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.
This post is quite old, but we kept it online for your learning and insights. Some parts of the application may look different today, usually much nicer - don't we all get prettier over time?
Interflora - Bad Link Penalty Analysis
I am sure you’ve heard about Interflora.co.uk being penalized by Google last week. In fact, well-respected fellows Martin McDonald broke the news and Dave Naylor already analyzed and closed the case. Irishwonder contributed as well to the discussion that Barry from SearchEngineLand already commented on and notable regional newspaper page rank penalties were documented, as well as Matt Cutts' “fresh reminder” about paid links.
Reminder: Google's guidelines on paid links that pass PageRank also apply to "advertorial" pages. See http://t.co/e7YluzHGas for more info. — Matt Cutts (@mattcutts)
Now I do second all of what my fellows said about the advertorial issue with regional newspapers as well as several link networks built with not-good-enough camouflage.
But there’s more to it!
When starting a simple Link Detox today (I’ve been OOO for 2 days for family reasons) I realized that Interflora had over 70% toxic and suspicious links and that made me dig deeper...
You can now read more about
- Cheap advertorials
- Tiered blog network links
- Simple paid blog post
- Page Rank penalties
- Mediocre or just plain stupid setup link networks
- LRT Power*Trust
- Oldschool directory links
- More and more bad stuff to be found
- How Interflora is doomed now
- how YOU feel now
Advertorial campaign on regional newspapers
No question – those advertorials were poor – very poor, and Wiep already joked about the mass advertorial
So *now* they tell me mass buying advertorials isn't the same as content marketing... — Wiep (@wiep)
and I think he’s damn right, especially if you look at pages they had (links have been removed very fast)
I mean marking the advertorials is one thing, but publishing those really bad auto-spun posts is another thing.
I know Interflora has been aggressive for Valentines Day 2013, like every flower-shop should be, but above samples just testify an ongoing messy link building strategy – quantity over quality. And while those domains could have had some trust in Google, they just all got penalized.
By the way, I quickly found those samples by using the “Restrict sites” feature in QBL, which is basically a “domain:” operator on steroids, ...
I gave it the 54 domains that Dave Naylor mentioned in his post having a PR penalty and got all the actual links to Interflora a couple seconds later.
(make sure you don’t let us remove the dropped links in this case)
And voila – some of those nasty advertorials on those penalized sites picture here
(3 of those examples are listed and screenshot above)
Ok, fair enough, those are really cheap made mass advertorials, and they won’t pass a manual review. Please note how all of them have FOLLOW links and are marked as paid links. Not the smartest thing to do. Dave’s opinion is very much that such cheap advertorials triggered the penalty. Is that really all, or is there more to see?
High level – what about Power*Trust?
Before diving deeper into all those bad links I thought that a competitive landscape analysis would make sense (using our Competitive Landscape Analyzer) and thanks to the quick mode those 70k links were brought back
within a couple seconds. I knew I wanted more, so started the detail mode right away against 10 competitors I hand-picked
The first graph already struck me
Please note those orange and green arrows are mine, should just emphasize the very obvious difference in those link profiles as measured by LRT Power*Trust.
As we can see immediately, that Interflora link profile lacks a good chunk of highly trusted and powerful links as compared to their competitors.
So whatever shady stuff we continue to find as we dig deeper into Interflora’s link profile, this isn’t good for them, and it was only a question of time when they got hit. In fact I wonder why Interflora wasn’t penalized earlier, given that a lot of shady stuff happened last year already. One possible conclusion might be – and I’m totally for it – that they just “overdid” it.
A big brand gets away with a lot of stuff, but at some point the threshold is reached and we found similar “limits” for traffic drops when we analyzed those unnatural link warning mails last year.
So that’s worth a closer look. Let me repeat, I totally believe that those recent newspaper advertorials for Valentine’s day triggered a threshold or forced someone at Google to take action, maybe simply because of one or multiple spam reports.
