Hello, and welcome back. Jessie and Shelby here, back from the final (😭) weekend of summer. Jessie spent these dog days on a lake, enjoying a kayak, campfire and cooking Vegan Balthazar. Meanwhile, Shelby went to a Blue Jays game by scoring last-minute tickets and built 25 per cent of a daybed by herself. Eldest daughters, unite! We don’t need help!

This week: Does evergreen still work? We convened an incredible lineup of SEOs to ask the question, “in this AI era, what’s the value of durable, repeatable and updatable content for publishers?” Find expert takes from Mark Williams-Cook, Louisa Frahm and Harry Clarkson-Bennett below.

HAPPENING SOON: Barry Adams will join our September 30 community call to talk updates to Google Search Console, presented in partnership with Trisolute News Dashboard! Stay tuned for more info.

Subscribe now

Throwing it back to one of our first-ever published topics: our love of trees. Photo by Michael Benz on Unsplash

Let’s get it.

THE INTERVIEW

WTF is SEO?: Explain in your own words what evergreen content is.

Mark Williams-Cook: Evergreen content is content that is relevant and can be updated a few times over a period of years with very little structural maintenance.

Something you can publish and reuse over time, unlike news. “How to make a great vegan lasagna,” for example. That’s not going to change too much over time.

Louisa Frahm: I’ve advocated for evergreen content alongside utility content throughout my news SEO career. Evergreen content is durable, authoritative content that can remain relevant over time.

It’s not content that needs to be massively revised on a regular basis, but if you want it to pop up in traditional search and AI modules during timely windows, it needs to be regularly refreshed, updated, and recirculated.

Harry Clarkson-Bennett: Resilient, repeatable content that provides value year after year.

WTF is SEO?: What historically has been the real value for publishers of evergreen content and how is that different now?

Louisa Frahm: Historically, it was a good way to showcase your brand’s expertise and unique value, and then create targeted topical hubs that can build long-term authority. The primary performance metrics were clicks and accumulated traffic over time.

Where that’s shifting — because the entire industry is shifting — especially with AI Overviews, is that click-through rates are down. We can’t evaluate search performance by clicks alone.

Even though clicks are down, effective evergreen content can still feed into brand visibility and reputation.

I previously conducted these types of search strategies at the LA Times. When a wildfire hits Southern California, they are known as an authority in the area. People know the LA Times will have an evacuation map or a wildfire preparedness guide. That connection goes beyond just clicks and to E.E.A.T: who has the experience, expertise, authoritativeness, trustworthiness? Where do you go when stress is high and you need timely information from a reputable brand?

In the past, the mentality was, “hey, this can build clicks over time” and that’s how you would evaluate it. But now, because of AI, the value proposition plays more into brand visibility and brand reputation. The way that you evaluate it has changed, especially as it’s become more important to build 1:1 relationships with readers.

Harry Clarkson-Bennett: Publishers were really news focused, which makes sense.

But then people realized how powerful publisher sites are in the search landscape. Evergreen content became far more valuable for publishers, particularly when people in more senior positions realized just how powerful a channel Google had become, and the volume — particularly in a last-click channel — that’s quite rare.

Publishers start going really so far out of their lane. It’s not wrong necessarily, it’s just the byproduct of this kind of thing.

Now, there’s a reckoning: “Does this really add value? Is this what your audience cares about?”

Mark Williams-Cook: The value of evergreen content used to be that you could drop users directly into friction and get them to do what you want.

It’s a bit like how supermarkets always put bread and milk at the back of the store so you have to walk past everything else, with the sweet stuff right at the till. This is what we do with websites: they create friction to get people to do the things that we want them to do.

All you had to do was put out evergreen content and you’d get rewarded. In a way, it’s almost an anti-visitor, anti-consumer design. As a user, you don’t really want that friction. If you could tell someone to go to the store and bring your stuff to your car, that would be great. Which is more of the paradigm we’re in now with the AI era. That changes the value proposition massively.

WTF is SEO?: What do publishers need to think about when it comes to the value proposition of evergreen?

Louisa Frahm: When everything can be summarized, it’s important to know what makes you you. What makes your brand special?

When I was leading SEO at ESPN, it was critical to know why people came to ESPN specifically. There was a large emphasis on stats and the fan trust that had been built over time, from the on-air product to the digital experience.

You can’t overcomplicate the process of understanding your audience, assessing what their needs are, and evaluating how your brand can specifically connect with those elements through your individual expertise. We talk a lot about this in the AI era because everything is so easily summarized in commodity content. What can actually stand out?

In the AI era, and for some time even before this boom, we’ve had more and more publishers all gunning for the same topics. We’ve seen how highly competitive the industry has become. Whether it’s ESPN with sports and stats, or the LA Times with Southern California culture, you need to know and really understand audience needs, and then you need to elevate your own unique expertise and connection to those themes within your coverage. I think that’s where the biggest value lies.

Harry Clarkson-Bennett: It used to be relatively straightforward to give an idea of how much traffic you’re going to generate from a specific set of pages — not perfectly, but reasonably. However we look at it, that has diminished over time.

The first protocol is you have to be really clear about what we want to create and why. What purpose is it serving? Who is it for?

You also have to have a really clear idea about what the content is doing. If it’s just for SEO purposes, that would be an amber flag. You need a more nuanced approach.

