In Search [Episode 32]: On the Health of Local Search
Don't forget, you can keep up with the In Search SEO Podcast by subscribing on iTunes or by following the podcast on SoundCloud!
June 25, 2019 |
The In Search SEO Podcast Community Question of the Week!
Help the SEO community! Share your top tips when competing with spammy listings in the Local Pack or when dealing with fake reviews!
Summary of Episode 32: The In Search SEO Podcast
This week we have none other than the grandmaster of local SEO, Professor Maps himself, Mike Blumenthal
to share his local SEO wisdom with us:
- Do local marketers need to hop on the paid search bandwagon?
- What is the real story with Google My Business monetization?
- The role of reviews in local ranking
Plus, we take a look at how Google is further developing its Topic Layer.
Google Adds More Layers of Subcategorization to the SERP [02:24 - 17:53]
If you recall, a few episodes ago, Mordy pointed out that Google was using the mobile SERP renovation
as the first step to ratcheting it up the number of changes it would making to the SERP.
If you remember we talked about all of the Google bugs (which are still popping up each week) and we talked about how Mordy thought that these bugs would indicate something big in the works. We also talked about how the redesign of the mobile SERP (you know, the new ad label that has no color and will smaller and colorless URLs with favicons, etc.) was an external representation of this makeover. An external makeover to symbolize an inner makeover.
Mordy actually predicted in the latest version of the SERP News
that we would see Google pick up the pace with the number of changes it makes to the SERP. If you have been following these things you’ll know that 2019 has seen far fewer changes than in previous years.
Mordy thinks there’s a pattern that can be seen here. A pattern of the changes being made and a pattern of when they are coming. If you remember, Google updated the look and gave a slight makeover to the desktop SERP quite recently. At the top of the page, you’ll see the headers/buttons on the menu now have both an icon and a name. For example, for images, it says "Images” and has an icon that represents images.
What happened after the aesthetic changes to the mobile and desktop SERPs was an opening of the floodgates. All of a sudden there were tons of changes to the SERP. Tests to all sorts of SERP features.
But that’s not the whole story because it wasn’t just the number of changes… but what has changed.
What’s changed is that there is an additional layer of depth that has been added on to many of the features you see on the SERP.
Let us show you.
If this were 2018 and you Googled store hours at Macy’s Herald Square you would get an Answer Box that showed the store hours for that store. Now, in 2019, not only do you get the hours, but you also get the business’ Google rating
Here’s another case. When you search for a general product you might see a refinement box. For example, if you do a search for bunk beds you might see dropdown tabs that will refine your search. In Mordy’s case, he saw a tab that allowed him to "refine by size.” Expanding the tab gave him a carousel of bed size options: twin, queen, king, etc. Another expandable tab Mordy saw for the query was "by style” that allowed him to get results on a specific type of bunk bed.
In other words, an extra layer of topical depth/broadness has been added to product queries.
Same thing with the store hours search. We have the business, its hours, and now we have another layer, its reviews.
Mordy thinks Google has leaped forward with its topic layer with its ability to add layer upon layer of topical breakdown goodness, of sub-categorization to an entity, to a topic, to a product, etc.
But that’s not all. If you did a search for bed sheets, it used to be you would get a carousel of bed sheets. Now Google is testing another tab where not only can you look at individual bed sheets but you can get a carousel of comparison sites
. Mordy thinks this has to do with Google wanting to avoid the critique that comes with just showing its own products. This does get Google out of that issue a little bit but it also reflects a better understanding and breakdown of the "topic” or "product” into additional layers.
Here’s one last one for you. You may know that Google Posts are shown in order of post date and only lasts for 7 days (generally). Well, now Google is at times showing a carousel of old posts from the business that relate to the user’s query. Back to the Macy’s example, if you searched for Macy’s perfume you might see old posts related to perfume in the Knowledge Panel. Again, another subcategorization, another parsing of a topic.
There were a bunch of other tests that you could tie into this. For the sake of time, we’ll leave them out for now, but in a few weeks, we will release the next edition of the SERP News
, so full coverage there!
