Episode Overview: Google often provides different SERP elements at once to answer user queries to help its machine learning system derive user intent. This is known as fragmented user intent, and knowing how it works is a critical component to understanding user intent in its entirety. Join host Ben as he concludes his discussion with G2’s VP of SEO and Content Kevin Indig as they discuss the various elements of fragmented user intent and define the valuable concept of “Search maturity.”
Summary
- Fragmented user intent occurs when Google processes a user query and provides a selection of SERP features in response.
- Google monitors which SERP features users frequently click to identify user intent for a particular query, both new and old.
- Google adjusts SERP features frequently to meet everchanging user intent, sometimes as often as on a monthly basis.
GUESTS & RESOURCES
- Kevin Indig: Website // LinkedIn
- The Voices of Search Podcast: Email // LinkedIn // Twitter
- Benjamin Shapiro: Website // LinkedIn // Twitter
Ben: Welcome to the Voices of Search podcast. I’m your host, Benjamin Shapiro and today we’re going to talk to one of my favorite SEOs, who every time we talk, he’s got a different job. Joining us today is Kevin Indig, who is the vice president of SEO & Content at G2. G2 is the world’s leading B2B software and services review platform. And until recently, selecting business’ software was difficult and inherently risky, but using G2’s real verified user reviews, they help you objectively assess what is best for your business. Now, prior to taking his role at G2, Kevin was the head of technical SEO at Atlassian. He was the director of SEO at Dailymotion, and once upon a time, he was an SEO at the world’s greatest SEO SaaS platform, our friends at Searchmetrics.
Ben: Okay. So far this week, Kevin and I have talked about his new role running SEO and content at G2, and some of the valuable ways that you can use reviews and review sites to benefit your business. Yesterday, we talked about something that Kevin’s been speaking about publicly, which is reverse engineering user intent using the SERP. The takeaway is you can’t just think about what your keyword set is or what your on page content is, you need to think about how that content is going to show up in Google because they are interpreting what the user’s intent is. And today we’re going to double click into that topic and talk about what Kevin describes as fragmented user intent. Okay. Here’s the last part of my conversation with Kevin Indig, VP of SEO and content at G2. Kevin, welcome back to the Voices of Search podcast.
Kevin: Thank you. It’s good to be on as always.
Ben: I said I would ask you again. Hey, are you still at G2? It’s been two days.
Kevin: It’s been two days, I’m still at G2. Hopefully one day SAP.
Ben: You’re the man of many great jobs. I hope you stay at G2 for a long time. They’re lucky to have you.
Kevin: Thanks.
Ben: Good. Yesterday we talked about reverse engineering user intent and using the SERP to really understand what Google thinks the consumer’s intent is and building content around that. And that’s great if user intent is always very clear, but sometimes user intent isn’t always a static thing. Talk to me about your concept of fragmented user intent.
Kevin: Yeah, you got it. So in yesterday’s example, we talked about some very clear searches that indicate a very clear intent, like say Miley Cyrus site cuts photo or image, or something like that. It’s very clear to Google and for everybody else what people want in this case. However, when you just Google the words site cuts, it’s a very ambiguous query, meaning there are lots of possibilities of what people actually want in terms of what their search intent is or user intent is. The thing that we’ve been observing for a couple years now is that Google shows different SERP features to make sure that it covers whatever need or whatever intent people have and we call this fragmented intent. It’s when Google shows different SERP features like say a local pack of featured snippets, some shopping ads, and maybe an image carousel.
Kevin: It’s very hard to then understand what Google actually wants. And they do that to make sure that first of all, they serve a lot of the different search intents. But second of all, they also either rate over time, meaning the more people click on a local pack, for example, the more they will prefer to show that local pack and adjust their SERP features. It’s interesting because Gary Illyes from Google actually mentions something like that last year in San Francisco at the Bay Area Search Meetup, where he mentioned that actually they look at tab clicks a lot to understand what they should show in universal search, meaning what SERP features they should show from Google images or news. But he kind of had … Take this as my interpretation of this, he kind of gave away that Google looks at clicks on some of these SERP features or taps to indicate and understand what people actually want. So that’s fragmented user intent.
