Episode 20 · The Search Signal
About this episode
A marketing team checks its dashboard and organic traffic is down this quarter. Rankings look fine and nothing obviously broke, so the question landing on the SEO lead's desk is whether the program stopped working. Two things are happening in search measurement at the same time that make that question harder to answer than it used to be. Searches are increasingly ending without a click, and the tools used to measure visibility, like rank trackers and Google's own Search Console, have gotten less reliable over the past few months.
In this episode of The Search Signal, Michael Transon makes the case that search is turning into a brand-building channel instead of a trackable performance one. Watch to learn why rankings can hold steady while traffic falls, how to tell whether AI and search visits are hiding inside your direct traffic, and how to build a monthly reporting routine that reads every signal together.
Michael's POV in 60 seconds
One thing
A marketing team's organic traffic drops this quarter and leadership wants to know if search stopped working. Rankings haven't moved and nothing obviously broke, so the easy read is that the program is failing, even though nobody can say yet whether that's true.
So what
Two changes compound to make that easy read unreliable. Click-through rates and search volume no longer predict traffic the way they used to because more searches end without a click. Plus, the tools built to measure visibility, like rank trackers and Google Search Console, are less complete and less consistent. A meaningful share of what looks like lost traffic is more likely sitting in the direct channel, mislabeled because AI tools.
Now what
Build a monthly reporting routine that triangulates rankings, AI mention rate, every traffic channel including direct, and what buyers say when asked how they found you, then read them together as one trend instead of expecting a single number to represent your search traffic. Agree with leadership on that scorecard before the numbers come in, so everyone judges a dip in traffic by the same standard.
Questions this episode answers
How do I know if a traffic drop is a real problem instead of a tracking or seasonal issue?
Check whether Search Console clicks and Google Analytics visits are moving together before reading anything into a traffic drop, since a mismatch usually means a tracking tag or a cookie banner broke. Compare the period against the same months last year rather than the prior period, so a seasonal dip doesn't get mistaken for a performance drop. Log the dates of known tracking breaks, like the week ChatGPT stopped passing referrer tags, so nobody mistakes a tracking change for a real one.
How do I measure AI visibility when there's no keyword to track?
Pick a set list of questions buyers ask when choosing a company like yours, run each one at least 10 times a month across every AI tool your buyers use, and track how often you're named as a rate rather than a single ranking. That rate will move month to month, since any individual AI answer can change the next time someone asks, sometimes within the same day.
How do I tell whether my direct traffic is actually hidden search or AI traffic?
Open the traffic acquisition report in Google Analytics 4 (GA4), filter to direct, and break it out by landing page, watching for growth on pages with long, specific URLs nobody would type from memory. That pattern, direct traffic that's been flat for years and suddenly climbing, is a sign AI or search growth is landing there instead of getting credited correctly. When ChatGPT links briefly stopped passing referrer tags for about a week, one product leader at Adobe watched visits credited to ChatGPT fall by roughly 30,000 a day on his tracked pages while direct visits to the same pages rose by about 26,000 a day.
Chapters
This Week in Search: A Shared robots.txt Line Opts Sites Out of AI Training
This Week in Search: Search Console Still Gives No Clarity on AI Clicks
This Week in Search: Reddit's Traffic Falls Despite Staying Highly Visible
Change One: Click-Through Rates and Search Volume Are No Longer Predictable
Change Two: The Tools That Track Visibility Have Gotten Less Reliable
Why Search Is Becoming a Brand Channel, Not a Performance Channel
Why Direct Traffic Is Likely Hiding AI and Search Visits
Why Shifting Budget to Paid Search Doesn't Close the Attribution Gap
Measuring Search Now Means Triangulation, Not One Number
Step One: Clean Up Tracking and Set a Baseline
Step Two: Report Visibility Every Month
Measuring AI Visibility: Pick Questions and Track Your Mention Rate
Step Three: Track Every Traffic Channel Side by Side
How to Read the Direct Traffic Channel Correctly
Step Four: Ask Buyers How They Heard About You
Step Five: Turn the Signals Into One Story
Which Signals to Weight When They Point in Different Directions
Step Six: Agree With Leadership on How Search Gets Judged
Why Losing Exact Attribution Doesn't Mean Stop Investing in Search
Resources mentioned
Take this further
Don't miss an episode
New episodes every week. Get them in your podcast app of choice, or watch on YouTube.
