One man, one billboard, 6.82cm of index finger and thousands of links...
It takes a certain type of desperate attention needing guy to decide to buy a £425 advertising billboard on the side of a Manchester road, luckily for the SEO industry we have such a man!
Going viral during dating is usually not something anyone would want but in this case it seems to have been a good thing!
Mark Rofe as you will all by now know purchased the following banner to try to get a date.
Its safe to say that the whole episode went rather big and since the billboard went up Mark has been flooded with signups, press interest and interviews. The interest has been the sort of exposure that any of us working in digital marketing could only dream of.
Exposure brings links and we all know thats what we really crave when we do our client campaigns so lets settle in and take a deep dive into what the billboard actually generated and what lessons we can learn from the last few weeks of madness.
The billboard went up on the 31st of January and press interest was high from the very start.
How many links did Datingmark.co.uk get ?
Where we stand today
As you will be aware the main link data providers can be a little slow to find the coverage we get but sometimes I don’t think we realise how slow they actually are. The link data sources that we used in this analysis and their totals at the time of writing were: –
It should be noted that Mark hadn’t actually set up GSC when we first pulled link data from the other sources but in the 10 days since they have continued to find new links (shown in brackets below)
3441 links in Google Search Console
Note: When we pull links from GSC the maximum it would give us was 1128, nowhere near the 3441 they suggest.
1423 links in Majestic Fresh (this is now 2200 ten days later – yes it took Mark that long to sort GSC)
Just 370 links in Ahrefs (Now 461)
Surprising how slow Ahrefs has been here finding the links that the other services have. I fully expect the link count to eventually grow to near the Majestic total.
512 links in Moz (more links in Moz than Ahrefs at this point!)
Buzzsumo (will explain later)
135 in Buzzsumo
As you can see there is quite a lot of variation across the link data providers!
If you then take their data and combine it into one master profile (We use Kerboo for this, a business I co-own and a specialist in link analysis)
From the master profile this is what we get as a total ‘link universe’ gained by Mark’s site.
Thats a lot of links for £425 and some either ego boosting or soul crushing embarrassment!
How did the different link data providers do ?
Next I took the link data as a whole and did a comparison gap analysis of the total profile.
When the data is parsed through Kerboo its cleaned to remove a lot of duplication and what we call ‘Ghost urls’ (Im looking at you Majestic!)
If I then compare the total coverage each of the tools has of the total visible link universe for Datingmark.co.uk – this is how they measure up
The best coverage by a tiny percentage was Majestic, look how low GSC was after only giving us half the data they claim they have!
What was missing ?
The BBC covered the billboard at least twice but didn’t link
As well as news interviews etc https://twitter.com/danbarker/status/1223904357000847361?s=20
Ain’t no links from sitting on the sofa
No links but nearly 200,000 visits from that coverage alone apparently, nice…
Many of the big nationals covered the site but a much smaller number actually linked to the site.
Which big sites did link (or linked and those links have appeared in the link data within the first month)?
Lets have a look at the quality of the links gained
By running the data through Kerboo we are able to gain some overall metrics that show the risk of the links gained and the value of those links.
LinkRisk tries to show where the links that Google might dislike are, generally any link that could be argued as potentially placed for the benefit of the linked to site or the linking site will trip the suspect band and these generally just need a quick review by a human.
In general this is a very good balance of risk and reward. The system is always very sensitive to bad and suspect links. When you look into these its clear that the vast majority of links that looked suspect were scrapers, often non UK and probably of very little real risk to the site.
As a general guide this balance of risk would align with any high quality and trusted domain, there are some great links in here.
LinkValue tries to show how many of the links are likely to pass their value in a world where Google ignores a large percentage of the links we gain.
The average profile has more than 90% of its links in the 0-1 band (little value), Marks site has 85% in those bands.
It’s depressing to note that even this site that has had no intentional links built at all still only ends up with 15% of its links probably passing their full possible equity.
Remember though that equity and risk are all relative. The impact that these links would have on the trust and authority of the site is a different argument.
I think that this is a good quality set of links, there are no major issues in the profile from a risk perspective.
