This case study was created using an LRT Superhero account.
Some of the use cases explained in this case study are not available in lower plans.
The LRT Superhero Plan (and higher) includes all our 25 link data sources and allows you to perform link risk management, competitive research, professional SEO and backlink analysis for your own or your competitor's sites. You get to see your website's full backlink profile picture and this can make all the difference for your SEO success.
We Don’t Need No Stinking Badges
100% Satisfaction Guaranteed. Outsanding Service. Certified Member. Such badges calm us, don’t they? Trust symbols, as they’re sometimes called, help us make decisions without thinking. And that’s the way we like it.
Unfortunately, there’s a dark side to these seemingly innocent images. Sites like Thumbtack tried to incentivize its members to add their badge to their site…in exchange for a follow link back to Thumbtack.
It worked…for a while. But then the all-crushing hammer of Google fell upon Thumbtack, leaving it black and blue and howling in pain. It all but disappeared from SERPs, relinquishing prime positions to its behemoth competitors.
The great mystery here was how a Google-backed startup like Thumbtack evaded detection for so long and was only penalized when an SEO called them out for “buying” links with badges.
As you will learn, there’s more here than you think…and it affects all sites, small or large.
Enjoy & Learn
Christoph C. Cemper
Table of contents
- We Don’t Need No Stinking Badges
- Why Was Thumbtack Not Spotted Sooner
- The Local Listings Market
- Silicon Valley & Venture Capitalist Involvement
- Major Competitors
- Thumbtack Analysis
- But What Of The Competitors?
- Getting The Links In Place
- Super Service Awards
- The Move To Content Marketing
- Conclusion – We Do Need Stinking Badges After All
- Author Bio
There have been several reports regarding the recent Thumbtack manual action.
Whilst this case study will look briefly at the site-specific circumstances which led to Thumbtack’s penalty, the main objective here is to look at why the site wasn’t spotted much sooner and why it took SEOs to notice the link acquisition methods.
The case study looks at the overall local search vertical and some of the main competitors’ link acquisition strategies to get a feel for what is “the niche norm.”
I feel it important to disclose that I do know this vertical well as a website owner.
I have a local listings site (tiny when compared to these market players), which I continue to work on as time permits, and have done so since 2007/8 when I began to work on the project.
It is the very reason that I chose to switch careers and move into SEO.
Whilst working on the project, and as a result of negative experiences with SEO providers, I decided to learn as much as I could myself.
My enthusiasm for search grew to the point that I decided to enter the industry full time.
My listings site is UK-based so it is not directly in competition with any of the US sites which are looked at in this case study. All of such noted information is very visible.
NB: All of the information in the document is widely available and in the public domain.
Furthermore, if it were not for the Thumbtack situation arising and focusing the attention on this vertical it is likely I would never have been involved in a case study of this particular nature, and the issues covered would not have been flagged.
This is a reaction to a currently live situation and hopefully of interest to the SEO community.
The Thumbtack Situation
Here’s a quick recap on the situation so far: Thumbtack is a major Google-backed investment which provides local listing services for small businesses, each of which have profile pages and look to generate business leads.
The organization was exposed for sending out emails offering progressive profile points in return for adding an aggressive money keyword link back to them along the lines of:
1. Copy and paste your profile link to your website with a description of your profile
If you would prefer to add a button or a badge instead of a link, you get one here.
2. Once you are done, respond to this email with the URL of the link, badge, or button.
I’ll verify the link and will credit the points and badge to your profile.
They were called out on social media by SEOs when @DonnyStrompf stumbled across their link acquisition methodology. Debate ensued around whether they were paying for links in their techniques, given that no actual financial payment was being made.
Their argument had always been that points did not mean currency.
Other commentators and (clearly) Google saw this differently because Thumbtack confirmed that they did indeed eventually receive a manual penalty.
So, rewarding / incentivizing people in any way for linking back seems to be the issue, regardless of whether currency or “points” exchanges hands.
Thumbtack subsequently disappeared quite dramatically from Google SERPs, not even ranking for their own name.
Even “drawing pin” on Wikipedia.org was ranking above them on Thursday for a brand search on “thumbtack.” Their Wikipedia organization page seemed to also drop, coming in also below the drawing pin Wikipedia page on Google SERPs for a “thumbtack” search.
