What can be learned from analyzing your website traffic sources in Google Analytics? More than 65 marketers share their most valuable insights here.
Marketing | Jul 2
Jessica Greene on December 6, 2019 (last modified on December 9, 2019) • 12 minute read
Despite publishing new content at a high frequency, our traffic was relatively flat.
Six months later, we’d updated 24 posts. The results? Well, this graph paints a pretty—and pretty clear—picture.
See where traffic begins to spike there in June? That climb began only days after we republished the very first updated post.
According to John, they’d tried a variety of different things over the previous six months, but nothing seemed to work. Traffic was, as you can see, relatively static.
That changed when we started updating old blog posts.
Over the course of the six months that we spent working on this project, John occasionally shared some of the results on LinkedIn and Twitter. Many people commented asking for more information:
In this post, I’ll answer all of those questions by walking you through four of the posts we updated, explaining exactly what I did to update each post, showing the results of the update, and sharing some lessons I learned along the way.
Early in the project, John used Animalz Revive to put together a list of posts that needed to be updated. One post in the report he received from Revive stood out from the rest: a post from mid-2017 that featured 10 Databox Google Analytics dashboards.
Historically one of the site’s top-performing blog posts, it had lost more than 50% of its traffic since its peak. Beyond that, it was a really important post for the product. Google Analytics is Databox’s top integration, and this post highlighted some of Databox’s most useful Google Analytics reporting templates.
Here’s a Wayback Machine link showing what this post looked like before I updated it.
After reviewing the existing content, searching for the target keyword—“Google Analytics dashboards”—in Google Search, and reviewing the top results, the reason for the traffic decline was clear: the post didn’t satisfy search intent.
“Dashboards” is a term the team at Databox uses to describe their prebuilt reporting templates, but it also happens to be a term that Google Analytics uses within its own product.
So when people are searching for “Google Analytics dashboards,” they’re looking for examples of dashboards you can build within Google Analytics. Databox’s post didn’t provide any examples of those. It was all Databox dashboards.
A quick Google search made the search intent discrepancy clear: all of the results that were outranking Databox’s post were specifically about dashboards you could build inside of Google Analytics.
The updated post, then, needed to satisfy that intent. We needed to focus more on dashboards in Google Analytics than dashboards in Databox. So that’s what I did.
But just because we needed to highlight dashboards in Google Analytics didn’t mean we couldn’t also highlight those same dashboards in Databox. In the update, I added screenshots of both products in order to both satisfy search intent for our target keyword and showcase Databox’s product.
And personally, I think it makes an even better case for using Databox. Seeing the two dashboards side-by-side highlights just how much more attractive Databox’s reports are.
The comment icon below the graph notes the date that the post was updated. Traffic to the post spiked a few days after the update and has remained consistent ever since.
Key Takeaway: A crucial step in updating old blog posts for SEO is making sure that the post satisfies search intent for the keyword(s) it’s targeting. If it doesn’t, you need to either update the content so that it does match search intent or choose a different keyword to target with the post.
One good reason to update your old blog posts is that your rankings are declining. Another good reason: your post was never really optimized for search in the first place.
That was the reasoning behind updating this contributor roundup about average session duration. While the content of the post was good—and the post itself was less than a year old—John believed that updating the post with a focus on optimizing it for search could give it a boost.
He was right:
Not only did traffic to this post jump after its update (the comment icon highlights the date the post was updated), but it’s continued to climb ever since.
The key to our success with this post was in optimizing it for multiple keywords.
Originally, the post focused only on the quotes our contributors had provided with their tips on how to increase average session duration. I wanted the updated post to be more comprehensive, so I targeted a variety of related keywords:
Most of those keywords have their own dedicated sections in the updated post, which has allowed the post to rank for multiple keywords and—better yet—multiple featured snippets.
Of course, the risk in writing a more comprehensive post is that you’ll overwhelm readers who just want one piece of information. Maybe they already know what average session duration is and how it’s calculated—they’re just there for the tips on how to increase it.
For that reason, we included a table of contents with anchor links so readers can jump to the section they’re most interested in if they’re not inclined to read the entire piece:
This not only delivers a better user experience, but it also often gets you an expanded search snippet highlighting some of the different sections within your post.
Key Takeaway: If you have blog posts in your library that were not originally optimized for search—or were not optimized very well—those are great candidates for updates. Conduct keyword research and analyze other high-ranking search results to find opportunities to make your post more comprehensive.
If you’ve read any of Databox’s recent contributor roundup blog posts, you know that we take a narrative approach to roundup-style content rather than just listing a bunch of quotes next to pictures of contributors. But that wasn’t always the case.