Its clear though that even though the site gained a lot of links that increase its trust and authority, it also gained some links downstream that probably don’t help the sites success.
Thats life though !
Some examples of the more problematic links that appeared
Amongst the scrapers and low value sites from around the world, Mark also found himself on a few interesting places… Like this Brazilian porn site for example (Auto translated from Brazilian Portuguese for amusement)
Ive also pixelated the obviously NSFW images.
The scrapers were often from news sites rather than Mark’s site itself. Or in this example, a Chinese scraper of CHEEZburger
Some examples of a few of the best links gained
Now, from a value perspective The Sun is a national media outlet and so we could argue that its a high value link. I think we all know The Sun is a terrible publication and the sooner they are out of business the better….
There are lots of high quality links from all over the world. Testament to the global appeal of desperation.
Even The Telegraph couldn’t resist the appeal of the story!
What about link velocity?
Many respected people in the link game will tell you link velocity is very important and its a clear red flag to Google if you gain a lot of links quickly.
Ive never been in that camp, I think this site is a very good example of why Google doesn’t care about the pace links appear.
Sure if you suddenly get 1000 PBN links thats a problem but its not the pace thats the issue, its the pattern.
Let’s stop pretending that link velocity is a risk factor and remember that the link graph is constantly changing and sites will often lose or gain large volumes for quite logical and natural reasons.
The full profile as expressed by Majestic TF value
Even though we didn’t have full coverage of every link in the Majestic data I was able to gather the Root domain TF for every linking domain. Here’s what the sites links looked like through that metric.
Average overall TrustFlow
Anchor Text Mix
Normal link profiles have an interestingly varied anchor text profile. Mark’s site has a predictably simple mix: –
The worlds pity....
The site generated links from all over the world but as with most press coverage dominated profiles, the majority is on the .com
Conclusions we can take from DatingMark.co.uk
Link data providers
Google would have you believe that all you need is GSC but they don’t have full coverage shown (even if they internally have more data than anyone) and you often can’t get the complete data from them.
Use multiple link data sources to get the best coverage
Even after a few weeks most of the link data providers only had about 2/3 of the possible links found. In this project it was surprising how bad Ahrefs was but they will catch up as the profile matures.
I included Buzzsumo in the project because this was a project that generated a lot of press links and it seemed logical that they might have more of these than some of the others, or at least faster than the others. As it turns out they did about as well as the main link data providers but certainly not better.
Links take time to be found, even if they’re on high profile and frequently crawled sites.
This site generated mostly great links, some really good links in fact. That said, not all press coverage generates links.
The attention the site gained also generated some lower value links from all over the world and even a number of less desirable scraper and NSFW style links.
Doesn’t matter how clean you think your profile is, there is always a set of links that need attention and potentially dealing with.
Site trust and authority
So does gaining this amount of links mean that Mark’s site will now be in a prime position to rank for whatever he chooses.
No, it will rank well for the thematic subject he gained attention for but that doesn’t mean the site has earnt its right to be an authority on the topic of dating. Sadly stuff is more complicated than that.
The site gained a huge number of links in a very short period. Is that a problem? Hell no…
Don’t worry about link velocity
What if we placed a cash value on the links gained?
So… given the success of the site and the links it managed to gain. What would the total value of those links be?
I asked the question on Twitter to see how people would go about estimating the cash value of a link profile. The excellent Ross Tavendale of Type A Media gave this useful answer
So following that general logic (and putting my own twist on it given the slightly different metrics we had in the profile…)
The total valuation of the links generated so far comes to….
£488,703 in links generated !!!!!
Are they really worth that ?
Of course not but it shows the power of an idea and a story that people can connect with and run with. (If you just take a one link per domain calculation it comes down to about £200k, still amazing though!)
Well done Mark, nice work chief…
Riseatseven are lucky to have you … probably….
And yes… for those that know me… I was the first (and as far as I am aware only) person to ask to buy a link on his site so far lol…
Like everyone he’s tried to date… The answer was “No”
Note: If you’re interested in the traffic that the site generated then I can fully recommend this excellent thread by the amazing Dan Barker… one of the most intelligent and nicest people in search…`