Again, all of these changes point to a pattern where external overhauls or changes to the SERP signify larger changes often enough. That larger change, or part of it at least, is an extra layer of topical depth on the SERP. With a more prolific use of the Topic Layer comes a variety of implications from increased product competition to a better contextual understanding of a business and so forth.
On the Health of Local Search: A Conversation with Mike Blumenthal [17:53 - 53:51]
[This is a general summary of the interview and not a word for word transcript. You can listen to the podcast for the full interview.]
Today we have the great sage of Local SEO, the co-founder of Local University
, the co-founder of online review engine Gather Up
, and the provider of local insights on the Understanding Google My Business & Local Search
. He is, of course, Professor Maps himself, Mike Blumenthal!
So am I wrong or are you a big outdoorsman?
I used to be a big outdoorsman. I was a guide for The National Outdoor Leadership School for a number of years in East Africa, Alaska, and Wyoming in the 70s and 80s. Since then I slowed down on that front. I do ride my bike a lot.
Let’s start today with Local SEO: spam, reviews, and all sorts of fun things.
A lifetime ago, I used to work for a property management company in NYC for about 8-9 years so I’m familiar with the big NYC plumbing companies, the most reputable locksmiths, etc. However, when I do a search for ‘plumber nyc’ I don’t see any of the people I would expect to see. I get a whole bunch of results for "24-hour plumber” this and "24-hour plumber” that. Is the algorithm busted? Oddly, the Local Service Ads are more aligned to what I would expect to see.
Well, for one, your search is broken. Most searchers tend to search on mobile, search on businesses nearby, and typically don’t use geo-modifiers in their search. If you do use geo-modifiers, and you’re within the proximity, Google does a better job. Sure, there is a bit of spam, especially in plumbing, where Google, over the past year or two, hasn’t done a great job of eliminating. So there is brokenness in Google handling local listings.
It’s a little bit of both. Google is interested in giving relevant search results based on location. But in service area businesses (SABs), there is a lot of spam and cheating as the economic rewards are so high. I think one of the ways Google responded to that is their Local Service Ads that look like pack results and need a higher level of verification to get in. They have been showing more of these in the SERP.
I wrote an article just the other day of Google testing a horizontal scrolling Local Pack
. When you looked above it the top of the screen had two Local Service ads and below were three Adwords ads, then an organic result, and then the Local Pack. So one way Google is dealing with this is by pushing the Local Pack and providing other results above it in that vertical and in that market.
For plumbers, it’s a very localized search. There are industries where people would want to look more broadly like, for example, car dealerships in New York City. There aren’t many and Google will need to expand its radius to answer your query.
So with the Local Pack test you saw. For Google, is that an admission of failure?
I had a couple of theories about it. One is a response to the actions the government plans to take that Yelp has been calling for the past 4-5 years. Certainly, Local Service Ads is an admission that SABs tend to be spammy. There have been doing a lot of work over the past six months asking SABs to reverify, and when they do they need to submit real signage of the location and to prove that they exist. It is a slow process as it’s a combination of physical verification plus AI and machine learning.
I’m not looking at culpability on Google. This is what you deal with today at every vertical and this is the decision a business has to make vis-a-vis these results, regardless of culpability and guilt. Google has succeeded in winning market share and succeeded in putting most of the other local sites out of business and this is your choice as a business or a consumer. To some extent, the reality doesn’t serve the business. Yes, it may suck, but in the meantime, you need to make a living so how are you going to do it?
You spoke about Local Service ads. If you’re a legitimate business and because of all the spam do you feel you need to go the root of paying for a Local Service ad?
I think it depends on the market, the vertical, and the location. We’ve seen AdWords in mobile creep up from one or two percent of the clicks to six or seven percent and maybe higher in this marketplace. And yet organic still has its part in that world. So even if ads took 20% of the clicks it’s still important to work on your organic local listings as 80% of those clicks are going someplace.