When the Databox team first started writing roundups, they went with the more standard list of quotes:
So part of the exercise for updating this blog post about using Google Search Console for SEO (here’s the original post courtesy of Wayback Machine) was just transitioning the post into the newer narrative style.
Another part, just like in example number one, was better catering the post to search intent. The original post assumed that readers had a more advanced understanding of Google Search Console, but reviewing the top search results for our target keywords showed that most people were looking for more beginner-level information.
When updating this post, I broke it into two parts: the first covers basic information about how to use Google Search Console, and the second features our contributors’ more advanced tips. The results:
Organic traffic to the post doubled the week after the update and has remained consistent ever since.
Key Takeaway: Having old blog posts on your site that no longer match your current style is a good excuse to dig into those posts and make sure they’re ideally optimized. Not only will you provide a more consistent experience for readers, but you just might increase your traffic in the process.
So far, I’ve shown you three examples of updated blog posts that produced big organic traffic gains, but not every post we updated was so successful. Some, like this post on digital marketing software, suffered from short-term losses after their updates.
The drop-off happened because I tried to target a different keyword when updating the post. Before the update, it was ranking in the number-one spot for the keyword “digital marketing software.” But I wanted to try to get it to rank for “digital marketing tools”—a higher-volume keyword.
Needless to say, retargeting the post didn’t work. In fact, it did much more harm than good.
When I noticed the post’s rankings dropping — which was very quickly, luckily — for “digital marketing software” but not concurrently rising for “digital marketing tools,” I reverted some of my changes to reoptimize the post for its original keyword. After that, the rankings and traffic recovered (and traffic even increased a little bit).
Key Takeaway: Updating old content isn’t necessarily a one-and-done process. You have to be willing to watch your metrics closely and go back in and make changes when needed. Definitely be willing to try different things, but also keep good records of the changes you make so you can revert those changes when needed.
Hopefully, the examples above answered some of the questions you might have had about how we generated so much success with our content update process. But to make things crystal clear, I thought I would also directly answer some of the questions we received.
I selected the examples above to hopefully illustrate the highly unsatisfying yet true answer to this question: it depends.
Sometimes, updating a post means adjusting its positioning to better cater to user intent. Sometimes, it means adding more content to make the post more comprehensive. Sometimes, it means rewriting the entire post and just recycling the URL.
Less often—but occasionally—it’s a simple matter of answering one key question that’s missing, structuring headers correctly, or writing a better SEO title. But in my experience with updating content not only for Databox but for other clients as well, having success after making only minor edits is rare.
To paint a clearer picture, I typically spend between 4-8 hours updating a single blog post. My goal—and the goal I’d recommend for anyone who wants to replicate Databox’s successes—is to improve each post I update considerably.
Improve your old blog posts by better optimizing them for search. Improve them by making them better resources for searchers. Even better: do both.
You can try it, but my guess is that you’ll probably just be wasting your time. Google knows what your article looked like before and after you updated its date, and it isn’t so easily fooled.
Beyond that, Google’s John Mueller has explicitly stated that Google doesn’t favor fresh content, so just having a recent date on an old piece of content is unlikely to have any significant impact on your rankings.
Pete Caputa—Databox’s CEO—typically shared the posts we updated through his own Twitter account, but other than that, we did nothing to promote the updated posts. We did, however, republish each updated post and ask Google to recrawl them.
The answer to that question is another entire blog post of its own—one we’ve already written:
While we initially used Animalz Revive to find posts in need of an update, we’ve realized since that we can get the same data using Databox.
Check out these instructions for how to find blog posts that need to be updated using Databox, or grab this free Decaying Site Pages & Posts dashboard.
Databox is also a great tool for tracking and sharing the results of updated posts. Here’s the Databox dashboard I use for client reporting:
If you want to create this report in your Databox account, here are the instructions for how to measure the performance of updated blog posts with Databox.
Some of the posts we updated saw dramatic gains. Some saw slight gains. Some were fairly static year-over-year. A couple continued to decline.
For the ones that continued to decline, it was usually because of heavier competition in the SERPs. I think if we had run a link-building campaign to secure more links to the declining posts, we would have had more success with those updates.
So will every post you update see dramatic gains? No. But if you’re doing it right, the traffic and rankings for your updated posts should start to climb—and your overall traffic alongside them.
When traffic stalls, the solution too often is to just start publishing more content: increase your publishing cadence, bring on freelancers, etc. But as these results show, sometimes the better approach is to just take care of what you’ve already created.
You don’t have to always be publishing new content to succeed with SEO. Google doesn’t reward quantity; it rewards quality. If you turn some of your budget and effort toward polishing what you’ve already produced, you just might start to see your static traffic trending upward.
You might also like:
Marketing | Jul 2
Marketing | Jul 1
Marketing | Jun 30