I like to say that advertising is not how we make money, but how we avoid losing money. But that’s no longer the case, because of what I suspect are changes to Google’s algorithm to promote utter crap instead of good original work. Our long-term goal is to improve our sustainability by leaning into membership, so this is going to be my biggest ask of the year: If you’re not a member yet and you love this website, please become a member.
This is the only way a site like ours can continue far into the future, given the recent change from Google. It’s not just us, of course, because other publishers are seeing the same thing. The difference is that we’re a member-supported business and have a lever to pull that most others do not.
This all gets a little complex, so I’ll try to make what’s happened as simple to understand as possible. Because Google is a black box, I’m just inferring based on what I’ve seen and what other publishers are telling me.
Google Discover Has Been A Huge Help In Growing This Website
First, an acknowledgment. Jumping from a popular website to start a brand new one in 2022 was a brave move for Jason and David, and I have a deep respect for those two and Beau for embarking on this journey. It became a successful and beloved website faster than I could have ever imagined, looking in from the outside. It turns out there are a lot of people who miss websites that were written for people and not algorithms.
The algorithms are important, however. We do a great job of turning people into loyal readers and turning those loyal readers into members. It’s a wonderful thing. The challenge has always been finding new readers, especially when the old levers of social media are broken. We don’t have a long tail of search, a Facebook page that got millions of readers in the initial social heyday, or many of the other advantages competitors have.
Google, though, provided a wonderful tool for us. This was Google Discover. It’s a feed that appears across Google’s many products that’s personalized to you. Without doing anything other than publishing, we were rewarded for writing great and unique stories that people loved to read. We were rewarded primarily with traffic, and it helped us grow the site from something that was a fraction of the size of the industry stalwarts to a site that was recently bigger than many of them. For a few years, our site’s excellent work was rewarded in that it was put in front of the faces of many people, who clicked and enjoyed our work.
And then this significant chunk of our readership coming from Google Discover mostly stopped working, and it stopped working around the time that David had a baby and went on paternity leave. We first assumed it was a short-term blip caused by being overworked and understaffed; what we later learned is that it was likely a Google Core Update that reduced our visibility on the platform (on top of already seeing search traffic dropping because of what we suspect is an AI-related cause). Just look at this:
That’s weekly traffic from Google Discover. You can pretty much see the moment when it happened. Suddenly, we went from having 4 million visits a month as measured by SimilarWeb — bigger than Road & Track, The Drive, and many much older publications — to 2-2.5 million. That’s a reduction of about 30-40%. To make it worse, when you drop down below a certain level, the amount you make per impression is less, so we’re being hit with the 1-2 punch of less traffic and lower rates.
What was great about Google Discover in the past was that it felt like a more even playing field. It was a place where our newness didn’t impact us as much because people clicked on our stories and read them. A lot! And then that changed. We don’t know why. We’re looking for technical fixes, but the fact is that this is happening to our competitors as well, so there may not be a specific or simple cure.
Here’s The Slop That Google Is Promoting Instead
I did a trial with a company that provides a monitor of Google Discover traffic to surface which stories are doing well. You can see what’s working above in the automotive category, and it’s largely AI hallucinations coming from websites that are in the “MFA” or “Made for Advertising” category. [Editor’s Note: We did cover that DIY plate also, to be fair. – JT] This is the worst of the worst, with terrible interfaces, awful toe fungus ads, and content that is both made up and boring.
This is the opposite of what Google tells us to do, both in terms of presentation and content. There’s a helpful section of Google’s own publication guide that talks about the company’s desire to surface “helpful, reliable, people-first content.” I think you’d all agree that’s what we mostly do around here. There are even more specific guidelines, called E-E-A-T, that we have always naturally followed [Editor’s Note: One of my founding documents here established the “EE Rule,” which requires that all articles be enlightening or entertaining, ideally both. -DT]
Google’s automated systems are designed to use many different factors to rank great content. After identifying relevant content, our systems aim to prioritize those that seem most helpful. To do this, they identify a mix of factors that can help determine which content demonstrates aspects of experience, expertise, authoritativeness, and trustworthiness, or what we call E-E-A-T.
Of these aspects, trust is most important. The others contribute to trust, but content doesn’t necessarily have to demonstrate all of them. For example, some content might be helpful based on the experience it demonstrates, while other content might be helpful because of the expertise it shares.
I can’t even link to the sites that are being promoted because I don’t want to be associated with them, but it’s really bad, and little of the content there seems true, trustworthy, or written by or for humans. There is no Mustang Pickup truck coming, nor a 2026 Chevelle SS, and the new Corvette doesn’t look like this.
We can’t do this, obviously, and we wouldn’t. Not only does Google tell us it’s a bad idea, but as journalists, we are ethically opposed to it. I also think it’s long-term bad business. What is valuable is a genuine audience, and we’ve mostly built that.
One alternative is to do “Trending” stories that are mostly rewrites of TikTok stories. The biggest car website on the web (I won’t name them) does this. It’s not that any of the employees there like it, but it keeps the bills paid. Could we do that? The problem, when you look at the numbers, is that they’re pushing out 100-150 of these a month, and they’re only hitting about 2-3 times a month. That’s a terrible ratio, and we’d have to drown The Autopian with slop (even if it’s slop written by people, and not AI).
We may not always be the largest website, but our audience is real and we have more direct traffic than even some bigger sites:
I’ve anonymized the data, but those are three other websites in our competitive set. We’ve been around the same size for the last year, although we’ve been bigger than the other sites at various points. You can see that we have a much larger share of direct traffic, even if organic (Google products) is lower. In general, we have a lot of direct traffic, which is good. We also have a lot of engaged time as people read our articles. Most of our stats have stayed approximately the same, and just to make it more confusing, we’ve performed better on Google News since Google rolled out the Core Update we think hit us.
There’s just more AI-powered slop, and human-powered slop, and it seems to be pushing out our human-made, non-slop writing.
We Were Kicking Butt, And Now We’re Kicking Less Butt
I felt great in April, because it wasn’t quite clear that this was happening yet. We’d had a long run of sustainability as a business, and it seemed as though we’d succeeded in our original mission. Memberships and partnerships were growing faster than planned, and so long as we could stay roughly where we were with ad income, we’d be en route to a great year. We even turned off ads for members, because it didn’t seem like we needed that revenue.
Now, that source of once reliable traffic is way less reliable, for no discernible reason, and it’s knocked out one of the legs of the stool.
As I said above, my goal here is to get you to become a member if you can afford to and haven’t already. Based on traffic patterns, there are a lot of you who read this site daily. About 12% of you are members, which is awesome, and I’m so grateful we have that many. It allows us to paywall very few articles so that more people can read the site, and it has allowed us to hire great writers.
If we can go from 12% to 24% then we’ll cover a lot of the gap we now have to cover because of this loss of Discover traffic. If we can get to 33% we’ll be in a position to support more writers with freelance work and, hopefully, more full-time work.
No one has had a better chance to make a great car website than us, and I think we’ve done it. If this doesn’t work, I’ll be haunted by the reality that either we’re not smart enough to figure it out, or no one is, because it’s impossible. We’re so close to getting there, and any help you can provide will be returned with more of the kind of work that brought you here in the first place.
Thank you. Now, if you’ll excuse me, I have to go read about the two vegetables that’ll kill all this arm fat, on the way to the Oldsmobile dealer to pick up my new Cutlass 442.
Top Photo: AI nonsense, Deposit Photos, Jason Torchinsky
First Appeared on
Source link