Ben: So that’s great, but Google can create the never ending SERP. How do we use that in SEO?
Kevin: That’s a good question. And by the way, I’m currently writing a blog article about this and how to do this. It’s not out yet, so you heard it here first, but the way that you can go about this is you can take a rank tracker like Searchmetrics, for example, and you can export all the SERP features for a query. And then you can cluster these SERP features according to different user intents. That’s where you need a bit of a subjective measure, but I think it works out pretty well, at least in the couple of examples that I’ve run through. And the way that it works is you would then for example, say, “Okay, what does a user who sees a featured snippet want?” It’s very probably to learn or understand something, right? It’s often more a bit of a shallow understanding, right?
Kevin: They might want to have a list or quick description of something. What about review snippets, right? That’s when probably the user wants to evaluate a product. And what about an image carousel? That probably also falls more into learning something or seeing something or getting a tutorial, right? So we are able to classify different SERP features into different categories or different user intents. And then we can measure that for a query. So you could say, for example, “Hey, this keyword like site cut shows four different SERP features, and three of those SERP features fall into the learn user intent. So there is a 75% chance that this fragmented user intent is actually geared towards a learned user intent. That is the whole kind of idea. And I tested that with a model a couple of times already and ran it through and it comes out pretty well. There is a small fraction of keywords or queries that have such a fragmented user intent that it’s not clear what Google wants in this case, but what it was also able to observe is that that usually changes over time and trends towards a specific user intent.
Ben: My takeaway here is that there’s a notion of keyword maturity for Google, and that there are some times when Google has search results and they just don’t really know what the user intent is, so they’re going to put everything on the page and they’re going to look at the user behavior and understand what is clicked on and that’s going to help them modify the page over time. So when you are dealing with a situation where you want to rank for a term that has a very fragmented experience on Google, how do you think about optimizing for that? Do you put your maps and your images and your various schemas and paragraphs and everything that you could possibly put into a blog post, just hoping something will show up? Are you trying to tackle everything or are you just focusing on what’s right for your user experience?
Kevin: Yeah. When you have a super fragmented user intent, which again comes up sometimes when the … As I mentioned, the keyword is not mature yet, or better said, Google’s understanding of the search intent for a query is not mature, yes. And in this case, you would, as you mentioned, adapt your content to what makes the most sense for your site. So if you’re a blog, it would need a bit more editorial side. If you’re a marketplace, a bit more on the lists and so on. And then you would want to pay very close attention to how the user intent changes over time. So when you would create such a spreadsheet with an overview of the fragmented user intent for all of your most important queries, what you want to do is you want to update that on an ongoing basis. I currently do this on a monthly basis.
Kevin: I think you could go down to weekly with certain rank trackers or certain tools, but at least on a monthly basis to understand how user intent changes over time. And if I was able to observe that in some cases, it changes quite a lot. You can even across keyword sets or for a single keyword, you can see how Google plays with these SERP features based on it’s progressing understanding of the implicit meaning or better said, of the searcher intent. And it’s actually very exciting because if you realize that saber set up a thousand keywords and you realize that all of a sudden 30% of those keywords are not a good fit for the user intent you can’t satisfy anymore, then that’s probably a market that’s breaking away for your business.
Ben: So my takeaway here is that you really have to stay true to your business, right? When you’re working in a fragmented experience, you really have to understand what is going to be true to the end experience that’s going to end up driving the most conversions, right? And just use the content that you already have to try to optimize, to reach as many people that are relevant to your audience.
Kevin: You can create more content. So that’s another learning that, or a lesson, that I’ve found from working on this model or developing this model. It does make a lot of sense to see if images show up or videos all of a sudden. And should that be a decision driver for me to invest more into videos? Tutorials are a perfect segue into that where it might make sense to have written content, but I also see Google showing more and more videos, we can dive into the reasons for that, for a lot more queries because a video is a much better format to learn something than a written text. I would say in 99% of the cases. There might be exceptions, right? But that’s one of the points that I want to make is think critically about the format that your content comes in and adapt if necessary or if there’s a benefit to it, right?
