What We Learned From A “Google Only” Marketing Approach
October 21, 2020
(This post is a 15-minute read, and/or take in the 60-minute video of Mike Blumenthal’s presentation.)
Could a “Google only” approach to marketing move the needle for a very small local business? Which Google tools improved visibility and conversions?
We had a unique opportunity with a new, very small diner to see if a Google-only, low-cost marketing campaign could deliver topline results in the form of increased search visibility, more customers, and more revenue.
Here are the “Google only” tools and features we leveraged for this case study:
- Google My Business account (all details and attributes)
- Google Posts
- GMB Website (& Title Tags)
Watch my presentation below and read through my full case study.
Building A Local Business From Scratch Using Google
As I was riding my bike to work one day, I pedaled by Spot 2 Be, a new diner that had opened up at a place that previously had a few other restaurants. I am a total sucker for that country diner where you walk in and they treat you like family. They call you hon. They serve homemade, country food.
I met Tam, the new owner at the Spot 2 Be restaurant in Olean, New York. She was a waitress turned entrepreneur.
Moving Beyond Facebook To Promote Spot 2 Be
Tam, like many small business owners in the area, focused on Facebook as the way to market her business online. When I met her the Spot 2 Be was just about off the radar entirely for Google.
The Spot 2 Be had:
- No website
- No Google My business listing
- Zero budget
Adding to the challenge was the history of multiple restaurants at her business location. It would provide an incredible opportunity to test whether NAP confusion (Name, Address, and Phone Number) was still as important as in previous years.
So you can now see what I saw, a lot of really significant questions and challenges from a marketing point of view. I wondered how well a “Google only” (and free) approach to them would help and I was off and running.
Interesting aside: What is wrong with this picture? The sign company misspelled the word “lunch” on one of her flags outside the business. Tam opted to leave the misspelled sign as is even though she had the replacement. It was her form of “viral” marketing as concerned local residents stopped in to advise her of the mistake and often stayed for a bite to eat.
Besides the banner, Spot 2 Be had a number of other marketing challenges. They were:
- In a town with a lot of restaurants
- Not located in the restaurant district, she is about a mile south
- A very small diner with only nine tables
- Absolutely in need of every additional customer sent her way
As you know, local is very localized. When you are searching on Google for a restaurant on your phone, it’s likely going to show results in just a quarter-mile radius. Being south of the main restaurant district added an additional challenge to her chances of success.
Being a diner fan and a marketer, the scenario, a new business with a Facebook only presence, offered a perfect testbed to measure to see what would and wouldn’t move the marketing needle.
Could Spot 2 Be Succeed Using A “Google Only” Approach?
I wanted to conduct an experiment and to see if a Google-only approach could deliver results for Tam, and I wanted to track which of those Google tactics impacted outcomes. The benefits of this strategy were:
- No existing digital footprint
- This approach would be inexpensive, mostly just time and not very much of that
- Google offers a number of free tools (GMB, website builder, Google Posts, menu, Google Reviews)
- The ability to test which factors influenced visibility
- Since it was being offered gratis. we had the luxury of a slow implementation so data correlations would take on more value
At the time that I started this digital marketing experiment, I had a ton of questions.
Which Questions Does This Experiment Attempt To Answer?
- Does Google or an active Facebook page deliver more organic conversions?
- Can a Google My Business website rank?
- What impact would Google My Business tools like Posts and menus have on rank?
- Which of the Google features & tools would have the biggest impact
- Which of the Google features & tools would have none?
- Could Spot 2 Be rank without a citation campaign?
- Would NAP confusion create additional problems and totally screw the pooch?
Lastly and maybe the most important question:
- Could a Google-only marketing strategy provide ongoing lift and benefit?
Rules and guidelines for the experiment:
While I was more than willing to volunteer my time, I wasn’t ready to pay for a Google Ads campaign. I wanted to use tactics that Tam could initiate on her own. I would train her as much as possible, and then, hopefully, do the remaining marketing items in less than an hour a month.