Tiered links – SUSP1
Taking a quick look at other suspicious links, we start with SUSP1 - we got 2382 or 46.6% of all links in this category. SUSP1
Page has no LRT Power*Trust™ and LRT Power*Trust™ Domain < 5 - a page without external links on a weak domain
Just have a look at this beauty. Nice pictures, some “unique” content and those two money keywords linked to http://www.interflora.co.uk/category/roses/ and
Well, some people might put the second link to Wikipedia (aka “Co-Citation to Authority”) but in this case it’s just a link to another free-blog.
That one indeed again links to only two sites.
http://www.interflora.co.uk/category/roses/ and http://flowersbypost.jimdo.com/
Hmmm... someone fancied roses as keyword and target url, did they?
Guess what, the next one again goes to http://www.interflora.co.uk/category/roses/ and rosesforskin.blinkweb.com (another free host) And that one again links to two URLs only
http://www.interflora.co.uk/category/roses/ and http://easy123-roses.livejournal.com/634.html
I think we can stop traversing the link list here – it’s enough to demonstrate how a “link-wheel” used to work.
Those free-hosts sometimes only had ONE page posted, a typical pattern. And needless to say, the links going to those free hosts were spam as pictured below
or just nada – no links at all. I have to say tough, that the “feeder links”, so those spam links powering those spammy free-host pages were very few.
Under usual blackhat practices one would have sent a couple hundred or thousand of such profile link to the free host parts of the link wheel. Here it was either intention to only get them indexed or scariness to “stay below the radar” with not overdoing the nastiest part of link wheel creation. Or the guys that set it up simply didn’t want or could spend their time making this a “full scale” black hat link pyramid.
Enough said, the pattern is clear and the implementation is so-so for blackhat standards. However, those links are still out there and removal is probably a big pain as passwords for automatically generated pages like those, 2 or 3 years ago, are probably lost. What we do not know at this point is, if those links were already disavowed.
Paid blog posts on abandoned sites – SUSP10
A quick review on those 23.8% (still 1216!) suspicious links according to rule SUSP10 reveal some insights – here’s the SUSP10 definition SUSP10
Domain has Link Velocity Trend < -70%, LRT Power*Trust™ < 16 and LRT Power*Trust™ Domain < 16 - probably part of an Expired Link Network or a domain that the public has lost interest in. So generally that means links from sites out of favor, have a quick look and judge for yourself... on this one http://www.miasprom.com/page/2/...
We found the roses again – linking follow to http://www.interflora.co.uk/category/roses/ and that page on http://www.miasprom.com/page/2/ really links out to every link spammer in the world as well, talking about nurse uniforms, Expedia discounts, shoes at Shopwiki, Mascot costumes and other kinky stuff. Anyways, looking at it, we identify a typical “paid blog site” where one of those poses went between 10 and 50 bucks onetime. Not a bad cut given that the post was most often simply posted automatically or triggered by the link buyer.
Another sample is this url at http://containergardeningexpert.com/articles/making-the-most-of-what-space-you-have/ and I think that screenshot says it all
Please note that flower deliveries link goes to another money term and page on Interflora http://www.interflora.co.uk/content/flowers-delivered-florists/ It’s been a while since this link was posted, and the site has obviously been abandoned since.
TOX1 sites, denindex
What it says is clear. And often a clear sign too... TOX1
Domain is not indexed in Google. Usually a sign for a penalty.
Read more A couple of those
And I leave it up to you to judge it that would be a link to keep. Looks like subdomain spam to me, and Google just recently mentioned that...
Google advises against Subdomain-Spam - like abcdef . blogspot . com ? :-) http://t.co/Um0fRvYP — Christoph C. Cemper (@cemper)
SUSP14 – Pagerank penalties
Once again a clear signal of poor links is when no pagerank (i.e. grey bar) is there and there are still links to the page. That’s not natural at all. SUSP14
Page has no PageRank™ but at least some weak links. This could be a sign for a page punishment by Google. Google PageRank™ is more and more inaccurate and many webmasters don't trust that metric anymore. And looking at some of these 224 pages (4.4%) we find
e.g. http://www.prettysinglemom.com/2012/11/flowers-from-heart.html and http://www.methinkyou.com/2012/11/flowers-and-decorations.html
which all have keyword-rich urls and pretty poor layout and text quality, of course including a money keyword link – what else. No need to red-highlight those for you, or?