So the answer is yes. I would take a mixed approach and I would evaluate each. I would put in tracking to make sure the expenditures in one place vs the other is paying off. I see a mixed approach to search in three things: AdWords, entity optimization/knowledge graph optimization, and organic optimization. All of these provide value to a business. It’s just the issue of the mix and the return on investment.
I want to jump to the topic of Google My Business optimization. For those not familiar, Google put out a survey that hinted to the search engine monetizing Google My Business
at some point down the road. One thing that came up in that survey is that Google may charge to verify a review. To see Google saying they can do this but will charge for it, does that make your blood boil a bit?
So I saw that and I immediately discounted it as it sounded like something written by a summer intern. The fact it had so many questions and they could’ve used their own survey tool made it seem unlikely that it reflected a higher level of thinking.
Issue two is that the reality is that Google is a corporation who is beholden to their stockholders to increase income. The other reality is that in local Google had a long practice of offering a freemium model where they offered many things for free and cheap which they’re not doing out of the goodness of their heart. They want the user data which they see as gold.
I don’t need a survey to figure out what Google plans to do. One reality is that if an expensive investment won’t generate $1 billion of income than they won’t do it. They need big money to change their income projections. Secondly, they see value in data and will continue to do things that will increase their data. For example, reviews. Google is now using review content to answer consumer’s questions about the business. That’s an incredible use of artificial intelligence and machine learning to solve a user’s problem and keep them on Google without having to go someplace else.
In that context, over the last four years, we have seen Google unleash the power of their local search engine to garner reviews. So when I do an analysis of a restaurant chain with over 200 locations I’ll look across their last four years of reviews. So on Yelp, they’re getting the same number of reviews per location as they did four years ago, maybe one review per location. Whereas on Google, they zoomed past Yelp where they’re generating 10, 20, or even 30 times the number of reviews per location on a monthly basis. Google sees huge value in that data.
There was a patent released in 2017 which indicates that Google is parsing these reviews to understand both entity details: what they do as well as how well they do it. They use reviews from a data point of view so they need massive quantities of data to effectively deliver on that promise of answering the consumer’s questions about the business.
So why not monetize Google My Business?
They are doing it. If you’ve looked at the history of Google local through 2015, they have announced they would do it through ads. They have increasingly localized their ad tool so that you can by now, for example, promote pins on the map, you have ads in the pack results, and you have Local Service Ads. These are all new formats in the last year or so.
My answer is they’re already doing it with great success and great monetary gain. Do they need penny ways to make money to piss people off? Probably not when they’ve got big ways to get their hands into your pocket.
I was talking to Sergey Alakov a few months ago and we were talking about how local is the natural segue to monetize voice search
. Does increasing the monetization of local segue into solving their voice search problems?
I think Google was ahead of the curve in terms of implementation of the knowledge graph and of understanding businesses characteristics. In a broad sense, the knowledge graph is a better way to answer questions than a link graph. So the question is if a monetized version of the Knowledge Graph, where they’ve done extra betting as with the Local Service Ads, is a better way to guarantee the relevance of those results. I think in some verticals, like plumbers, that Google has a legal and moral obligation to vet these businesses at a higher rate than they have been doing in the past and Local Service Ads are the best way to do it. It makes sense they would deliver those one-or-two answers that voice supports via more knowledge about that business with that knowledge likely coming from a paid ad because it includes serious vetting.
Yeah. I think it was a month or so ago, there was some speculation that Google was using Local Service Ads as a way of providing voice search answers for local queries. From my perspective, that’s great. It’s much better than the average local result.
Right. Google is concerned with delivering a good experience and a good answer and they got to where they are historically by doing that. So if they believe they’ll be giving a better answer in some verticals I think they’ll make that choice.
Let me jump back to the Q&A Feature which we spoke of a few minutes ago. I find this feature to be very undervalued and underutilized. Is there potential for immense amounts of spam? In other words, why can’t one business ask some really nasty questions against its competitor which will show them in a bad light?