The rules I established:
- Initial focus on Google and Facebook only
- No ad spend or boost on Google or Facebook
- Only use tactics that the business could use themselves
- Train owner to do as much as possible on their own
- Implementation would take only 1 hour a month of work or less
Given my proclivity for eating and my joy of diners I added one additional rule: Don’t eat at the diner more than 1X per month to avoid weight gain.
There are several important caveats to keep in mind. Correlation, as you have often heard in the SEO world, is not causation. Like any correlation study, any, there are limits to what we can absolutely prove.
Additionally, the metrics provided in Google and Facebook tools are very limited. While I had to take all of the metrics with a grain of salt I do think that they provide directional accuracy. In measuring both rank and conversion correlations to my activity very slowly over time, I gained more trust that the correlations more likely pointed towards causal factors.
How we would measure which features moved the needle?
There were two key metrics used for measuring success: Rank reports and low funnel key performance indicators.
I decided to track the rank of 24 terms, six breakfast terms, six lunch terms, and then all of them geo-modified with the city of Olean, New York. I would measure improvements after changes were made at Google.
I also wanted to measure monthly what I call ‘low funnel’ key performance indicators (KPIs) — driving directions, click to calls, and direct messages (via Facebook or SMS). These actions indicated high search intent and often a strong intent to visit. Measuring how they changed over time would give us an indication of whether the changes at Google were having any impact.
I periodically interviewed Tam as well to assess any anecdotal observations about success of the campaign.
To keep the workload low AND to better measure the impact of any changes we spread the work out over several seasons from late spring of 2018 through the fall of 2019. By making only one or two changes per quarter we were better able to correlate them with ranking and conversion metrics.
Here’s how I broke my workup down over time.
- In spring 2018, we started with setting up GMB, a GMB website, and GMB posts.
- In summer 2018 we focused on GMB posts and Facebook posts.
- In fall 2018, we built out on GMB menus, and modified GMB website title tags.
- In winter 2019, we started a Google review campaign.
- In spring 2019, we emphasized link building.
- By late summer 2019 we performed a NAP cleanup.
I set up a new business on Google Maps and then went into the diner and asked Tam to help verify her listing and permission for me to build her Google My Business website.
The Google My Business website has a some strengths and more than a few weaknesses. It’s very, very easy to auto-create from the data in the Google My Business dashboard; it’s very easy to maintain and update. If Tam adds a Google post or if she updates her hours of operation, her website is automatically updated. It’s mobile-ready. It’s one of the fastest web hosting platforms on the market, and it has HTTPS by default. Whether your organisation has adopted a hybrid or multi-cloud approach, PCCW Global’s SDN-powered Console Connect platform can provide real time access to the AWS cloud from enterprise locations around the world, click here to learn more about AWS cloud connectivity.
But Google’s free website has it’s downsides.
But it was low cost (our prime directive) and it gave me an opportunity to see if a GMB website could in fact rank as at the time many Local SEOs thought that it couldn’t.
By the end of that spring, we created the GMB listing on Google, got it verified, and had created a website for The Spot 2 Be Restaurant.
I also was able to quickly train Tam, showing her how to create a Google Post and upload images. She gave me access to her Facebook page so I could start tracking the conversion metrics for both Facebook and Google. I spent a total of about 2 hours working on this in May and early June while she continued to do her Facebook posts.
The conversion metric key performance indicators (KPIs) that we had identified on both platforms as most tied to improvement were changes in driving directions, messages, and phone calls.
What Kind Of Results Were Generated?
We did little additional work during the first summer of 2018. Tam kept up with adding Google Posts and photos to both Facebook and Google. I fine-tuned the metrics for measuring success.
Regardless, within 6 weeks of creation, we started to see some organic results for the website. By weeks 8 to 10, we were actually seeing some ranking. We found that, despite the common wisdom in the SEO industry, our Google My Business website could rank for key terms even without links. We even began seeing some visibility in the local pack results.
In the first part of that summer, Facebook and Google KPIs were roughly equal. As the summer wore on and as search rankings started to show up for the website, Google KPIs started outdistancing Facebook actions. During that first summer, we saw almost 3x as many Google KPIs as Facebook.