SUSP15 – Typical web directory links
Yadda yadda, web directories are dead, links still up, as no one bothered to remove them. Not looking good at all
And the point here is – links from pages like http://www.submitocean.com/shopping/giftspage-8.html might be old, but Google still sees them.
They influence your google-credit-rating, no matter what. And that link simply looks and feels bad
BTW did you realize the name of the site?
SUSP6 – Same Domain Owner – Link Networks
Finally as Irishwonder also brought this up, there’s simple minded link networks, and bad content with money keyword phrases add to their worseness. I promise this is the last suspicious rule I will comment on before leading on... 161 links or 3.2% are SUSP6, which is SUSP6
Domain has the same REG (Domain name registrant) as other linking domains - possible Link Network. And it looks like this
In fact a sample page looks like this
Lovely, ha? Be aware that the page here http://www.barnstaplepeople.co.uk/Gardening-summer-June/story-16218432-detail/story.html looks just as spammy as http://www.belperpeople.co.uk/Award-success-Interflora-RHS-Chelsea-Flower/story-16175370-detail/story.html and http://www.beverleypeople.co.uk/Gardening-summer-June/story-16250337-detail/story.html
It’s obviously lowest quality, isn’t it?
I am frank here – since we started Link Detox last summer I’ve seen so many bad link building results that I got really shocked.
Not shocked about people delivering such poor results, but actually clients accepting these as work results, and merely quality-checking on what they got from their suppliers.
So let me say, I'm not surprised here, and I wouldn't be surprised if no-one at Interflora had checked these "results".
Interflora is doomed – because of the deep dive.
Interflora have simply overdone it. As Dave and Matt pointed out, those advertorials were a really bad idea, but looking at these links from 2 or 3 years ago everyone and his dog will realize that Interflora has been doing unnatural link building for years. Not surprising, given that it’s a pretty boring topic to write about a flower shop. An undisclosed Google engineer once said to me about a heavily spammed (and undisclosed industry) that they simply don’t have enough good signals. That means, Interflora got away with it, probably because they were just GOOD ENOUGH for that industry. Every vertical is diffent – competition rules are defined by country, keyword space and language. Whoever echoes stuff about X percent of something and Y to Z ratios has simply never worked in more than one industry or none at all. Anyways, Interflora is now undergoing a deep dive. They HAVE to file for a Google reconsideration request, and given they lost (almost) everything organically that’s a wise thing to do. They HAVE to cleanup everything we saw in this post and a LOT more.
What Interflora should do...
To be very clear, Interflora got away with a lot of bad link building for a while – they just tripped the 70% mark, but actually that’s pretty late given that we saw 3 major penguin updates in 2012. However, NOW a Google spam engineer will soon or already has been looking at all these old nasty things, and will ask for them to be removed. Is it simple, no. Does Google care? No. Guess what, Interflora has been getting away with all these spammy tricks for years and should send a big chocolate cake to thank Matt Cutts for not taking action against them earlier.
Are you clean?
The question now is, are you clean? What would you do if you trip a filter because of those old and bad habits? Don’t get me wrong. There used to be a time when forgiveness was pretty en vogue at the Google spam team, but nowadays you can get in deep trouble for your old spamming games. It’s very similar actually to banned Adsense and Adwords accounts. Once you’re blacklisted they just refuse future business. You should take care about your backlink profile today, not when you get booted. Obvious pitch: If you want, you can just go and get a link detox account to review you own site.
Let me know - what have YOU found about the Interflora Link Penalty?
Christoph C. Cemper
[…] Backlinks if done correctly can help improve your website’s traffic, rank, reputation, and even relationships. If done incorrectly backlinks can lead to a Google Penalty as happened to Interflora a couple of years back. […]
It’s amazing to read through this again a few years later, especially the comments and see just how far the Google approach to link building has changed.