Let’s first take a step back and look at the feature broadly. Google introduced it from August to December 2017 as a way to increase their long-tail understanding of a business. If you look at the way it’s shown on mobile it presents the business almost like a website with a tabbed interface. The Google Q&A serves to supplement that information much like an FAQ does on the business website. Google implemented it as a way to create timely answers to very specific questions that consumers might have. And then when a question is asked they would send it to a Local Guide, which they have 50-60 million of, to get a quick answer.
Now, this opens up possibilities on multiple fronts. On a positive side, it gives the business the opportunity to preload its own questions and answers into the Q&A. Answering potential consumer questions will save the consumer the phone call and save them the trouble of looking for the answer.
Initially, when I did my research, the quality of answers I saw was that 75% of them were selling opportunities and 23% were initially reputation related, inappropriate questions that violated terms of service. 11% were reputation related while the others were irrelevant questions.
Since then I have noticed that many of these bogus questions have been taken down by Google and some still get leaked through. The business can report it and it will be taken down so it behooves the business to monitor their business profile because it’s the most viewed information about your business in the world, even more than your website.
So yes, it’s open to abuse as is any crowdsourced platform. As Google being machine learning driven, it tends to look at 90% accuracy as being accurate enough. Which is annoying as it means about 10% is inaccurate, but a business needs to understand that this is how Google views the world. If it’s largely relevant, that’s good enough. As a business owner, I want precision. I want it to be accurate. This pushes the workload back down to the business in order to ensure it is accurate.
So yes, it’s open to abuse, but I think any business paying half attention to this problem would be monitoring their Google profile regularly and doing something about it.
I want to preface the next question by saying that when it comes to YMYL issues on the SERP, particularly medical issues, Google’s Medical Knowledge Panel has taken a lot of care and effort (with help from the Mayo Clinic) to make sure the information there is accurate.
But in dealing with the Q&A feature for a health-related business, Google is dealing with content created by users and the business itself. Does the Q&A feature work for a medical/financial business where there is room for "irresponsible” information to be put out?
I think a bigger tragedy that I have seen is when a consumer in desperate need asks for help for serious depression or suicidal thoughts. It’s conservative thinking that this is going into a real-time answering system when in reality it goes into a tumbler where maybe the answer comes, maybe it doesn’t. Maybe the answer is accurate, or maybe it isn’t. I think Google hasn’t done a good job of educating their consumers about their feature. I’ve seen some examples of tragic consequences and they’re very sad.
This is the problem with doing things at scale with machine learning with user-generated content. We’re seeing these problems now that our society’s allowed us to be that modus operandi.
I think Google failed in two areas: One, they haven’t educated consumers on its features, and two, they haven’t done a good enough job of building reporting so that businesses can immediately be notified about new questions. Recently, in January, they upgraded their API to support Q&A reporting which is good enough for small businesses, but not good enough for large businesses with multiple locations.
This is typical Google. They bring out a feature, they throw it against the wall, they see if they can get 85-90% reasonable answers, they think they can train their AI to keep the bad down, and then they invest in it slowly over time. They build early and reiterate often. In the real world of local, that approach to relevance can be very annoying.
It is interesting as it would seem that Google could’ve locked off this feature from certain business types.
They do, it’s less visible on schools and such. So they could but they don’t. Google views success differently than us.
I want to end off our serious questions with one last thing about reviews. In one Google help document the search engine says, "Google review count and score are factored into local search ranking: more reviews and positive ratings will probably improve a business's local ranking
.” This seems to imply that ranking is impacted just by the number of reviews and the rating of these reviews. But what about the content of the reviews themselves? Do you think Google in some way "indexes” it? The idea that they can verify a review seems to also imply that they can analyze its language. Do you think that the very content of the review will or does come into play in some way, shape, or form?
I think the content is probably perceived by Google as the most significant part of the review. They published a patent in fall 2017. I wrote up an article about the patent
and my understanding of it on GatherUp in February. In the patent they note the following: user reviews may be gathered from one or more of a blog or social network postings, emails, articles written for websites or from printed publications such as magazines or newspapers, or postings made to a user review section of an online vendor or marketplace.