Kevin: Sometimes you might rank pretty well with your written content, but then all of a sudden Google shows a carousel of videos and that should make you think about like, “Should I maybe create a video out of my written content? Should I repurpose a lot of my content into different formats?” And I see the same thing with podcasts right now, there are more and more podcasts showing up for all sorts of queries. And as Google is already pretty good in even understanding the different parts of a video and which part is the best answer to your question, I expect them to do the same thing with podcast content as well. So that means that we’ll have a lot more rich media content ranking for different keywords and we as webmasters have to think about how to create content for that. And this fragmented user intent model is a great way to make that visible to whoever gives the resources to executives, to customers, or even to drive your own strategic decision.
Ben: Makes me think that we should keyword stuff this podcast and talk about things that we think are relevant to Google. Google is number one, YouTube is great. What other products do they have? I love my Pixel. I have an iPhone. Anyway.
Kevin: Not the approach here.
Ben: Serge is brilliant and good looking.
Kevin: A true estio at heart.
Ben: Yeah. As you think about other than podcasts, keyword stuffing, when you’re looking at fragmented user intent, this is not just something that is important because it’s what Google shows on the page. It means that the consumers at large don’t have a clear understanding of what they’re looking for in a search term. What’s the order of operations? What are the things that you should do to make sure that when you’re running into a situation with fragmented user intent, that you’re still using your time valuable and putting your best foot forward, not only the search engine, but to your consumers as well?
Kevin: Sure. So first of all, you want to monitor how SERP features change over time for the keywords that are relevant to you. That’s the first step and it’s so important because it gives you an idea of whether you’re on the right track or not and if your traffic might diminish over time. There’s one thing that we haven’t mentioned in depth yet is that some of these SERP features will take away some of your traffic. On data and you rank the features at snippet, that’s great. But if you don’t, then you’ll see way less traffic. So different SERP features have different implications on the overall traffic that most organic results get. So the first step is to monitor that closely. And again, you can use a tool like Searchmetrics or whatever rank tracker you prefer, because most of them by now show the different SERP features.
Kevin: Number two is you want to regularly export your keywords with SERP features and then classify the SERP features into the user intent that you think fits best or you can use my model. I’m publishing this blog article pretty soon with how I use it and how I understood it. And then the third case is you want to make sure that you repeat that every month at least so that you understand how the search intent for the same keywords changes over time. And then as a fourth step, I would say, you want to monitor your traffic closely and see if you get more or less traffic, depending on whether Google shows more or fewer SERP features for your keywords.
Ben: I think that the concept of search maturity and keeping an eye on what the keywords that you’re trying to rank for and what the SERP features are is something that really tells you not only what Google thinks the intent, but really what your consumers are looking for so you can bring the right search experience to them and deliver as much value as possible. And that wraps up this episode of the Voices of Search podcast. Thanks again to Kevin Indig for joining us today. If you’d like to get in touch with Kevin, you can find a link to his LinkedIn profile in our show notes. You can contact him on Twitter. His handle is Kevin_Indig. K-E-V-I-N_I-N-D-I-G. Or you can visit his personal website, which is kevin-indig.com. Just one more link in our show notes I’d like to tell you about. If you didn’t have a chance to take notes while you were listening to this podcast, head over to voicesofsearch.com where we have summaries of all of our episodes and contact information for our guests.
Ben: You can also send us your topic suggestions or your SEO questions. You can even apply to be a guest speaker on the Voices of Search podcast. Of course, you could always reach out on social media. Our handle is Voices of Search on Twitter and my personal handle is BenJShap, B-E-N-J-S-H-A-P. And if you haven’t subscribed yet, and you want a daily stream of SEO and content marketing insights in your podcast feed, we’re going to publish an episode every day during the work week. So hit the subscribe button in your podcast app, and we’ll be back in your feed tomorrow morning. All right. That’s it for today. But until next time, remember the answers are always in the data.