However, according to a check on their keywords, they still had some 89,000 keywords ranking in the top 10 (likely long tailed).
So, it’s a pretty heavy blow.
A few days later they were sending a very different type of email to their members. They asked those previously incentivized to add money links to now “nofollow” (or remove) the links, with Thumbtack admitting they had “messed up.”
Barry Schwartz @rustybrick in a SERoundtable post on this showed a copy of an email sent to @NorthEasternExt asking for the link to be nofollowed.
By Saturday, June 13, 2015 (less than a week after receiving their penalty), Thumbtack rankings looked to recover dramatically, indicating the possible lifting of the manual action.
With lifts on location- / service-based terms such as “catering miami,” “computer repair denver,” “nutritionist Houston” moving up by as much as 70+ positions right back to position 1.
Why Was Thumbtack Not Spotted Sooner?
Why did it take an SEO to out Thumbtack if they were carrying out such large scale automated link building that was clearly against Google’s guidelines?
Before we answer that, let’s recap on where they went wrong:
- They blatantly offered profile progress points for keyword-rich anchor text links in emails to site members.
- They likely automated the process of link acquisition (via the emails) using their CRM system.
- They over-egged the exact match money anchors? (or did they? More on that later)
- They were obvious with anchor text linked widget badges such as these (once someone took the time to check manually).
- Are they more visible to the SEO community than others in their vertical? It’s just a thought, but they do have many SEO and web design businesses listed as members on their site. SEOs, by their nature, will question these types of techniques once discovered.
A link velocity check on the Thumbtack domain indicates they’ve employed this tactic aggressively for quite some time.
So, why have they only just been penalized? Surely, keyword-rich anchor text building on this scale would normally have triggered some type of Google filter? Why were they not spotted sooner by any search engine velocity filter systems or “too many money anchor” filters?
Various commentators noted, upon reviewing their non-brand anchors overall. that their links look like spam on a large scale with a vast array of many, many different types of money keywords.
So, again, why no detection from Google’s filters?
How did they get away with it for as long as they did?
IT’S THE NORM FOR THIS VERTICAL
(from an anchor text perspective and sitewide / domain wide backlink profile perspective)
THEY JUST GOT CAUGHT
OVERALL THEY BLENDED IN QUITE NICELY, THANK YOU
They squeaked by because their profile does not stand out from the rest of the vertical: They blended in and did a little bit better.
In this industry, the sheer volume of inbound anchors to these local listing sites is diverse (often with a combination of location, state and service type anchor), and typically these anchors add less than 1% each in anchor text commonality.
That’s why Thumbtack was getting away with it while reaping the benefits via sitewides from an army of thousands of small businesses eager to gain points and grow their own business.
Now, by saying that this is the norm, there’s no implication that all of the competitors are sending automated emails offering points to their site members in exchange for a keyword rich anchor text. There are other ways to reward members in exchange for followed links at scale from tens of thousands of sites with low power and a bit of trust.
Badges and awards, of course
Of course, “badge wearing sites” are more often than not sites belonging to local small businesses who likely have not been overly involved in the “penguin fodder” link building practices of the past, so that’s an added bonus.
The Local Listings Market
To get an idea of what the norm is in this vertical, it’s important to look at the local listings search market overall and who some of the key players are (and their investors).
Silicon Valley & Venture Capitalist Involvement
It’s vital to remember that Thumbtack is a Google backed venture with them contributing up to a $100 million investment round in August 2014.
It’s no surprise that Google chose to invest in a local services listing platform – it was missing from their portfolio.
This Moz article authored by Johann Beishline @LeadersWin in August 2013 entitled, “Watch out, Google Will Dominate The Web For Decades To Come” refers to the various areas of search that Google was looking to gain significant presence in, and local listing sites was missing from their inventory of verticals to dominate.
The potential in this market is huge with annual revenues heading towards the billions if, as Johann states, “you throw Groupon into the local category” with an additional $1.61 billion.
HomeAdvisor is one of over 150 brands owned by IAC (Interactive Corporation), a publicly traded company with annual revenues in excess of $3 billion (as of 2014).
