My original hypothesis was that landing page optimization was often under resourced by internal product, engineering, and design teams, despite the impact landing pages can have on conversion rates. To test that hypothesis, we built a proprietary platform to build and optimize landing pages, and this has been our primary focus since our launch.
The core of our platform is our component library. It's a repository of generic page components that we've built over the last 2 years. We write styles to reflect each client's unique branding which our components inherit and we use those components to build production landing pages. Our landing pages typically live on a sub-domain (ex: try.example.com) which allows us to work independently of internal engineering teams.
We've focus on three core landing page use cases:
Quiz landing page examples (Editors Note: I first tested a quiz approach with Hubble in Oct 2017 which has subsequently been copied by a number of their other agency partners over the years)
Though there are a number of excellent WYSIWYG page builders out there, we decided to build our own system for a few reasons:
In conjunction with our platform, we have a strong team of CRO folks who've spent the last 2 years optimizing landing pages, and we bring their combined learnings to each new experiment. I believe that our strategy work is the most critical aspect of our service. It allows us to work in close partnership with our clients' teams and agencies so that our landing page strategy is generally aligned with campaigns, creatives, and objectives. I've found that many of our clients simply lack the time & know-how that's required to make landing page testing impactful, and our service helps to fill in those gaps.
Landing pages can be hugely impactful, but by definition they are almost always limited to paid channels. Direct/Organic/Referral etc. traffic sources will never find their way to a landing page, unfortunately. We wanted to help our clients optimize those traffic sources so we built an on-site optimization platform.
We've built our own audience segmentation system to allow us to target audience segments with specific experiments and to set and hold control groups for our experiments. Any of the dimensions you could use in Google Analytics to create a report we can use to create audience segments that we can display certain experiences to.
Our on-site optimization use cases are often quite customized for our clients, and I intend to share more of these examples in future blog posts.
For a subset of our clients, we are assisting on the growth side (media buying, creative testing etc.), particularly those that don't already have large internal teams or other agency partners. I'll have a lot of interesting results to share in the future on this front.
If you are interested in learning more please let me know at email@example.com. I hope to share new blog posts here on a regular basis going forward.
Another attempt to look at platform-wide dynamics post iOS14.5 through the lens of Thesis clients' data.
A feeble attempt to try to understand how Facebook's performance as a user acquisition channel has changed since the iOS14.5 rollout.