Full transcript
Michael 00:00 – 00:39
My perspective on this after doing this for almost 15 years is search work is starting to look a lot less like a performance marketing channel where you could track the click to a sale and it's starting to look more like a brand channel where a lot of what it does is putting your name in front of buyers who end up deciding later and do so without a click that you can trace. Hey, welcome back to The Search Signal.
Michael 00:39 – 01:33
I'm Michael Transon. I'm the founder and CEO of a search marketing agency called Victorious. And here at The Search Signal, our goal is really simple. We're here to help marketers uh understand and also take action on what is happening in this big, quickly evolving world of search marketing. We ourselves have been running campaigns for nearly 13 years. We also work with some of the fastest growing and biggest brands out there. So what we do is we bring our experiences. From working with those clients, plus a lot of our own first party research to The Search Signal to help you with your own work. So today I want to answer a question that I'm hearing a lot from a lot of different marketing leaders right now, which is if my organic traffic is down, is my SEO still working? And to answer that, we have to talk about two important things that are happening in search measurement right now at the same exact time.
Michael 01:33 – 02:33
The first one is that more and more searches are ending without a click. And that has changed numbers that we used to be fairly and confidently certain about. Things like uh how many people click on each search and how many searches there are in your industry or your category, and then also how people just even use Google in the first place. And then the second is that the tools that we have used to see how visible we are, meaning like the the rank trackers and even Google's own Search Console, they are giving us less reliable and less complete data than they have really ever given us. So we're gonna go through uh both of these changes and we're gonna talk about what's causing them. And then we'll also talk about how we as marketers should measure search now that it's turning into, in my opinion, more of a brand channel than a performance marketing channel.
Michael 02:33 – 03:06
This is a really honestly strange time to be measuring search. And it's because the way that visits are getting tracked and credited is just changing. And it's changing at the same time as the search experience itself. But I also think that this is the most exciting time to be investing in search in a very long time. And I feel that way because Google's AI features and tools like ChatGPT, they are growing faster than anything I have personally ever seen.
Michael 03:06 – 03:52
And buyers are getting more and more out of them than most of the other ways that they research what to buy and how to buy it. So before we dive in and get into these two changes that are happening in search measurement, let's first start by talking about what happened this week in the quickly evolving world of search marketing news. Let's start with a new way that just came out this week to opt your website out of AI training. So on October 30th, Google, OpenAI, Microsoft, and Anthropic all agreed to uh honor one shared line in your robots.txt file, which if you're not familiar, uh, this is the file on your website that tells search engines and also all the AI bots what they're, you know, functionally allowed to read. Now that one line can opt you out of training.
Michael 03:52 – 04:37
For all of those four companies all at once. And this is instead of you having to write a specific or separate rule for each of the different companies. Anthropic is honoring this right now. And then the other three that said that they're going to start honoring it sometime next year. What I would point out is once Google supports this uh shared line in robots.txt, they plan to treat it as also blocking Google-Extended unless your file has its own rules specifically set for Google-Extended. Now, what is Google-Extended? Um, That is the setting that controls whether your pages can show up in AI answers in the Gemini app. And it does not affect just regular Google search or AI Overviews, which you know we're probably familiar with.
Michael 04:37 – 05:13
Those are the AI summaries that are at the top of Google's results. So if your visibility in Gemini happens to drop next year, One of the first things that you will want to check is whether the line that you set in the robots.txt file that we just talked about is causing that drop because you won't see anything in your analytics that are going to point you to robots.txt and that line being as the reason for seeing the drop. So next in the news this week, there is an AI report in Search Console. I'm not sure if you've seen it or not, but uh Search Console, if you're not familiar, this is Google's free tool for seeing how your site performs in search.
Michael 05:13 – 05:46
And today's news is whether or not that report is ever going to show actual clicks. So right now, the report shows how often your pages are appearing in AI Overviews and AI Mode. Uh, And AI Mode is Google's chat style version of search is at the top now. And what it doesn't show is whether anyone actually clicked on your website. So this week, Google's team said that they want to give more data, quote, they said data that makes sense. But they also said Google has nothing to announce around clicks yet.