Other brands owned by them include Match.com, Ticketmaster, Ask.com, Investopedia, Price Runner, Tinder, and Vimeo, to name just a handful, along with some country-specific sites such as Trade Advisor (UK) and 123 Devis (France).
Likely a few of those names will be familiar to you – it’s an impressive stable.
The HomeAdvisor domain name
A bit of history: Back in 2000, Microsoft was keen to enter the real estate market with the vision of providing an automation platform for home buyers, amassing over $100 million in investment in order to launch HomeAdvisor Technologies. More background is here via Wall Street Journal reporting.
HomeAdvisor technologies is listed on the Microsoft Investment History page below.
And it’s also on the Wikipedia page showing a historical list of assets owned by Microsoft Corporation.
However, Microsoft’s vision for HomeAdvisor Technologies had changed by 2001, as reported in this Wall Street Journal article.
Microsoft had decided against its plans for the platform with a spokesperson saying “the main thrust of the HomeAdvisor Technologies venture” – building software applications to help mortgage companies automate more parts of the home-buying process – "just didn't match up with our goals and objectives with MSN."
They let the homeadvisor.com domain lie idle, but remain registered since 1998.
In 2011, IAC opted to rebrand their original local home services site “Service Magic” as Home Advisor, buying the domain from Microsoft.
Indeed, according to WhoIs, as of June 13, 2015, the homeadvisor.com domain name was still registered to Microsoft Corp. (Maybe WhoIs simply did not update their database recently).
Historical evidence of the “Home Advisor” history has left a trail in the backlink profile below, with anchors such as “Microsoft is in the real estate business,” “homeadvisor / service magic,” “Microsoft homeadvisor” and “Bill Gates is in the real estate business!”
Service Magic rebrand to Home Advisor
Bill Gates is in the real estate business
Microsoft is in the real estate business
Originally set up by Angie Hicks in 1995, Angie’s List provides a crowdsourced review platform for those looking for home services. It reportedly had 70,000 subscribers in 2013 but has failed to ever have a profitable year. Recent reported figures show revenues of $78,896,000 for the quarter ended June 30, 2014, with a net loss of $18,223,000, according to their Wikipedia listing.
We are looking at some extremely big venture capital and Silicon Valley backed platforms with massive revenues across these 3 sites alone.
But what’s the estimated value of their organic traffic to each site versus what they pay for via AdWords?
HomeAdvisor – Organic - $2,640,410 – Ads - $706,221
Thumbtack – Organic - $3,309,966 – Ads - $109,852
With the exception of HomeAdvisor, each of them spends less than 10% of the value of their organic search on paid search – so they rank massively organically, and they rank for a lot.
A drop in organic traffic to these sites is catastrophic, and likely very costly.
There’s plenty of search volume to be had on these location-based terms when you add it up.
Every state and city multiplied by every category and service the site lists amounts to hundreds of thousands of search volume pcm (if not millions). For example, “boston photographer” has a search volume of 590 pcm. Multiply that by every city in the US. Get the picture?
How do they all get that organic traffic, and what’s the norm for their vertical that allowed Thumbtack to get away with backlink building in the way they did for so long?
Let’s just take a quick look at Thumbtack on its own to see if there are any clear “other” reasons (aside from the progress points rewards for links issue) that might make them stand out like a sore thumb in their vertical.
Let’s run a Link Research Tools DTOX on their site to check for the types of issues you’d normally associate with a manual penalty.
We’re leaving “dropped links” in the mix because, at the time of this report being compiled, Thumbtack was already “working with Google” to resolve their issues. Therefore, we want to see as much as we can without significant change from en-masse dropped links.
You can see from their LRT PowerTrust metrics (power x trust) the “monster” nature of this site: 42. Wow. That’s 6 Power x 7 Trust.
These kind of metrics are huge and rarely seen. They’re very powerful and very trusted.
To give you an idea of how powerful this site is, it ranks right up there with ebay.co.uk and groupon.com on PowerTrust metrics, and it beats the leading UK-based comparison sites.
But what of the links?
The link audit priority split between high priority (often the types of links you see many of on penalized (algorithmic or manual action) sites, medium and low is surprising.
Only 1% of high priority links to audit?