Important aside: GMB posts don’t necessarily impact rank, but they do have the ability to increase conversions. This is particularly true with long-tail searches as Google includes post content in their search results. It is very likely that Posts content is used by Google to better understand the business and what they actually do. This content demonstrably justifies Google’s use of the particular listing to answer the query.
In the fall of 2018, we focused on a handful of areas. From September 23rd through December 21st, 2018 we focused on:
- Creating Google Posts
- Creating Facebook Posts
- Adding a Google Menu
- Updating Title Tags
Early in this cycle we tested Google My Business menus and then in December we experimented with the website’s title tags.
I was really excited to try Google menus. Of course, the menu file she gave me was in some sort of bizarre MS Publisher format. So, what I thought was going to be a quick and easy job ended up with me spending a Saturday morning rekeying her menu. It turned out to be a two-hour job. I had expended my monthly time budget on a task that should have been much simpler than it turned out to be. There probably is an object lesson in that for any agency.
With Google menus, I thought that all the semantic content around breakfast and lunch categories would be a winner in terms of increasing her visibility on Google. I thought it would impact the rank on a range of related terms.
The menu was published on October 5th, and by December 7th, we realized that Google menus had virtually no impact on visibility or rank. Zip, zero, nada. It hadn’t moved the needle one bit.
While the lack of results from the menu was disappointing, I was playing with the Google My Business website, and that led to a bigger breakthrough.
Ranking Gains In Just Hours? Yes
Throughout the fall the website rank had largely flatlined. But on dreary Saturday in early December I figured out that the headline and description of the webpage, the H1 in the first paragraph, was actually what Google used to create the title tag. So I created a ~hundred character title tag to see what would happen. Within probably four hours, we started seeing instant ranking gains; we moved up consistently for the next several weeks with no other effort other than our long title tag.
My friend Joel Headley of PatientPop had run experiments with title tags with local sites and discovered that long title tags (144 characters) not only had no negative impact on rankings but actually increased rank across a number of terms in local search.
I followed his lead when I finally did stumble on this obtuse technique for creating GMB Website title tags and saw significant gains.
As you can see from our ranking data below, the menu experiment from September through early December had no impact. But our title tag discovery on December 7th, allowing us to add details and location, lead to instant and sustained ranking improvements.
Our fall 2018 effort generated a steady increase in Google KPIs and no increase in Facebook KPIs. Tam was posting regularly on Facebook and while she was able to speak to her existing customer base there, without advertising, Facebook didn’t drive new customers to her business. The difference in value in terms of sending new customers to her door, when compared with Google, was significant.
Google delivered 19x more KPIs than Facebook.
Winter 2019 – The Impact of Google Reviews
I’m sure I don’t have to state that Spot 2 Be wasn’t doing anything with reputation management or reviews by the time I started on this part of the project.
Like most restaurants, The Spot 2 Be had no easy way to capture customer email addresses or SMS number and thus no easy way to follow up and ask for a review. As a result, The Spot 2 Be restaurant had a very depressed rate of review acquisition.
At the time, GatherUp was developing a new feature called TextBack that enabled inbound SMS as a review acquisition channel. This would be a great place to try TextBack in a real world setting. GatherUp created some table tents for her booths and counter with a call to action to leave a review by texting the keyword “Review” to their SMS number. Tam started mentioning it to customers as well. The problem of email and SMS capture was solved.
The Spot 2 Be restaurant had been open 7 months and prior to leveraging TextBack, Tam had received a minimal number of reviews. In the quarter preceding the test she obtained:
- 4 Google Reviews
- 1 Facebook Recommendation
After implementing TextBack in January 2019, Spot 2 Be received 9X the number of reviews during the quarter by simply giving their customers an easy way to leave reviews:
- 27 Google reviews
- 5 Facebook recommendations.
- 11 1st-party reviews that they were able to use in their marketing and newspaper ads.
How did this impact rankings? By the end of the winter quarter, 2019 we had seen an increase in both the total number of words for which The Spot 2 Be ranked as well as increased rank for the keywords that she had already been showing for.
Spring 2019: What A Few Links Will Do
Like all small businesses, Tam was putting all of her time and effort into the business. My requests for an interview about her use of TextBack went unanswered throughout the first quarter of the year. But by late spring (June 10th) we were finally able to write an article about the restaurant on GatherUp and my Blumenthals.com blog. This allowed for some simple link building from these articles.