Michael 05:46 – 06:23
So some people read that as Google saying click data isn't coming at all. And John Mueller, who is also on Google and on that team, said it was a weird way of reading it, and that's not what they meant. So nobody outside of Google really knows what's going to happen yet. And for now, nothing Google gives you shows you really how much traffic its AI features are sending to the site separately from regular search. So your AI reporting for now still has to work without that number. And third, this is very important for today. Reddit, which reported its earnings on uh October 29th.
Michael 06:23 – 06:56
Let me explain this. So Reddit is one of the sites that Google, as we probably all know, shows most often. And its CEO, Steve Huffman had said that its search traffic, Reddit's search traffic was down meaningfully this quarter. And he said it was because of you guessed it, AI Mode and AI Overviews, zero clicks. Um, He said that he expects it to keep falling next quarter. So Reddit's plan is to focus on the people who go straight to Reddit without searching and do a logged in only experience.
Michael 06:56 – 07:27
And funny enough, the stock fell like 15% after the earnings report came out. So why do I bring this up? Uh Reddit is still all over Google search. They're one of best performing sites, and they are still losing search traffic. And this is the same pattern, maybe a lot of you are also seeing on your own website, this trend where you might be as visible in Google as you ever have, but getting fewer visits from it.
Michael 07:27 – 08:05
So I picked these three news reports this week because each one of them is just about the gap between how visible you are and what your reports could actually show you. And because of that gap, we're gonna talk about this today, but traffic, search traffic on its own can't really tell you fully whether your search program is working. So let's dive in. let's start with the first change, which is that um a lot of the numbers we used to be fairly certain about uh in search that just are not reliable anymore.
Michael 08:05 – 08:44
For uh years, click-through rates, for example, which is the if you're not familiar, this is like the share of people who click on a specific result in the search results. You know, the first result, the second result, so on. Those click-through rate curves have stayed fairly steady. So if you knew where you ranked and you knew how many people generally searched for a keyword, you could calculate and predict roughly how much traffic you would get. And, most people used Google the same way when they would ever go to Google.com, which was type in a search, look at a list of links, and then click on one of them.
Michael 08:44 – 09:14
That has changed because more and more of these searches are ending without a click. And that's what I mean when I say and people say things like quote zero click. Uh The results page looks a lot different now. There's a AI Overview at the top that answers the question. And since I think late August, Google has been opening up some of those overviews to full lengths automatically, which pushes the regular results even further down the page.
Michael 09:14 – 09:50
So point being, the same ranking is getting you less real estate and then fewer clicks than it did just a couple of years ago. And then on top of that, some of the searches that used to happen on regular Google are happening inside of AI Mode or on ChatGPT instead. So there are fewer searches that are happening in traditional search in your category or industry to really begin with. That's what happened to Reddit this quarter. right? It's still one of the sites that Google shows the most on the internet. And its search traffic fell anyway, which, like we said, its CEO had tied to AI Mode and AI Overviews.
Michael 09:50 – 10:55
So your traffic can fall. And it can fall a lot further than your visibility falls or changes. In fact, your rankings can actually hold steady. Your visibility can even go up, and your traffic can still go down because fewer people are clicking and fewer people are searching in traditional search. And you know, Your SEO team or your marketing team didn't cause either one of these changes. It's just what's happening in the market. And fewer clicks on top of that, in my opinion, doesn't also mean search is doing anything less for your company or your marketing program because people don't have to click anymore to decide who they want to work with. And a lot of them will just see your name in an AI answer, and then come to you later through a different or another channel. So now that we've covered that first big change that's happening, which is that clicks and searches don't behave the way that they had used to, I wanna also talk about the second change, which is that the tools that we use to see our visibility have gotten a lot less reliable.
Michael 10:55 – 11:37
Okay, so if traffic can't tell us whether our search program is working, the next place that you would look at is visibility. Which means where you are ranking in Google and then also how often AI tools mention you. And the tools that we use to measure visibility have also gotten less reliable over the last few months. So tools like uh Ahrefs and Semrush, uh, two examples, they track rankings by sending uh automated searches to Google and then they just record which sites come back in which position. But starting in July of this year, Google began routing the links on its results pages through its own redirect.