Let’s look at them.
They’re just a few random links from poker sites with “visit my website” in the anchor and some deindexed domains (mostly just sites no longer really in use).
Meh. Overall, compared with the huge number of backlinks, this is just a tiny drop in the ocean.
It’s unlikely to be dramatic enough as an overall percentage of this site’s profile to cause a penalty.
NB: It's still likely best to get rid of these in 99.9% of cases, but it’s just a tiny, tiny portion of what they have incoming.
Looking through medium priority links shows domains sharing IPs / footprints (possible link networks).
The aggressive anchors we’ve already heard about.
Of course, when we look at some of these referring URLs we see this type of widget “powered by Thumbtack’” but with the money anchor in the link “graphic & web design.”
So many of the anchors are location / state / service combinations. Take this example below:
Those anchors are helping them rank highly for location and service combos US-wide.
However, when we look at their external anchor text cloud, these types of anchors are not visible.
But they are massively diverse and “blended” overall with nothing screaming out as the obvious one or two money keywords (if you exclude [img no alt-text] as not-money).
Hence, overall it starts to look quite normal-ish.
To be on the safe side, let’s dig into the [img no alt-text] anchors and see where they are coming from.
The biggest contributor looks to be some type of “locksmith” domain type name network (e.g. http://locksmithinderbyct.com/ and http://locksmithincheshirect.com/ )
There are lots of them. Hmmm…notice the “inderby” and “incheshire” in their domain names? Now, maybe my geography is not as up to scratch as many people’s, but isn’t “Cheshire” and “Derby” in the UK, rather than the US? Either way, they’re domains built around combinations of locations and services.
They look quite similar, don’t they?
And there are many of these all with the locations and locksmith in the domains.
But when you look a bit further down the page on these sites, you’ll see something interesting.
Thumbtack isn’t the only site being linked to. They’re linking to all the local listing sites, and the links are to the root / home page of each.
Therefore, I would conclude that Thumbtack is likely not behind this network of “cookie cutter” sites.
NB: If I were Thumbtack I would likely disavow these domains to be safe.
But where are all the other anchors we saw before?
[img alt no text] is contributing 13.8% to their overall cloud.
All the others are so diverse that they each contribute a very tiny percentage to the backlink cloud. They’re almost undetectable.
But What Of The Competitors?
Its all about the vertical norm in SEO as we know with every niche having its methods of link acquisition which blends in.
The gambling or personal injury backlink verticals will be dramatically different from that of charities for example. Of course, the point is to blend in (and do a little better) in order to win in organic SERPs.
If they were doing excessive shady link building, why didn’t they stand out before an SEO outed them?
Running a Competitive Detox Report (CDTOX) with Link Research Tools provides LRT PowerTrust metrics on the competitors.
All are equally powerful monster sites: very trusted and very powerful. They’re all right up there with Groupon, Ebay and the UK-leading comparison aggregators we looked at earlier.
Looking at their risk for a negative backlink profile returns all compared sites as being “Below average,” with Thumbtack the least risky of all.
It’s almost odd, given the huge number of referring domains.
And all have a tiny percentage of “high priority” links.
Digging into the URLs, very few (virtually none), of the usual suspects (low end directories, social bookmarking, forum spam, article spinning, etc) are present.
But all of the sites are accumulating a lot of links at quite a pace over the past 2 years: HomeAdvisor more so than Thumbtack.
And a lot of sitewides, too…
…from a lot of domains.
Let’s classify the 60,000 URLs (max 5 per domain) gathered in the Competitive Detox Report (CDTOX) according to anchor text (brand, compound, money, other) across the sites (including Thumbtack).
It’s important to note how keywords are classified as “money” here.
We have to remember that these are local listing sites, so the locations as well as service types are important.
You can’t just look at primary category terms such as “lawyer,” “home cleaning,” “cleaners,” “web designers,” etc. You have to look at those and the combination of location-based words, such as “atlanta,” “texas,” “utah,” or “TX,” etc.
Thumbtack overdid it, but not enough to make them get a penalty without an SEO calling them out and raising the alert.
Even with all the money linked anchors to the profiles, the other sites still have half of the amount of money keywords in their anchors (of the type that Thumbtack have been called out over – location and service combined).