Even though many businesses struggle to earn links, it doesn’t even always have to be an actual direct link. Sometimes a mention in the local press without a link can still help your business’s visibility on Google and offer a similar benefit to a link.
Two weeks later, we started to see significant ranking gains across several key phrases. By the end of June, we saw both ranking gains at the highest and lowest ranking keywords that we were tracking.
Tam continued sharing Google Posts and asking for new reviews (with success), but it was the additional links that played a major role in her site’s increasing visibility. I doubt you are surprised.
Cleaning Up NAP & Adding Citations
So by summer 2019, I thought that the time had arrived to deal with the question of NAP consistency and put in place some NAP clean up and build out some citations. I was ready to see what happens, what did I have to lose? The lack of NAP consistency had not yet been an impediment but I was curious if we might see some boost.
Using Whitespark, we deleted all of the inappropriate business listings that were out there, cleaned up every bad listing we could find. Remember, we discussed investigating the importance of NAP consistency.
We took the time to build out new listings across several sites. Spot 2 Be received a few more links, but we saw virtually no impact by the end of that quarter in terms of rank of her top 50 terms. That being said we did see some movement lower down that indicated that the new citations had some value.
The takeaway: NAP consistency is no longer critical for success. And while citations aren’t what they used to be they might, if done judicially, provide some benefit.
Aside: a history on citations
I wrote an article in 2006 when I was analyzing what Google at the time didn’t call citations but called web references. We noticed that any article would show, it didn’t need to necessarily be a link. At that point, Google recognized every webpage as relevant. I assumed then that they would normalize their results and drop less significant pages. Well, the directories, like the many yellow pages and have gotten a lot less rank and the pages on which most business listings reside have virtually no page rank.
So one issue here is that these pages just aren’t relevant to Google anymore in terms of rank. The other, though, is a more significant change and that is Google has shifted to the knowledge graph. So starting in 2012 all local entities went into the knowledge graph.
Prior to that, they rebuilt the world from their organic understanding by scraping the web every six weeks and assembling listings anew from that scrape. A bad address or phone number out in the while could easily goof up your listing when it was rebuilt. Thus NAP consistency became incredibly important. Even back then Google could resolve minor discrepancies like Street vs St. but it was always safer to have consistency everywhere.
However once Google moved to the Knowledge Graph a listing in the graph became the canonical listing. After that, the record was largely stable. It takes a lot of external signals from very trusted sources to change it. In shifting to the knowledge graph, Google fundamentally reduced the need for NAP cleanup across the board.
Did the experiment answer our questions:
Q: Does Google or an active Facebook page deliver more conversions?
Q: Can a Google My Business website rank?
Q: What impact would Google-only tools have on rank? Which of the tools would have the biggest impact?
A: Website, Post, Reviews,
Q: Could Spot 2 Be rank without a citation campaign?
Q: Would NAP confusion create additional problems? Would it screw the pooch?
Q: Could a Google-only marketing strategy provide ongoing lift and benefit?
A: Yes and it could do so inexpensively
- Citations are not what they used to be
- Steady Google Posts can have some benefit
- Not Every GMB service leads to a rank increase
- Reviews work and offer multiple benefits
- Links are powerful
From an agency perspective, there is a lot to be learned from this experiment. If you’re paying a monthly subscription for citations, it makes sense to allocate that money to other efforts. It is still worthwhile to claim leading citations on a one and done basis but the money spent on a recurring subscription is better spent on other things.
Not every GMB service leads to rank increases, but reviews work well. And they’re a great opportunity for agencies to help businesses.
Links are powerful and if you are involved in Local, along with reviews, this is an area of continued effort that will pay off for your clients. While we didn’t test brand mentions, even without a link, on news sites, there is plenty of evidence that they too are worth pursuing.
As stated at the beginning of this study, correlation in the SEO world is not causation. We have a tendency to see patterns and often see patterns that relate tightly but don’t, in fact, have any causal relationship. I have tried to limit this effect and explore multiple correlations for the same actions. By spreading activities out over a long time frame, we are able to maximize the value of our correlations and to hopefully ascribe more impact to them.