Michael 11:37 – 12:13
So a tool reading the pages sees a Google address on every result instead of the website that it leads to. And then on top of that, at the end of August, uh Ahrefs said that the change had caused, quote, a temporary inconsistency in its data. Then on top of that, in September, uh rank tracking companies started reporting that Google was sending result pages with the wrong data in them to the tools it suspected were trackers, which we talked about last week in our episode about our Q3 Quarterly Search Report. Now, Google, like we said before, has not said anything about this.
Michael 12:13 – 12:47
Uh, And the vendors' own estimates of how much those data sources are affected. They run from as small as, which I don't totally believe, one to two percent at some companies to what I believe more, which is like a fifth or more at some other ranking companies. So point being is the same keyword can show a different position depending on which one of these tools you buy for and which of these tools you end up asking. Now, another tool, Search Console, that's Google's internal product. It comes straight from Google, but it only covers your own site.
Michael 12:47 – 13:25
So you can't see your competitors in it and the position there on top of that. That's an average that GSC uses across every search you appear for. And then on top of that, it also counts AI modes and AI Mode clicks together with regular search. So you cannot separate them out. And its AI report shows that you appeared in AI Overviews or in AI Mode without showing, like we said earlier in the news, whether anybody actually clicked. We used to be able to put a third-party tracker and a Search Console report together to get a full picture of where we stood.
Michael 13:25 – 14:01
And right now we can't do that with the same confidence we've been able to do in the past. Now, AI search, on the other hand, is even harder to see. And that's because the AI companies are running their own database of web pages and nobody outside can see into them. And then on top of this, we've talked about this before. A ChatGPT conversation, does not have a keyword that you can track. Uh, And that's because like your company might come up, I don't know, 40 messages into a conversation, and that conversation's private.
Michael 14:01 – 14:36
So the best anyone can do for AI visibility at this point at least is a like a representative sample where you are running the questions that your buyers ask. And you run them a lot of times and you see how often you're named. Now, as we know, if you've been listening to any of these episodes, we talk about this before, but AI tools give a different answer almost every time you ask. So getting named once does not mean you're gonna be named the next time. And we're not gonna get data on individual AI searches the way we've had it for Google, at least anytime soon.
Michael 14:36 – 15:31
I can't see how we'll be able to. So that's the second big change that we've been seeing related to tracking. And now that we've covered both of these changes, which are that again, traffic is falling further than visibility. And that visibility itself is also now harder to see, uh, let's talk about what that means for how we actually start to judge search today. My perspective on this after doing this for almost 15 years is you put those two changes together and search work is starting to look a lot less like a performance marketing channel where you could track the click to a sale and it's starting to look more like a brand channel where a lot of what it does is putting your name in front of buyers who end up deciding later and do so without a click that you can trace.
Michael 15:31 – 16:13
The problem is that a lot of companies are still judging it with performance marketing metrics. These are things like just looking at organic visits and only giving credit uh for a sale or a lead to the last thing that somebody clicked. And those metrics miss a lot of what search is doing now for a business. A lot of buyers hear about you in an AI answer and then they're going come to your site later through a different channel. We've had leads that come into Victorious through one of our Google ads, for example. And then we ask how they heard about us, and they say, Claude. So in that situation, the ad platform, if we were to track it that way, took full credit for the leads that started actually in an AI answer.
Michael 16:13 – 16:47
And if we had not asked them how they heard about us, we would have never known the answer. And now the other place a lot of buyers end up is a very important channel moving forward, which is called direct. This is the channel in your analytics that it uses for any visit that it can't trace back to a specific source. So a visit gets filed as direct when it arrives with no referrer, which if you're not familiar, it's like a note that a your web browser passes along about the page that somebody clicked from to get to your site.
Michael 16:47 – 17:25
And no tracking tag on the link. So one analytics vendor that uh, I was looking into this that was tracking AI traffic said that I think somewhere around seven in 10 of the AI visits that it could identify came in with no referring tag. And I would say, based off of what I'm seeing, the phone apps, like the app, the ChatGPT app on your phone, those are the most likely to leave the referrer off. And we know this because the Gemini app on the iPhone doesn't send a referring tag at all. So a lot of your AI traffic is probably sitting in direct.
Michael 17:25 – 17:56
And only some of it, maybe two or three out of 10 visits are showing up in the AI assistant channel that Google had added into Google Analytics in May. And then in October, just a couple of weeks ago, we did get something close to proof that AI visits landed in direct. So you might have seen this in your own uh website analytics for about a week in October the links in ChatGPT stopped carrying the tag that tells your analytics a visit came from ChatGPT.