But it’s the “compound” anchors here which are interesting. Compound anchors are made up of a combination of a brand mention and anything else. They’re anchors across all sites which include “find [category] [location] service on [target site] domain” e.g. or “reviews on [category] [location] services on [target site] domain.”
Even anchors you would normally categorize as “other” are more often than not very long tailed themed “what do we want to rank for” anchors.
They contribute to all of these sites in some way because there’s a huge percentage of followed links with a very diverse range of anchors with tiny percentage contribution.
But, it’s not just Thumbtack that’s linking money anchors from member sites (typically badges or awards, or find me at (target site)). Here in Angie’s backlink profile you’ll also find a lot of money / location combo anchors but in such tiny quantities that they’re diluted.
Both “Thumbtack” and “Angie’s List” feature heavily on this list of optimized location- and service-followed anchor text when filtering on the location “San Antonio.” (Albeit Angie’s List has started to switch to just an image alt optimized location / service anchor combo.)
It seems that many of these money keywords have been dropped from Angie’s List’s members at some point.
There are some images with no alt or title in them, as in the case of these domain wides from members to Angie’s List below (all are followed (of course) and passing page rank from related sites).
Look at these anchors to HomeAdvisor: themed anchors with “LLC Reviews” appended en masse. It still all contributes to their overall organic relevance.
It’s certainly likely to help them rank for “clean carpet,” “home maintenance,” “pest control,” “heat and air conditioning,” and so on when you add it to the internal strength and relevance of the site.
But they’re not money anchors right? They have LLC Reviews after them, so that’s ok.
There’s also phrase-matched alts on images, too…
…as well as some fairly obvious money keywords in alts on images.
There are also links from small solar panel business sites with optimized images for keywords they likely want to rank for themselves, but they’re sending that authority away.
Getting The Links In Place
But what if these small business owners don’t know how to link via their badges and review links to these sites?
Angie’s List will show their members how…
…and give them a review link and graphics, too.
There are even awards to be achieved…
Super Service Awards
…for which the winner gets to provide Angie’s List with a followed link back to their site.
There are badges for all types of exemplary service: Elite, Gold, Best Of, and Super Service.
And, of course, if the small business wins one of these, they get to send a followed link back to the awarding website.
They could even get a “seal of approval” for which, you guessed it, they get to proudly link back with a followed link.
HomeAdvisor will also lead the way to the badges.
And they will also provide advice for business owners on how to link build.
According to their article on “Low Cost Ways To Drive Traffic To Your Website,” they advise to “link, link, link: Establish a linking strategy on your site…” and
“…having your link on their site, and their link on your site nurtures a relationship…”
Do you mean like reciprocal linking via thousands of badge-wielding members who link to you?
Isn’t that against webmaster guidelines which prohibits “Excessive link exchanges ("Link to me and I'll link to you") or partner pages exclusively for the sake of cross-linking”?
Isn’t also “Widely distributed links in the footers or templates of various sites” against webmaster guidelines?
Isn’t that just like many badges from small business local sites with followed links in their footer?
It’s like this sitewide followed footer link via an intermediate site www.Abouthomeadvisor.com with template level footer links from www.homeadvisor.com and then out to “International sites” like www.tradeadvisor.co.uk.
They further offend with a handy sitewide followed template footer link from TradeAdvisorUK back to HomeAdvisor.com.
It seems it’s not just Thumbtack then – all of these sites are employing followed badge techniques - it’s just that many of the other competitors do it via alt tags and images on badges.
Thumbtack just pushed it too far with the automated email requests rather than “encouraging” members to “wear their badge with pride” (maybe as part of their onboarding process – who knows?).
It seems rewarding with badges of honor (for which they’ll provide a few followed links each) is not the same as rewarding someone with progress points for their profile.
Is it an exchange of products and services or is it just a reward?
Do the sites who are linking know that they are passing PageRank?
That’s why they didn’t trip a filter.
It’s the norm in the vertical (for anchors, at least).
The Move To “Content” Marketing
I use the term “content” very loosely in this context.