So point one is that they look everywhere to review content. Point two is they noted in this patent that this is about understanding entity attributes. In other words, they are looking for a business’s entity attributes, they are looking for reviews to understand what the business does and how well it does it. They looked at positive mentions of what they did in aggregate as increasing the likelihood of that business to show that attribute.
Google thinks of rank in terms of prominence and relevance. So reviews have some say in prominence but not as much as people think. And relevance, i.e., the content of the review, in terms of expanding the reach of a local business, plays a huge role in reviews. in some ways, I think that that's the most important part of reviews, by using them to understand more about the business.
Here’s an Interesting tie-in between Google reviews, Google Q&A, and Local Guides. Firstly, with local guides, they incentivized local guides to create longer reviews. Google gets that more content helps with understanding. Secondly, there is a new feature in Google Q&A in the US when a consumer starts typing a question Google looks at the word in the query and starts servicing reviews that match that query. They extract those nouns and verbs and highlight in the review the ones they think answer that question.
This demonstrates what the patent is talking about when Google looks at the content of the reviews to understand the attributes of the business. And now they’ve taken the next step of not only understanding the attributes of the business but to answer unpredicted questions about the attributes of the business.
Optimize It or Disavow It
If you had the terrible choice between uploading a pretty bland, boring, and otherwise irrelevant image to your local listing or creating pretty bland, boring, and otherwise irrelevant Google Posts and Q&A questions and responses, which would be the lesser of the two evils?
The way I see it the business should get their bicycles and go on a bike ride because bad content and bad images is a waste of everyone’s time particularly the consumer. And I can see marketers thinking that somehow more images are better even if they’re terrible.
Wow. You’re the first person on our podcast to give such an answer. Which is why you’re so cool.
Thank you so much, Mike, for coming on. I really do appreciate it. It was a lot of fun.
Thank you for having me and I’ll be glad to come back anytime.
SEO News [57:08 - 01:03:31]
Google Stealing Lyrics From Genius?:
A bit of controversy as Google was accused of scraping lyrics from the site genius.com
. Google denies the accusation, but it has taken the SEO world by storm.
This is interesting as Google has said that they pay for the lyrics, but what about other content they display on the SERP? Mordy wondered that if you wrote a recipe and Google scrapes it and puts it on the SERP, shouldn’t Google pay you? If you’re going to pay for one, pay for both!
WSJ Reports on Fake Google Map Listings:
It was a rough week for Google. Days after the "lyric” incident, the Wall Street Journal ran a report highlighting how fake Google Map listings really hurt people
This whole issue of fake Google map listings is old news so why is it being discussed now? According to Mordy, the culture is changing where people are asking if Google should be regulated. Whether you agree or disagree, the reason these articles are coming out is that it fits into the new cultural dynamic of Google being too big for our own good.
3D Ads and YouTube Live Ads:
Google is bringing 3D functionality to Google Ads
. A new ad format lets you place a rotating object in video format within an ad.
Also, Google is releasing a special Google Ad format for YouTube live broadcasts.
New ‘People Also Considered’ Ads:
A new Google Ads test has a carousel of smaller ads entitled ‘People Also Considered’
appearing under one large ad.
The title ‘People Also Considered’, has been driving Mordy a little crazy. It seems indicative that the first advertisement isn’t your best option because people ‘considered’ other ads. It gives the user a feeling that before they make a buyer’s decision they need to click on all the other possible options.
Fun SEO Send-Off Question [01:03:31 - 01:07:54]
If Google had a theme song, what would it be?
Sapir thought that the Requiem for a Dream theme song would be a great fit as Google scares her sometimes and we’re all addicted to it.
As for Mordy, since Google has been under fire with this whole lyrics snafu their theme song should be Hit Me With Your Best Shot by the great Pat Benatar. It’s sort of a message, like c’mon SEO industry… you think you got something on us… hit me with your best shot.
Tune in next Tuesday for a new episode of The In Search SEO Podcast