We’re on a historical blog optimization kick over here on the HubSpot blogging team. “What in the Sam Hill is ‘historical blog optimization,'” you ask?
Basically, it means optimizing our “old” blog content to generate more traffic and leads. And by “old,” I just mean posts that already exist on our blog, whether we wrote it three years ago or last month.
A lot of different tactics can fall under the umbrella of historical optimization: updating and republishing blog content so it’s fresh and up-to-date, optimizing posts for search to improve their search rankings, and conversion optimizing posts that still generate a lot of traffic but don’t convert visitors into leads as well as they could.
I’ve already written about why marketers need to stop neglecting their old content. In this post, I’m going to make an even stronger case, focusing on one specific conversion optimization tactic: keyword-based conversion optimization.
That sounds boring, so before you bounce from this post, let me rephrase: I’m going to show you how we increased the number of leads we generated from our top-ranking blog posts by 99% — and explain how you can do that, too.
How We Approached Conversion Optimization
Our First Approach: Focusing on the Calls-to-Action (CTAs)
Over the past several months, ever since we discovered that over 75% of our monthly blog post views were also of old posts, we’ve been on a mission to improve the conversion rates of our old posts.
We started out by targeting our highest-trafficked old posts to see if we could improve their lead generation potential. Our approach was mainly to optimize the posts’ calls-to-action (CTAs) based on the following questions:
- Was the primary CTA’s offer the most relevant offer we had for the post’s particular subject matter?
- Did the post include a slide-in CTA? If so, same question as above.
- Was the CTA design outdated?
- Were there other CTA opportunities available that we hadn’t thought of at the time of writing?
The results of this approach were hit or miss — sometimes the conversion rate of the post would improve, sometimes not. Sometimes, the conversion rate would even decrease.
What we realized was, while we were using relevancy to inform our offer/CTA choices, it was still mostly a guessing game. There was really no way to know whether a visitor would prefer one offer over a another one. Our approach wasn’t data-driven enough.
Then, we had another idea …