage multiple).
StatusPage.io is a SaaS subscription model where our customers can sign up an account and set up their status page for free. When users sign up their status page is hidden by default. When they're ready to make it visible to their customers, they pay to make it public. So we meaured people from visiting the site, to signing up, to setting up their page, to paying to launch it. You might be surprised just how big of a drop-off you're getting in conversion due to any given step.
Your funnel should look something like this.
We also gathered some qualitative data about why users weren't converting:
- Watched a bunch of people poke around the marketing site to find where they were confused or had questions
- Asked current customers what they found valuable about our service
- Scoured old feature request emails for patterns
For us, there was so much low hanging fruit in watching a bunch of people check out our marketing site. I suspect there's a good amount of diminishing returns there, but if you've never done it, the value you get out of it is tremendous. Once we better understood why people weren't converting and what we could do to fix those, we set to redesigning the marketing site and helping them use our product once they had signed up.
Things We Did To Increase Conversion
Better Social Proof
In our original design we highlighted a few of our beta customers. After a few months we had landed a few customers that were household names in the tech community so we updated the customers section the reflect that.
Demonstration Videos and Screenshots
One of the biggest reasons visitors don't convert is that they don’t understand what they are about to sign up for. We overcame this obstacle by adding a section that highlights all of the parts of a status page. We also added a section with videos going through the motions of creating an incident and updating a component.
Quotes For Each Feature
You probably already have a page that shows off all of your features. Valuable is that is, it can still be better. Your customers don't think in terms of the features you have, they think in terms of the benefits they get from using your product. We added customer quotes next to each feature detailing the specific benefit that feature provided.
Created A Better First Experience
We've had a "wizard" setup for a long time and it's been really effective. After a few months of gathering data on what actions users took that made them more likely to convert, we were able to change some of the steps of the wizard to reflect that. Once users have completed the wizard and land in the interface, a progress bar showing them what other things they could do to fill out the rest of their status page follows them around until they launch their page.
Lessons You Can Apply To Your Startup
Show Don't Tell
One of the biggest reasons visitors don't convert is that they don’t understand what they are about to sign up for. Our product was a webpage that customers use to communicate with their customers during downtime. Telling people that's what they get isn't nearly as powerful as showing them what the status pages look like and all the features that came with them.
Social Proof Is Vitally Important
Having launched in February, we're a very young company. When you're just starting out, being able to show that bigger, more established companies like New Relic and Disqus are paying for your product lends a lot of credibility.
Design For The Novice, Configure For The Pro
Not much to say that hasn't already been said here but I'll just reiterate. Your biggest enemy as a startup is the back button. If it's not immediately obvious how to use your product, new signups don't have a problem abandoning your website in search of an easier-to-use solution.
Yes, your product and feature set will grow over time. Just make sure you're appropriately hiding all the little nobs and switches of your product such that they're accessible to power users but not overwhelming to new signups.
An On-Going Process
All things considered, our conversion rate still has a LONG way to go. We'd like to get it somewhere in the mid-single-digit range. We're going to continue talking to customers, testing new ways of talking about ourselves on the marketing site, and iterating the on the product. We'll use our current conversion as a baseline and write a followup of all the progress we've made in a few months.
Discuss this post on Hacker News.
Now that StatusPage.io is participating in the summer Y Combinator class, we're in the process of winding down our consulting gig that funded development in the first place. The setup we used to fund development for StatusPage.io was unique in the contracting/consulting world, but it afforded us great periods of working full-time on StatusPage.io while still paying the bills each month. Cognitive overhead and context switching are vile, vile creatures, and systematically eliminating these will free up your physical and emotional energy to maintain a startup and a consultancy at the same time.
If you aren't going to raise money for your startup, transitioning your consultancy to a "one week on / three weeks off" model can give you a manageable balance of both"
Let's be clear: enabling this model is not easy, but it's heaven when it works. Since RoR developers can charge well into the $100/hr rate bucket, billing a continuous 40 hours for only one week every month provides for a modest salary while you pour your heart into what you want to be your full time product business, and actually works out best for both parties in the long run.
To make this successful, though, there are a couple rules that you must adhere to to ensure both parties remain happy with the development progress.
Rule #1: Get the perfect client
Selecting the right client is key. I can't understate this, so please reread the first sentence before continuing. There's an extremely limited subset of clients that will work well with you on a one-week-on-three-weeks-off basis, and it's your job to hunt them down wherever you can find them.
This mythical client will likely have the following characteristics:
- Mostly professional services type of business, with an increasing focus on productizing their business or launching a new software product.
- Funded, or doing a significant amount of revenue such that they won't mind paying $4k+ for the week you're actually working
- Knows what it takes to build software, and trusts you're good. Having only 40 hours leaves 0 hours for negotiation or discussion around anything. Don't even bother. They say what they want, you build it, they pay you.
- Is organized, and is available during the week that you're "on".
Rule #2: They do all the planning
Since you're only "on" for 40 hours at a time in one continuous motion, there's no room for planning and you should be spending all of your time executing. Getting a client on this setup means they won't be using your time to waffle on specs or get feedback on wireframes, that skill should be honed internally and they should only send you what they think are final mocks and specs. This is never a perfect process, and you're bound to answer ballpack questions like "is this possible" or "can we fit this into 1 sprint" type of questions, but they're usually quick responses.
Rule #3: Get them to do customer dev
When you're off building the next AwesomeCo, they need to be doing customer dev. This will inform the next sprint of work that you do, and keeps them active verifying that the newly shipped code a) functions properly and b) is accomplishing what you set out to accomplish by building it in the first place.
Rule #4: They're awesome clients, so be flexible sometimes too
Sometimes they need a couple weeks in a row, and sometimes they have deadlines, or showings, or are pitching at some launch conference. Be nice every once in a while when they really need you, and make it clear this can't be a pattern. Finding one of these clients is gold, and have some empathy where necessary.
Rule #5: Minimize the overhead of mental context switching
Set up a GMail filter, schedule any off-week calls for your car ride home, do whatever you need to do to get them out of your head when it's your startup time. 3 weeks will never seem like enough, and it's always hard to tear away after a solid flow session lasting 21 days. During these 3 weeks, turn off anything related to the contract client and let them know you're off limits for anything that doesn't 100% need your immediate attention and expertise.
On the same token, those 40 hours better be the highest octane hour you can give them, and they should be smiling when they go to cut you a check. You should be exhausted. They should be happy. You'll go home and sleep for the weekend as you gear up for another awesome 3 weeks. Use SelfControl.app or whatever other "self control" tools you can use to stay focused and alert. States of flow - learn about them and strive to be in them. The Zuckerberg "wired in" scene from The Social Network isn't a joke, and your client deserves this from you.
Rule #6: Exercise, sleep, eat well
This goes without saying. I put it here because it can take planning and explicit action to make sure you're doing it. Train your brain to reward yourself when doing these activities. The human side of your brain recognizes they're important, and it's up to you to force your lizard brain to accept it as well.