This is Week 3 in the 10 Growth Experiments in 10 Weeks series. We’re running 10 experiments to see which growth strategies are the most promising for Kiip’s business model and current state. Follow us on Twitter to get more marketing updates.
Week 2 was AdWords Experiments, which revealed Kiip’s poor 0.5% landing page conversion rate. This week’s post on landing page conversion rate optimization (CRO) is aimed to solve our conversion rate problem.
Conversion rate is the most important metric for your company
Bill Gurley, general partner at Benchmark, explains the importance of conversion rate best, “No other metric so holistically captures as many critical aspects of a web site – user design, usability, performance, convenience, ad effectiveness, net promoter score, customer satisfaction – all in a single measurement.” Even beyond Gurley’s statement, conversion rate not only captures critical aspects of your website, but also how well your company fits in it’s competitive marketplace. If visitors are coming to your page actively seeking to buy car insurance for dogs, but no visitors convert into customers, your company either (1) has a poor website or (2) does not have a winning product that fits the car insurance for dogs market. The most scientific design in the world cannot compensate for a product that has no market.
For Kiip, we isolated the problem to (1), we have a poor website for converting traffic interested in advertising on the Kiip network. Since we have a product used by hundreds of advertisers, we know there’s at least some product/market fit. However, we struggle to explain the virtues of our product on the current landing page. With a few weeks of effort and diligent testing, we were able to refine our message, improve the user flow, and convert traffic properly.
You have no excuse to not test and optimize your landing pages
There’s a surplus of amazing products to help you optimize your landing pages. The best are Optimizely with it’s simple integration and fast A/B testing, HubSpot with complex integration and an integrated inbound marketing platform, and Unbounce sitting in the middle with quick and powerful landing page creation tools. How do you decide which option to use? It depends on your current setup and what you want to test. If you want to test smaller changes like button color or text, but do not want to completely overhaul your page, use Optimizely. If you want to have a fully integrated inbound marketing platform complete with email automation, landing pages, and more, use HubSpot. If you want complete changeability for your landing pages, but don’t want an expensive integrated inbound marketing platform quite yet, use Unbounce.
We chose Unbounce because we’re not quite ready to make the HubSpot investment, but need to make more major changes than Optimizely is built for. Our current conversion rate analytics setup is entirely through Google Analytics goals. Checking our current metrics, it shows that users are excited to click “Register”, but once they hit our form, completions plummet. So we wanted to test a flow that removed the 2-step registration process, and made it extremely simple for visitors to enter their email address to sign up for more information.
Unbounce is perfect for our needs, a complete conversion funnel overhaul to remove the registration page step. 22% of our visitors make it to the registration form, which is impressive, but only 2.5% of those users complete the form, for a full-funnel conversion rate of 0.55%. Below is the original version of our conversion funnel, both landing page and registration page:
To test the hypothesis that a one-step registration process would generate more conversions, we built a simple landing page with three variations. One variation with only an example Kiip user flow section, another variation with only a “three reasons to use Kiip” section, and the final variation with both a user flow and three reasons section. The variation with both the user flow and three reasons to use Kiip sections shown below:
With four landing page variations ready to test (one original and three with Unbounce), we used Google Content Experiments to split traffic between the pages. This enabled 50% of the traffic to go to the original kiip.me/selfserve and 50% of the traffic to go to my.kiip.com/selfserve, our Unbounce subdomain, which then got evenly split between the three Unbounce variations.
Reducing steps is more important than a beautiful design
The results of the experiments confirmed the hypothesis that a one-step registration process would generate more conversions. The top Unbounce page had a 3.05% conversion rate. Over the same time period, the original version finished with a 0.67% registration rate.
As you can see, the Unbounce page with both the user flow and three reasons section converted traffic the best. The complete CRO process from landing page creation to splitting traffic with Google Experiments took about 2 days to setup, and the experiment ran for 1 month. A 450% increase in landing page conversion in return for 10 employee hours and a $50 monthly Unbounce fee is roughly a 3X ROI in one month.
2000 visitors a month (average monthly traffic)
$25 per Registration (willingness to pay per registration)
$335 value = 13.4 Registrations (2000 visitors * 0.67% conversion rate)
$1500 value = 60 Registrations (2000 visitors * 3.05% conversion rate)
$1165 increase in value
$350 = $50 unbounce, $300 employee hours
$1165 / $350 = 3.3
Our key metric is registrations as the platform still requires a person-to-person sales process to elaborate on the virtues of advertising on the Kiip network. The platform is not quite ready for full marketing automation (but we’ll be there soon!). We’re willing to pay $25/registration since 10% of registrations convert into customers with an average LTV of $250.
Landing page optimization never stops
This was a great start for us, and only leads to more exciting questions on landing page conversion. Our next test will be combing the beautiful design of the original landing page with the simple user flow of the Unbounce pages by putting an email input box on the original design.
Please reach out to firstname.lastname@example.org if you have any advice or questions on our landing page conversion process!