Michael 17:56 – 18:31
There was a product leader at Adobe. And uh, when he noticed this, he looked at about 500 pages on their site where ChatGPT was sending like a meaningful share of traffic to the site. And during that one week, visits credited to ChatGPT on those pages fell. I think he said like 30 or 31,000 visits a day. And direct visits to the same pages rose like 26,000 visits a day. Also, he looked at the activity and the new direct visitors also behaved a lot like the ChatGPT visitors had before the change.
Michael 18:31 – 19:05
So he read the shift, and I also read the shift as the same people just getting filed under a different label. And we've honestly suspected for a long time in the SEO community that direct hides regular search traffic too. Um Man, like 10, 15 years ago, there was a large e-commerce company. It was pretty well known at the time. They had taken their whole site out of Google search index, they had de-indexed the site for like six hours. And some of its pages had very long specific addresses, URLs that no one would ever type in from memory. And during those six hours, direct traffic to those pages fell by more than half.
Michael 19:05 – 19:45
Now, you think about it. If someone de-indexed a site, which would impact just the organic search traffic channel, why would direct traffic also go down significantly? There's only one answer. It's that more than half of those direct visits had been coming from Google the entire time. So taking a step back, a lot of the marketing leaders with big budgets right now are uh asking why, in this whole shift and change to attribution and tracking and visibility, why they shouldn't just simply move more money into paid search because the tracking still works there.
Michael 19:45 – 20:28
And I would say, the Claude leads show that paid tracking has the same blind spot because it only covers the click that you paid for. And moving budget into paid doesn't get your company into the unpaid results or into the AI answers, which is where a lot of your buyers are deciding which companies to consider before they ever see an ad or ever have an opportunity to click on an ad. So, you know, now that we've covered why search is becoming more of a brand channel, in my opinion. And why the traditional performance metrics on their own miss a lot of what it does now. I'd like to spend some time talking about how we as marketers should measure it now.
Michael 20:28 – 20:58
So the way that we as marketers should respond is, in my opinion, to actually report on more than we used to. And that's because there's no single number anymore that can tell us whether search is working. Uh, I definitely recommend traffic and rankings need to stay in whatever reporting we do, but we need to layer in and add on the signals that are picking up where just traffic performance is missing. Uh, We used to honestly have a through line, like a very clean uh path from the organic click to the conversion.
Michael 20:58 – 21:49
What we have now is something that I would say is more like triangulation, uh, which means that we measure search from uh several different directions and then we weigh them against each other. And then we tell one cohesive story about whether search is growing. And there are six different ways I would recommend setting this up. So um, the first thing is housekeeping. We need to clean up our tracking and we need to set a baseline before we read anything. So the first thing that you should do is check the clicks in your Search Console and then the visits in Google Analytics to see if they are uh telling the same story and moving together. Because if they are not, you might have a tracking tag or a cookie banner problem that might have changed on your website. Once you've cleaned that up, the next thing I would do is compare against actually the same months last year, not last period.
Michael 21:49 – 22:21
Because we don't want to mistake a seasonal dip that happens in our business for a performance drop. And then I would uh add notes in Google Analytics of the dates of the known tracking breaks that we've talked about, like that week in October when uh ChatGPT and their links lost their tags. And then also at the end of September, maybe when Gemini stopped adding them, like we talked about earlier. And this is so that nobody on our team, executives or you know, tactical members of the marketing team, read tracking changes as a performance change because they're not.
Michael 22:21 – 23:05
Then I would pull at least the last year of data, two years if you have it. And that means how often your pages show up and how they show up in search and how many clicks they got, how many people searched for your company by name. I would also pull out direct traffic to each page. And then also, if you're tracking it, the leads or sales that came from search. Now every signal we're about to talk about gets read as a trend. And you cannot see a trend without a starting line. So that's the first step. The second step is to report your visibility on a monthly basis every single month. So for Google, I would use Search Console, not a third-party tracker, as your official source for your own rankings.
Michael 23:05 – 23:43
And we need to read that search by search, or page by page, and for a set list of the searches that actually matter most to our business. And then we would keep our searches that are branded, our company's name searches kept separate from everything else. Um, We can still use a third party tracker. I would recommend using it for less like general direction of your competitors and your category, but we need to make it a rule that nobody on our team reacts to a random drop in Semrush or Ahrefs until we have checked the same search in Search Console.