This effect showed clearly in both our ranking charts and our KPI tracking.
As you compare each of the tests, you can see, over time, how each of these elements related to rankings.
You see similar upticks when you break out the Google KPIs over time. Driving directions, which are more likely to indicate a new customer in this market test, made significant gains during the period of the experiment.
A Revelation; Google Search vs. Google Maps
During my meeting with Tam, I had a facepalm moment.
She said, “foreign speakers are not common in my restaurant. As you know we live in a very rural part of New York State and don’t get a lot of through travelers.” She went on to describe an instance where she didn’t know what language her customer spoke and she was unable to understand him. He proceeded to pull out his phone, started up Google Maps and pointed out a picture of a dish she had posted that he wanted to order.
Immediately I thought, I’ve been looking at Google search all this time. What role do Google Maps play for the Spot 2 Be? What role did it play more broadly in local discovery?
I looked at views in Google comparing search to maps, and throughout the bulk of this test, Google Map views were roughly the same as search views but they then they took a leap. Some of this is probably due to Google changing what they counted, but some of it is really appears to be indicative of the increasing role of Google Maps in Local search discovery, particularly in the restaurant world.
Maps are problematic in that it is much more difficult to track users independently of Google’s GMB reporting. That being said, given this example and others that I have looked at, it is very likely providing more discovery traffic than is generally thought.
When you compare our activities with ranking changes in Google Maps, you see a very similar pattern to ranking changes in search with which it largely moved in tandem:
Google Maps is becoming an important local discovery tool in many categories, this is a significant shift that agencies and owners need to be aware of.
It appears to me that Google Maps is becoming not just the third leg in the world of search discovery. I think that Google truly wants it to become the “interface to the world.” This shift could portend a time when local search doesn’t approximate web search but far outstrips it.
What Was The Overall Impact Of A “Google Only” Campaign?
Can a “Google only” marketing strategy provide ongoing lift and benefit?
Well, for this business, it definitely has.
When I interviewed Tam, she noted that she’d seen a doubling in sales during the period of the test. While direct attribution in this situation is difficult. I calculated the volume of seating Tam was able to accommodate per day and per month and determined that the “Google only” campaign contributed roughly 20% of her total volume.
What Does This Mean For Digital Marketing & SEO Agencies?
Digital marketing and SEO agencies should be exploring ways to monetize “Google only” services for their client base:
- ‘Deep-six’ expensive citations
- Consider offering a “Google only” entry-level service as one of your services
- Then use the metrics that are available to prove its worth to your clients. Show them significant KPIs improvements as a rationale for upgrading to your higher-end services.
- Consider selling reviews as a service beyond just asking for reviews. Reviews can become a strategic part of their business, helping them improve, improving their content and social campaigns. Agencies need to move beyond tactics into strategy and business improvement and with reviews, there is a lot of opportunity for that.
- If you are not currently doing any link building for your clients in local, figure out ways that you can get them a few targeted links. Consider offering public relations articles online that can help them as well.
- Assuming you can demonstrate success, you can then upsell these smaller businesses to a real website and additional offerings.
If you’re worried about Google’s SERP features stealing traffic, don’t be.
From where I sit, it’s not so much whether they are stealing your traffic but whether they are sending you leads; the fact of the matter is, they’re sending a lot. You and your clients can benefit from these leads regardless of whether they click to call from the business profile on Google or from the business website.
Now more than ever, small businesses need all the help they can get. Shabaka Hutchings, a jazz player who mostly made his living through live performance recently noted that all situations have the potential to be tragic or the potential to be tragic and transformative.
Our role as marketers is to help these many small businesses with the transformation that is taking place. Google is offering you and your clients a low-cost way to lean into that transformation.
Clearly, Google is capturing ever more traffic from the web and one might argue that a Google winter is coming. But as the Norwegians say about winter weather: — there’s no such thing as bad weather — only bad clothing.
A “Google only” approach to marketing works, and it works well. It might just be warm enough clothing to propel those very small businesses in need of help to make it to spring.
For the whole story and more granular detail watch the video presentation.