In the fierce competition for organic search, companies attempt to gain authority by trying to provide the content which Google advises us to do in its webmaster guidelines below:
“The best way to get other sites to create high-quality, relevant links to yours is to create unique, relevant content that can naturally gain popularity in the Internet community. Creating good content pays off: Links are usually editorial votes given by choice, and the more useful content you have, the greater the chances someone else will find that content valuable to their readers and link to it.”
Instead of creating high quality content, all of these major players simply encourage badges on tens of thousands of related small business websites with low power and a bit of trust to dominate in local search.
They naturally outrank the websites of the member services who pass authority via their followed linked badges of honor.
An army of small businesses link, link, linking to them and drive up their authority with alt optimized anchors on locations and services.
Of course, they are never more than a few pages or enough anchors to become a major issue (1 themed anchor here, 3 there).
As we previously mentioned, because the vast majority of their linking members are not accustomed to link building, they’re also unlikely to have suffered any kind of penalty themselves - pure juice.
Remember Matt Cutts’ famous “I’m sticking a fork in it” comment?
His rationale for “sticking a fork” in guest blog posts was largely due to the nature of link builders reaching out with the offer of FREE “quality content” (it often wasn’t).
The only commitment on the webmaster’s part was to place a dofollowed link back to the site suggested by the contributor. It took advantage of hobby and enthusiast bloggers unwittingly contributing link equity and welcoming free content.
Is there a difference between this and encouraging sites to wear a dofollowed free badge which in return passes PageRank (and a huge amount of trust metrics when you have thousands of them combined) to local business members in order to dominate in local search?
Likewise, recent advice from Google via webmaster hangout (office hours) to designers and developers who traditionally had tagged footers of projects with a followed sitewide link “designed by x company,” or similar, was to ensure those links were nofollowed. @rustybrick reported on this advice here.
Whether the incentive to link via money keywords or themed phrase keywords or any other type of followed link is the offer of additional profile progress points or via an ego boosting award badge, it gives these sites a huge advantage over all small businesses looking to compete in the local market.
It goes against current recommendations in Google’s Webmaster Guidelines.
“Widely distributed links in the footer or templates of various sites”
Conclusion – We Do Need Stinking Badges After All
Now that Thumbtack has started to instruct their members to nofollow all of the links to their site, they are at a disadvantage compared with the other analyzed sites who still have followed badges.
Is it time to also “Stick A Fork” in dofollow footer and boiler plate area badges, and put an end to David and Goliath advantages?
What’s the point in lamely trying to make the effort to create content to gain popularity when you’re up against armies of Silicon Valley PLC-backed sales teams “selling in” the benefits of linked “badges of honour”?
Believe me, getting people to add badges to their site is very much a “hand-holding” exercise, particularly with non-tech savy small business providers who have not encountered HTML. (I openly admit I have attempted it in the past, but it takes an onboarding team to implement this as part of the inception process or the member FAQ UX, which needs large scale investment.)
Going forward, I will instruct the few badge-wearing members of my small site to nofollow their links to me and ensure that any badge codes include nofollows, too.
Yes, reward people with badges for achievements on levels of service and such, but don’t abuse it en masse for link equity.
Nofollow them all and level the playing field for everyone.
One rule for all.
I’d love to say in the words of Mel Brooks’ ‘Blazing Saddles’, “We don’t need no stinking badges,” but, unfortunately, it appears we do if we want any hope of competing with heavyweights – unless they are nofollowed.
This case study was written by Dawn Anderson, Organic Search Consultant at Move It Marketing, and proud user of LinkResearchTools and Link Detox.
A word from Christoph C. Cemper
It’s comforting to know that Google will penalize one of their own sites. If nothing else, it’s like a coach making an example of an errant player: don’t be like Thumbtack or you can join them in the SERP penalty box.
Get-link-rich-quick schemes like this aren’t sustainable. Though you may succeed at first, Google’s hammer will find you and smash you. Don’t be like Thumbtack and their competitors. Please do yourself and the Internet a favor and create actual content. It’s harder but better, and it lasts a lot longer.
Our goal is to provide our user community and clients with quality service and knowledge. Our Certified LRT Professionals and Xperts are key to achieving this goal.
I look forward to Dawn’s future work, and I personally recommend working with her whenever you get the opportunity.