Michael 23:43 – 24:13
When traffic falls, the report should also say whether fewer people are clicking or fewer people are searching. So impressions are the number of times that your page shows up in the results. So if your impressions are showing steady, but your click-through rate is falling, it's fewer clicks. And then if your impressions are falling, it's fewer searches. So that's for traditional. for AI, I would recommend you pick a set list.
Michael 24:13 – 24:53
We've talked about this before of the questions your buyers ask when they are choosing a company like yours. I have so many podcast episodes about this so far. So I'm not gonna relitigate this, but just to keep it high level, we pick those searches and then we run at least one of them at least 10 times a month. And we do it in each AI tool that your buyer uses. What you wanna report on is how often you're named as a rate, right? So like out of I don't know, four out of every 10 runs you're named or not. And then watch how that rate of mention moves over months because, we've talked about this before, any single answer can change the next time that somebody asks, literally in the same day or the next moment after that.
Michael 24:53 – 25:38
That's the second step. The third step is to report every traffic channel that your visibility can affect. And this is so important. And we need to track it side by side each month because a buyer who finds you in search or in an AI answer can and will come back through any of them. So in Search Console, you can filter your searches to the ones that include your company's name and then track those branded searches as their own line. Since more people searching for you by name is a sign that your visibility is reaching buyers. And then in Google Analytics, we can use the AI assistant channel, which when they send the referring tags, can count visits from tools like ChatGPT and Gemini, again, only when they pass a referrer.
Michael 25:38 – 26:09
Um, I would treat that as the minimum AI is sending you. And this is because, like we said earlier, AI visits without a referrer, which is about seven in ten times, still land in direct. And if Perplexity, for example, sends you traffic, we need to uh have whoever runs your analytics add a rule that counts Perplexity as an AI because some people have actually found that the AI assistant misses that one specific AI tool. And now I would also keep paid in the same report uh as well.
Michael 26:09 – 26:44
And that's because a buyer who you know heard your name and an AI answer might come back by clicking on one of your ads, the way that our Claude leads did, like I talked about earlier. But what you are looking for is across all of these channels, a uh a rise in performance along with your visibility. And just to say one more thing about. the direct channel. For direct, we cannot hand all of it to SEO or AI. So the way that we and us at Victorious as marketers think about direct is as a trend line that sits next to our own visibility.
Michael 26:44 – 27:18
So like let's just say direct was at your company steady for years. And now it's suddenly climbing a lot faster than it used to, right? And maybe faster than some of your other uh more steady channels. Um And that's happening while your rankings and your AI visibility are theoretically going up. In a case like that, search and AI are the most likely reason why direct is going up, especially when the growth is on pages with, you know, for example, long or specific addresses, because people aren't going to be typing those in manually.
Michael 27:18 – 28:00
So to see which pages the direct growth is landing on, you can just open up the traffic acquisition report in your GA4. You can just filter it to direct and then you can break it out by individual landing pages. Um, But what I wouldn't do is take a fixed random share of direct and then just suddenly call it sixty percent of your direct traffic is suddenly gonna be search traffic because we don't know. every business is gonna be unique. The fourth and very practical step that I would recommend is actually just asking people how they heard about you. So for a B2B company like ours, or if you're in B2B, you could just make it a required field on whatever lead forms you're using. You could just list ChatGPT, Claude, Gemini next to some of your other channels.
Michael 28:00 – 28:37
Or you could just do like an open text field. It's really whatever works for you. Um, For an e-commerce company, you can also add a, you know, a one question survey after they check out. But store that answer in your CRM and I would put it next to the source that your tracking recorded. And then every month you can just look at how many leads you know, name an AI tool and how many of those you're tracking as credited to something else. Like again, I said earlier, the Claude leads that we get from our site that actually come through Google Ads. And that comparison can show you what AI is doing for you that your analytics just is not built to see right now.
Michael 28:37 – 29:10
And then the fifth step is taking all those first four steps and turning it into a story about what search is actually doing for the business. Because a traffic number, like we said, on its own is gonna just invite the wrong conclusion. So let's just say, for example, your organic traffic is down like 15% this quarter. Um, The version leadership can follow can say something like: uh, our rankings on the searches that matter most held steady this quarter and impressions in our category fell.