[…] SEL article concluded that there appears to be a lot of bad SEO going on but these are a few of the most obvious tactics they used that likely drew the Google […]
[…] analysis revealed that the advertorials weren’t the only issue, however. Seventy percent of their links were labeled toxic or suspicious according to Link Research […]
[…] So, when it comes to tactics people have employed (or purchased) to help rank their sites, well, little surprises me any longer and this week, we saw another big brand in Interlflora punished for taking their spammy SEO practices one step to far. […]
[…] they’ve been well covered already by Martin McDonald, David Naylor and (in excellent detail) by Christoph C. Cemper. The general agreement is that the penalty was most likely a result of sponsored blog posts (via […]
[…] “Thanks for letting me know. Unfortunately I cannot allow this, as paying for links is actually against googles terms, therefore if google came across a post on your site marked in this way, then your site could get penalised, as would the brand. I could allow this to be marked as ‘guest’ ‘feature’ or an ‘associated’ post, but ‘sponsored’ is something that google has picked up on, therefore risks you losing your page rank. Let me know whether this would be possible at all, as I couldn’t agree to it otherwise.” My response: That’s not quite accurate - It is not Google that stipulate whether something can be labelled, they only give guidelines on follow and no follow. In terms of labeling, the law states that all paid-for content must be labelled or I could be prosecuted. If you need any further info, I’ve found this very useful: GO HERE Let me know how you wish to proceed. Their response back to me: Would it be possible for you to label this as featured or guest post? It isn’t actually advertising that we are looking for – it is SEO. This is why we request that the links are very subtle and discreet, as if not, google can penalize blogs, in a similar way to what happened with the Interflora situation. This may be helpful to you - https://www.linkresearchtools.com/case-studies/interflora-penalty/ […]
Great post Christoph - very detailed and very insightful. The Link Research Tools really do allow in depth analysis of link profiles. I already planned to get an LRT account for my agency, but this case study managed to convince me that I definitely have to take action NOW lol - thanks Christoph and all the best from Munich! :-)
[…] Kurz nachdem es Interflora erwischt hatte, kam es zu einer weiteren manuellen Aktion von Google: Dutzende britische Tageszeitungen verloren ihren Pagerank. Das stand direkt in Zusammenhang mit Interflora, weil die meisten dieser Tageszeitungen eben auch “Advertorials” für Interflora.co.uk enthielten. Hier kann man also von einer Folgeaktion sprechen. Christoph Cemper hat hierzu einen unglaublich ausführlichen Blogpost geschriebenn und bringt viele …. […]
Every website has good and bad links. The point here is because it was a famous and big brand Google put its efforts to revoke manual spam penalty so early, I haven’t seen any quick recovery before when there was penguin introduced for penalizing spam links. If so than what the hell was wrong for small business owners who’s income is dependent on it. Why google is not taking these steps for them also and revoke their penalties this early. The small business owners hardly get reply so soon after a month or more they get the reply.
“The only thing which hits my mind is Rich gets Richer and Poor gets Poor.”
I am a florist with over 25 years of experience in the flower industry, including 13 years of experience as an Interflora member pre-incorporation and 1 year after incorporation. Obviously this situation is very interesting to me as it could have a big effect on our industry especially over the looming Mother’s Day period. Reading this and some of the other comments and articles about it I’ve found it a little frustrating that no one seems to have much of an idea how the Interflora system works and why they so vigourously pursue order gathering in this way. I thought those curios might find my post on a florist forum of interest. You can read it here http://www.floristnews.co.uk/flower-chat-public/23979-how-does-interflora-work.html
I’d like to get the message across to our consumer while this iron is hot how much better off they are to deal directly with a local florist in the area they want flowers delivered to rather than using a middleman such as Interflora. Dealing direct with a bricks and mortar florists, especially those that display the Interflora branding as they are generally tried and tested, will get you a far better quality and value for money product than placing your order through either what is know in the trade as an ‘order gatherer’ or ‘relay provider’.
This is an excellent directory of real florists www.goodfloristguide.com. If you can’t find a florist listed there for the area you want to send to give them a call, they will happily give you details of a reputable florist that can help you. The directory is produced by people who really care about quality and service and whilst they do not charge for a listing they do have personal knowledge of all of the recommended florists shown on it.