A big part of that comes down to something that publishers have always done that I quite like: the composite metric for success that takes in 50 different things and every six or 12 months you review it and tweak it.

I like the idea of having something like that for your content so that you can start to value the right things and get the desk to focus on the right things.

For something like evergreen content, are you really likely to get a macro-level conversion from it as a publisher? In some cases, yes, but I think that is a harder sell now in a more commoditized age, so you need to have some pretty clear micro-conversions for that.

On the quantitative side, we have to take engagement data a lot more seriously: reducing the value of traffic and increasing the value of engagement.

Then there’s a qualitative look: Does this fit your brand strategically? Does it fit what your audience cares about? Combine those two things to get a good steer on whether you should invest more or less in this category.

Mark Williams-Cook: For our SEO clients, we go through a brand process to understand what they are. So many clients come to us and don’t really know why they’re different or why people should choose them. You cannot rely as much on that friction to create value as much as you used to.

AI has reduced the difficulty to follow what everyone else is doing. You can build tools, workflows, monitoring systems, create briefs, use data to get you 90 per cent of the way there to have something ready to go. Because everyone can do it, it immediately becomes worth a lot less.

AI can’t lead with new ideas, content, perspectives. The reason it works is, 10 years ago, you could publish good content and it would be a document on the internet that would mainly rank based on how many links it got.

The internet now is like the dominoes are stacked closer than they ever have been: if you put really good content online, because of TikTok, YouTube, Instagram, everything is so tightly connected that really good ideas tend to pick up through whatever channel a lot easier than they used to.

There’s more risk in not trying to do these big things than there ever used to be before. We tend to work on these pieces multi-channel, which gives you some resilience for when SERPs change.

The engagement we get clients do something that makes people react lifts everything.

Louisa Frahm: What has really shifted is that we used to define performance by, “Did it do well on search? Did it rank? Did it get into Top Stories?” The front page of Google was a primary channel for measuring performance.

Now, you should still make search-friendly content that’s based on user intent and the timely needs of that moment, but you shouldn’t stop with the first page of Google in terms of reach and where you might potentially connect with target audiences. Content that was originally grounded in search can also do really well on other platforms. You need to know your brand and understand which other channels make the most sense for cross-promotion.

For a long time, newsrooms have been siloed between search, social, video and other business segments. But because Google isn’t producing the visibility and results that it once did, we need newsroom departments to work much more closely together to elevate timely breakout opportunities, reinforce brand authority and prevent workflow redundancy.

That’s a bigger part of the evolution of the game with this type of content: thinking beyond just the first page of Google.

WTF is SEO?: What are the specific things that are part of the new checklist for evergreen SEO?

Mark Williams-Cook: In AI platforms especially, recency — or freshness — of content being updated seems to be a big factor in determining what’s cited. On their own outside of search engines, [AI platforms] have no real way of telling what’s good, bad or spam. If the content is similar and one was updated a week ago and another a year ago, probably the week-ago one is better.

We know for a fact that Google has a “last significant update” tag. They use a binary yes-or-no flag when they look at content to decide if it has had a significant update. Update the last modified date, and that’s what gives it that bump.

One thing available to us now with new tooling that was previously very difficult to do at scale is to monitor content for intent shift. COVID-19 is a great example. In 2020, everyone’s desperately Googling signs. Then over a few years, all the searches changed to, “Can I go outside now?” The shift in the questions people had is vastly different. That happens both seasonally and as the public zeitgeist shifts. We need to keep evergreen content relevant and meet these standards.

Louisa Frahm: We’re in the multimodal era of search now. Text might have been enough in the past. If you want guides to surface in AI platforms and modules, tapping into multimodal search best practices is important.

The text is still essential, but incorporating images and video, making the overall user experience as engaging as possible, as well as giving AI platforms diverse content to cite is an important evolution. This multimodal search development is pushing publishers to incorporate more content formats beyond just text.

Harry Clarkson-Bennett: Although publishers have been vaguely decimated, they do have an audience. Not that many brands have an audience. The byproduct of having an audience, particularly if you’re a subscription organization, is that you have opportunities to create unique data, which is normally quite difficult to do.

If you have a willing panel of subscribers, the joy is that the audience feels like they are part of the paper. It’s reassuring for them to get data from people they feel a level of kinship with.

You can’t get away from the fact that structure and headers are how bots understand your article and content quickly and efficiently. While we want to do brilliant things, use audience data, have videos, and have strong E.E.A.T, the fundamentals still matter. If you’re a big publisher, the better you can make these things, the better.

WTF is SEO?: What is one thing publishers need to stop for SEO?

Harry Clarkson-Bennett: Stop chasing clicks and straying outside of your lane. You’ve gotten too big and bloated. It’s time to nestle back into who you are. This is the sweet spot.

Mark Williams-Cook: Use AI less.

Leaning on what Harry said, I’m seeing so many websites get bloated with content that is just meh. It’s not adding anything new. In isolation it looks good, and it feels cool that you managed to pump out that article so quickly, but it’s not going to do you any favours. Less is definitely more now, because that’s the biggest problem Google’s facing at the moment: this humongous tsunami of content, trying to figure out what’s good and what’s not. The easiest way they can do that is just exclude a lot of it. You don’t want to be in that pile.

WTF is SEO?

Did AI kill evergreen content? Let's discuss.