Michael 29:10 – 29:41
So what that tells us is fewer people are searching on Google. But searches for our company by name went up. And direct visits to our product pages went up. And more of our leads said they found us through ChatGPT or Claude. When you read that together, that says search is reaching more buyers while fewer of them click. And that gives leadership a valid reason behind the traffic number.
Michael 29:41 – 30:26
Now, when those signals point different ways, I think we as marketers should put more weight on certain ones than others. So for example, I would put more weight on rankings than on traffic when we are judging the efficacy of our SEO program. The whole point here is that uh, you know, zero click searches and search volume are mostly outside of anybody's control. So you can't judge whether your team is doing well by the things your team doesn't control. And also, rankings move first. So they can also show whether the work is succeeding months before leads or even revenue can do that. And I would also, again, we've said this before, read this story and this narrative on a quarter over quarter basis, because, like we've said many times, a single month can just swing a Google update or a holiday.
Michael 30:26 – 30:58
And that's besides the point that most search work takes uh a couple of months to show up in actual rankings. But when your visibility and your traffic channels and what your customers tell you are all moving uh the same way over a couple of quarters, you do have a case that you can defend that the search program is working. Even if you cannot trace a sale back to a single search that someone did on a single um uh time that they went to Google search.
Michael 30:58 – 31:31
And when your AI visibility is going up and let's just say your branded search is flat. The most likely uh explanation is time. And that's because buyers who hear your name from an AI tool usually don't search for you that same week. So I would give it a couple of months before you read anything into like a flat branded search number. Um, And when, for example, branded search and direct are going up and your visibility isn't, something else is probably driving the demand.
Michael 31:31 – 32:03
And that could be like a different marketing campaign you're running or a press release that got picked up by a bunch of publications. And we always want to make sure that we're not giving search credit for things that it shouldn't get credit for. I said I would give you six steps. That's five. And the sixth step is to take that story to leadership and agree on how search gets judged now and into following quarters. And this is before the numbers show up in the reports. If leadership is still grading search on traffic alone, the rest of the report just won't make any sense.
Michael 32:03 – 32:34
So, we need to set a target for each signal that we talked about, holding your rankings on your priority searches and uh raising how often AI tools name you and growing the number of people who search for your company by name and growing the share of leads who said that they found you through search or AI with the leads and the revenue from search alongside of them. Uh Traffic needs to stay on the scorecard. I recommend you keep tracking it, but it needs to be next to rankings and next to impressions.
Michael 32:34 – 33:08
So a drop in clicks or traffic gets read in the actual context. Then at the end of the quarter, you are grading the team against what everybody, you and leadership all agreed to. So when your traffic, your organic traffic is down, like we said, starting from the beginning, we have to rule out seasonality and broken tracking first, and then check in Search Console, whether your rankings dropped, or whether people are just clicking less or searching less in traditional search. Then we show what search is doing for the business by reporting on more than traffic.
Michael 33:08 – 33:41
This is your visibility in Google and AI, the traffic channels that it affects, including and very importantly, direct, and what your customers actually tell you about how they found you. And we read these together as one cohesive trend and one cohesive story. So anybody who thinks you can invest in search right now and get exact attribution for it, they're fooling themselves. I understand why that has been the expectation. And it's because Search was one of the most measurable channels we had for a long time.
Michael 33:41 – 34:13
And then it changed pretty quickly. And honestly, it sucks. And I'd still rather be investing in search right now than at almost any other point in the last 13 years that we have been doing this. More and more people are asking AI tools and search engines which company to work with, what company to to talk to, what product to buy. And the answers they get are more useful to them than most of the ways people have shopped for products and services anytime before. So measure search the way we talked about today.
Michael 34:13 – 34:44
And I would strongly recommend you keep investing in it. That's today's episode on whether your SEO is working when traffic is down. It's a big episode today. Thank you for listening to this episode of The Search Signal. If you're getting something out of our show, we would really appreciate it if you could just take a minute to rate and review it wherever you are listening, because that is honestly how other marketers are going to find us. And also, if this was useful to you and you have not yet, please go ahead and subscribe so the next episode finds you.
Michael 34:44 – 34:47
That is it for me today. I will see you again for the next episode next week.