I hope that helps explain things from a florists perspective :-)
@timmayer but then, without tools and monitoring you maybe get this crap after a couple years of “luck” http://t.co/QFEngM7iHe
[…] The Link Research Tools post was very interesting in terms of identifying the types of links which are likely to be having a negative impact. In terms of if one type of link is worse than another, I’m not sure it quite works like that – anything that looks paid and over-optimised is always going to stand out to Google as a potentially unnatural link, irrespective of the type of link or domain it’s hosted on. […]
Unfassbar starke SEO-Analyse! RT @cemper: Deep Dive into the Link Penalty of Interflora http://t.co/El7JZaMLBq #seo
[…] Interflora – Bad Link Penalty Analysis by Link Research Tool, which probably is the post that will resist the pass of time, as it is not just about the Interflora penalization, but a wonderful example of how to do a links’ analysis; […]
FAIL: Once again Google reminds us that it will punish its search customers with poor results in order to punish a website!
A search for interflora now results in me finding vouchercodes! Well done Google!!!! I am looking for a website and you send me somewhere else. Idiots!
Interflora’s Google Penalty - a Deep Dive & hugely insightful look into their Spammy Link Profile http://t.co/dESyg5z0qZ #seo
Interflora Penalty – Link Penalty Analysis http://t.co/MxT0oCmC7J #svSEO
@cemper btw my tweet on #Interflora w/your webinar http://t.co/CkyufsLsZx is my highest since fall. So many need more info about Links & SEO
[…] you like data? I bet you don’t like data as much as Christoph Cemper. His Interflora Bad Link Analysis post on February the 24th was simply amazing. He concludes that over 70% of the links pointing to […]
@cemper Can’t wait for tomorrow’s Interflora Penalty Webinar!! http://t.co/CkyufsLsZx
@keyserholiday no doubt.But I was looking at @cemper post here >> http://t.co/jfJTt3INpf << clearly something is too pushy.
Yet more brilliant analysis of the reasons behind the Interflora SEO penalty: http://t.co/ZhE6U8eOi3
A surgical dissection of the Interflora penalty http://t.co/UW3N10OHm8 by @Cemper. Chock full of #SEO link building analysis. ^
One of the best analysis around the Interflora penalty: http://t.co/9qOtHwkP9A #seo
[…] diesem Fall sind vor allem Footprints, die niedrige Textqualität, die hohe Anzahl der Links, das Linkwheel und die harte Verlinkung aus sponsored Posts der Seite zum Verhängnis geworden. Das einzige was an […]
Juiciest Article so far on the #Google Update #Interflora Penalty!! http://
http://t.co/CkyufsLsZx via @cemper @vivitron @AussieWebmaster
>how easy is it then to do bad backlinking against a competitor?
It is NOT easy which means it is not cheap. Negative SEO is something not only questionable from ethics, but also from economics.
You can sign up to a blog network subscription for $97 a month and take out a competitor within about 3 months. It actually helps them in the short term, before they get hit.
The only reason Interflora recovered from this is because they are the recognised leaders in the UK and really should be number 1.
Google knows its fighting a losing battle - they absolutely cannot combat link spam algorithmically. we have 140 clients ranking very well with very similar tactics, and occasionally we’ll lose one and rerank them with a new domain within a couple of months. We’ve had 3 client sites hit this week (all uk sites)!and my guess is some VP started going on the rampage in a google uk managers meeting and immediately ordered a task force to go out a manually hit a bunch of websites using similar tactics. Everything will calm down and go back to normal in a few weeks because its costs money to manually police the web - money they don’t want to waste.
[…] (Los datos del gráfico son sensiblemente distintos que los presentados por Christoph Cemper aquí). […]
Their agency is already scrambling to fix the problem. I’ve published advertiorials for Interflora in the past and they’ve already asked me to remove them. Which is fine, they were crap anyway.
Fascinating insight - the fact is link building is getting harder and harder these days. Article directories, web directories, link wheels, forum profile spam and many more have gone by the wayside but what intrigues me is why Google still holds backlinks in such high regard given they can be manipulated so easily - especially with the mass link building software that is cheaply available. Surely a move to greater emphasis on social signals must be in the pipeline!
Its worth remembering, though, that its easy to be critical of past link building practices but the fact is they worked, and Google did not always discourage them at the time. I therefore seems a bit harsh to penalise them in the light of changes that happened down the line.
Lets face it, most SEO companies wont admit it, but its all a game and every kind of link building, link baiting or link begging is grey hat at best.
Grey Hat=walking uphill towards a cliff and never know which step too far. Esp if the cliff itself moves backwards… http://t.co/n2WatqMW1B
Wow … This kinda reminds me of the last time Google Attacked the Flower Industry :: What? Ten Year Anniversary of The Florida Update?
Roses, roses…roses! > Interflora Penalty – a Deep Dive into Bad Links http://t.co/h6vyA1RbsL
Bachtiar, actually I think they got away for a while because they are such a highly visible brand, and just that made them suffer now - it’s a perfect example for Google to give and very visible.
Very visible because of who SEO’s? We are Google’s unpaid PR team.
Good point and always the discussion if we should reflect on what Google does. Actually sharing of insights and analysis of ranking changes/algo changes have been the basis of the whole SEO industry, starting in forums, going to blogs, social, conferences etc and I don’t see a reason to stop that.
In fact Google themselves got just A LOT smarter in their PR actions, no question on that, but not responding wouldn’t make sense either IMHO.
Interflora has just enough good signals in a spammy niche. Then paid advertorials crossed the line = manual penalty http://t.co/NRiiRIhNv6
Very interesting post - Great shame for those that might have had their “SEO” and “Linkbuilding” done by the same company used by Interflora - who no doubt would have used the Interflora brand to bolster their credibility.
Excellent, showing the value or LRT. Thanks for sharing!
You are very welcome Nathan!
I think that’s a really interesting point about them ‘not having enough GOOD signals’ in certain niches and helps illustrate just how different this can be from industry to industry.
As the obvious and most stinky example in the world the pay day loans market is still super spammy but I have a few new clients in other similar financial sectors where the abuse is rampant and a lot of folks still seem to be getting away with it. But, if you dig in, there is just no one doing anything solid or natural so I do wonder how does an honest site compete in these kind of niches?
The one thing I have seen in some spam heavy categories is that there is often no one working on the long tail and some good, honest content can jump right up to the top of the page if you can just convince your client to stop focusing on their trophy keywords for a few months to show how that can build even more traffic.
Still, it’s reasuring to see that they are still continuing this crackdown and whilst they may not have signals today, they soon will have.
Great analysis by the way, very cool.
I cannot agree more - every industry is different, even the something called an “industry” i.e. financial sector works different in different sub-categories of it, think Mortgages vs. Payday Loans.
great case study over the whole advertorial saga everyone’s freaked about. Interflora just has a shit link profile http://t.co/3Z1wT3zqoS …
Is your #SEO without reproach? Here is the antithesis: http://t.co/no4dZeATWV
The newspapers selling PR passing links has been an ‘issue’ for ages, so it really isn’t a huge surprise that they drew some fire, eventually.
What is surprising? Why it wasn’t picked up by one of the Penguins!
The amount of historical links built, many of which weren’t penalty fodder (directories/bookmarks), are a concern for so many companies now…. Identification and pro-active cleanup is on the minds of many.
Solid post research and a great read, thanks.
Robb, very valid point re: Penguin Updates. I always had the feeling that Penguin 3 didn’t fully rollout to UK despite shakes. Different brand structures could be a reason… gotta sleep on that
[…] Después de los análisis de ayer, también Gordon Campell hizo el suyo: “Interflora first, then the Newspapers, next you!” y Christoph C. Cemper: “Interflora Penalty – a Deep Dive into Bad Links“. […]
Excellent Analysis. Looks like a great tool. Can’t wait to have a shot:)
Great - let me know if you need anything or we can set you up for a full Superhero trial.
So do you think now that all websites within the flowering niche will be affected, is it as simple as Google have flicked a switch on the flowers niche.
[quote]An undisclosed Google engineer once said to me about a heavily spammed (and undisclosed industry) that they simply don’t have enough good signals. That means, Interflora got away with it, probably because they were just GOOD ENOUGH for that industry.[/quote]
Or is it that Interflora have just topped Google as the worst offenders and that’s why they looked into the niche?
as my friend said
There were a lot of pretty obvious links and a link network to boot. However, as it is a super competitive industry, this was to be expected — they are no worse than the results that now replace them. In fact I’d say the user is now worse off because these guys represented an end node for the customer experience and not a thin affiliate or junky result.
that being said, I don’t think Google has specifically picked the flower shop niche now. This statement referred to even more competitive industries, where spamming is an ongoing thing and 8/10 results are typically auto-spammed sites.
I just think that Interflora overdid it, and chances are that others will follow, but then, which flower sites should Google list, they all basically play by the same rules,
so doing a deep dive on their competition probably brings up similar old sins.
This is one of these warnings Google puts out. The flower industry is supposed to run for cover now and clean up its act. I simply hate the fact that Google cannot find all the offenders and penalize all at once that is the only way the link manipulation will stop. Website owners would have to stop taking short cuts to rank and it would be better for users and Google. But they still after all these years cannot get it right. While it still works as long as Google has not caught you it will continue. They have no problems catching the Mom and Pops and they took out thousands of them with Penguin never to return but the big brands just keep on breaking the rules and they will keep breaking the rules because the chances of Google taking them out are small. Even if they do take them out they know they will get back pretty quickly. Its the unfairness of it I take issue with.
The SEO industry plays right into their hands with the publicity they give to these isolated cases in other words we do Google’s work for them. The next time they penalize a big brand how about the SEO industry gives it zero publicity. We need to stop doing Google’s work for them if they want to get their linking issues resolved they need to fix the algo and reward content and engagement not links but that would take a huge investment by Google and they dont need to while they have thosuands of SEO’s hanging on their every word and spreading the word for them.
It does seem to be whoever is a whiter shade of grey does well, I don’t think that there are any results within the top 10 of most searches that are 100% white hat. But the question is will that ever change? I can’t see google changing their algorithm enough to show wholly white hat websites as the content may well be relevant but the services/products will more than likely not be what the users are looking for.
Certainly every site has a few bad apples in the basket, but what Interflora did was manipulation at an extremely large scale and in a very obvious way. The fact that they didn’t get smacked earlier shows that Google is pretty forgiving when it comes to huge brands.
The problem with any grey hat stuff in general is: You are walking uphill towards a cliff and you never know which step is one step too much. Especially if the cliff itself moves backwards…
WOW, great comparison with the “moving backwards cliff” - love that phrase.
Google is an ever changing set of rules and the future in link buildign means Risk Management and Risk Assessment, something we we introduced with Link Detox on an automated level, but it needs action as well, especially if you’ve been stretching limits like Interflora and others in that niche.
Another through post. I think the mistakes have been obvious and them getting away for 2-3 years must have been good for them until now!
Love your simple in-depth style of writing about SEO research. Cheers.
What I find very remarcable is that Interflora (or their SEO company) thought that Google would not take action against all these advertorials.
As you point out all of them include the words “advertorial” / “advertorial feature” / “commercial feature” on the page where they were published and include follow links; something that must make Google suspect.
I was offered some advertorials from websites in the UK last year, but all of them had in common that they would include the word “advertorial", which seeemd very spammy to us.
Gordon Campbell also points out something similar: “Interflora first, then the Newspapers, next you!“
A few months ago I received a phone call from a company trying to sell me advertorials on large media sites. It seemed like a really good offer at the time, especially after Penguin considering the trouble that you now have to go to get decent high-authority links.
The only thing that I still don’t understand is why Google didn’t take action before Valentines Day 2013.
thanks for your feedback! Honestly I believe that Google had no special intention to “keep them up” for Valentines day. They probably saw that spike and/or got reports and took time to react, like every major corporation has it’s slowness. I wouldn’t interpret anything into the timing actually.
This is a message from Google to say that nobody is safe nor matter who you are. If you fail to conform to the guidelines, you will get hit. Having said that, you would think such a big brand (or their marketing agency) as Interflora would avoid these obvious spam link building tactics. Link wheels, spun content distribution and network sites has never been good.
As the new lead SEO at a company in a very competitive area I have recently been leading a complete link structure review alongside a competitor analysis and the findings have shocked me. Whilst our current link profile is not perfect it was the absolute disregard for the rules that our competitors, some of who are household names, have. If the link structure that Interflora has means Google sanctions then expect a few more big names to drop in the near future. In the meantime our SEO approach is to become without reproach, it is a little more time consuming to do things right but worth it in the long run.
Over 20% of sales for US ecommerce ($42B) goes thru affiliate channels via CJ, Linkshare, etc. Many times fashion bloggers, couponers write articles instead of posting ads on pages. What rules should they follow in light of this? Anyone know? Thanks
As always great analysis. I agree with you concerning interflora sending Google a big chocolate cake.
What I find more interesting about this case study is Interflora’s brand status, and how long it took for Google to penalize them. I personally believe that someone tipped off the spam team and it wasn’t an algo bubbling up the site to Google’s awareness.
Brand’s simply can get away with so much more, and what you shared in this post is proof of that. Lesser sites with not as poor linking tactics didn’t even make the first Penguin cut.
Dave Naylor hit it right on about how Google wants brands to be at the top, but has to slap their hands when shady tactics are being done. I would be very surprised if we didn’t see Interflora back in a few months, just as we did with BMW, JCPenny, and Overstock.
Again, I really appreciate your in depth case studies, on both the bad and good of SEO. Looking forward to seeing much more! Cheers!
What is wrong with directory websites?
What is wrong with a ‘dead’ directory website?
Steve, Web directories are sites made to spam the Google index and are a cheap and obvious “trick from the past” - just forget about directories, once you cleaned up your old sins.
Are you kidding me? Directories were the way people searched the web before Google. But now we have a big a$$ company spreading propaganda and all over the sudden it is a “cheap trick".
Not “all over the sudden"!
Directories lost their value to users since a behaviour shifting towards search engines took place. Like social bookmarking it is a technology changing from “users & webmasters use it” to “webmasters use it, but users don’t anymore"…
Google could perfectly disregard links that come from directories. But they chose to punish instead. I had to remove myself from a few directories that were bringing me some traffic. Why is that? Google is already the main entry to the web and intend to become the only one. Directories lost they values? Fine, just stop ranking them and disregard the links. But let the users decide by themselves and don’t dish out penalties.
See my rant about google’s tyranny: http://blog.linkody.com/seo/google-tyranny
Ah nice I had the feeling it would be something along these lines, so the question is how many other brands have links from these sites that will be next to get hit?
The big problem with any SEO trying to do a link removal with .co.uk links is the difficulty in getting the contact details of the domain or website owner as typically WHOIS searches don’t turn up anything outside of postal details.
Awesome analysis as usual.
The irony is that a lot of niches are still dominated by brands with a very similar link profile. I guess it’s a case of then not triggering the manual review which is why they get away with it?
Very well said!
A good friend also commented
There were a lot of pretty obvious links and a link network to boot. However, as it is a super competitive industry, this was to be expected – they are no worse than the results that now replace them. In fact I’d say the user is now worse off because these guys represented an end node for the customer experience and not a thin affiliate or junky result.
In fact as to quote another friend of mine:
It’s amateurish… they overdid it so much, thereby breaking rule #1: don’t make google look stupid.
My point, if it didn’t come across yet, Interflora is just ONE of the many possible future penalties. The tactics we see here are tactics that were somewhat “OK” only 2 years ago, simply because “it worked” and you got away with it.
I am sure if we deep dive other competitors in that space, we’ll see similar things.
And yes, the manual review was the reason the penalty fired, very sure abou this,
and could actually create another case study how well (apart from Power*Trust) they blended in.
This post has 44 feedbacks awaiting moderation...