Switch from .com to .co.uk to clean link profile - thoughts?

[FranticFish]

I'm about to get involved in a major website upgrade for a web-based business that has been online for about 10 years with one domain during that time.

Presently, Google is everything to them, and they do well by Google. However, this year we plan to move away from this over-reliance on one source of traffic, by undertaking a major content strategy on the site and also by building lots of local partnerships.

Our main aim is referral traffic, but we'll obviously be taking links where we can get them too, and herein lies the potential problem.

The site has a lot of old 'SEO directory' links from about 7/8 years ago. I have first hand experience of Google's amnesty on old links - as long as you don't get any new ones, you're OK. But if you do anything to change your profile, then you risk being out of action for several months at the very least.

They currently publish the site at the .com, but they also own the .co.uk.

So my plan is this...

1) Mirror the site on the .co.uk, but noindexed in its entirety.

2) Do any outreach using the .co.uk domain name; if we spot any new links coming in to the .com, we'll ask people to switch to the .co.uk

3) Build up as complete a profile as we can of the links the .com has, and work out those we want to switch

4) When I feel the .co.uk has at least as much clout as the .com, we'll then ask the owners of all the good links that the .com has to switch to the .co.uk.

5) We then let Google into the .co.uk, and take the .com back to a single page with a click-through to the new site (no redirect)

The aim is
- to avoid mass link removal (for all I know, some of those crappy links could be helping if they're grandfathered in), and
- to avoid pointing any potential crap links at the new site

Would very much appreciate any advice / experiences.

[netmeg]

Personally I'd be afraid that I'd be throwing the baby out with the bathwater on this one, so I don't think I'd do it.

[FranticFish]

I can't have a moratorium on new links to the site though, and I have to do something. Leaving things as they are isn't an option either.

I have seen first hand that new links can cause old links to be re-evaluated and punished (with the penalty only lifted when all the old links are removed, which took months) so starting a campaign for the old site carries significant risk in my eyes.

[RedBar]

However, this year we plan to move away from this over-reliance on one source of traffic, by undertaking a major content strategy on the site and also by building lots of local partnerships.

I don't understand this, what do you expect to gain by this change? Hopefully to get traffic to the site from alternative sources?

Do the local partnerships have plenty of organic traffic or is theirs coming from Google?

I'm with netmeg on this at the moment, on the surface it seems to be rather a strange thing to do.

Why can't you not carry on with the current set-up but start adding local partnerships?

Yep, I'm puzzled.

[Shai]

I would say that's a very risky strategy. Especially if the company is so reliant on traffic. I presume you are not under a Penguin penalty already?

Instead, I would create a complete list of link pointing to your site and manually go through them one by one and check for spammy links. Categorise those links into 3 or 4 levels of toxicity and create 3 or 4 separate disavow files; One for each level of toxicity. I would then disavow the links in stages. Start with the worst and slowly, as you gain more good links, disavow the rest. If possible, make sure that the files you create have relatively similar amount of links. Especially, be careful with the first file by making sure it is not massive sub-section of the overall list.

To be extra careful, you may want to attempt to actually remove some of the link on those lists. At the moment it not necessary but may become more of a requirement in the future.

I would definitely stick with the .com unless something drastic has already happened to your traffic and even then there are other options to consider.

Good luck.

[FranticFish]

Thanks for all replies, much appreciated.

what do you expect to gain by this change... traffic to the site from alternative sources?

Exactly that. There will be some 'advertising' to this, 'nofollow' banners, mailing lists etc. However we're also going to be asking for links to all the new content we'll be adding. I'm not about to ask people to 'nofollow' editorial links to me - if I've earned a link through content, then I want the juice.

I've already thought about disavowing, and it concerns me a great deal.

The key to my mind is that I cannot guarantee to gain new links at the rate I'll be removing old ones, because I'm not buying them. My experience (only on one site but it was a tough lesson to learn) is that as soon you do anything to a link profile that's been untouched for years then you risk a big fall (between 5 and 9 pages). In the end I came back far stronger but it took time, and the business didn't depend on Google. Top rankings were a bonus, not a necessity. Here, I can't risk a loss of traffic - it would be the end of the business in a few months. But we have to do something.

I feel caught between a rock and a hard place. The site is ranking off old crud. If I remove the old crud I have every reason to believe it'll fall. If I add new links then I have every reason to believe the crud will be re-evaluated and it'll fall.

This is why I thought of the idea of leaving things as they were on the current site and building a completely different and ultimately far stronger and risk-free profile, then making a switch.

It appears I'm the only one that thinks that's the lesser of two evils!

Would be very grateful for any experiences of disavowals in stages.

[RedBar]

The site is ranking off old crud.

Are you referring to old crap no longer valid or evergreen quality?

This is why I thought of the idea of leaving things as they were on the current site and building a completely different and ultimately far stronger and risk-free profile, then making a switch.

Is there no reason not to keep the original site and create a completely new site and run them alongside each other with the obvious crosslinks?

I have done this extremely successfully with all SEs for at least 15 years. Heck, it's only the cost of a new domain name and you are creating a completely new site, but do not forget whichever way you do it that G's current ranking of new and even established but changed sites is running months behind.

[FranticFish]

old crap no longer valid or evergreen quality?

In Google's eyes, who knows? But I'm talking about links from a factory-SEO, one-size-fits-all approach. Obviously placed, not given.

I've slept on this, and have arrived at pretty much the same conclusion as you.

The site is a regional directory (county/area > town/city), and there are areas that currently under-perform on the current domain.

What I'm going to do is this:

1) Mirror the old site on a new domain, with every page noindexed
2) Add all the new content to the new domain
3) Pick an under-performing area, noindex it on the old domain, add our new and better pages and extra content, then let Google in to crawl it
4) Do PR, outreach etc to drive traffic and links to those inner pages of the new domain
5) Wait and see what happens

Hopefully, the new pages on the new domain will start pulling in referrals and also ranking. We can then work our way through the new domain gradually 'turning on' more areas, one at a time, and removing those same areas from being crawled and 'turning them off' on the old domain.

[Shai]

Sorry... still think you are making a mistake. :)

Just remember that you are introducing quite a lot of unknowns into the mix. Google does not always behave as you would logically expect and by doing what you are attempting to do, you are risking quite a lot of damage to the existing site in my opinion. Behaviour such as making virtual links between very similar sites/brands and even transferring penalties between them without a hardwired 301 redirect worry me when things like this are attempted. Although saying that, the only experience I have of this type of move is once when we split one brand into two sites (.com for outdoor furniture only while the .co.uk did indoor and outdoor) but we did not carry out the noindexing of pages. We just re-wrote the duplicate content. The results were not great. (feel free to PM me for urls)

With regards to the disavow, yes, we have quite a lot of experience in doing just that. In fact, we are starting on a similarly segmented disavow process as we speak. I think you will be surprised at how little your ranking will move. I really think that the vast majority of the type of links you mention hardly have any effect on the rankings. And with regards to the fear you have about messing with the hornets mess, just remember that all you are doing is turning the links to NOFOLLOW and that only happens once the links get re-crawled. So its still a gradual and due to the nature of the links, often slow process. Little worry about shaking things up there really.

Saying all that, your preferred method is more exiting so would be great if you keep us informed. Just remember to block using robots.txt from the outset rather than just Noindex as preferably, you want to stop the initial crawl of the site rather than just tell it not to index the site.

[FranticFish]

Thankyou - lots of food for thought there.

I think the best course of action is not making any decision now (we've got to do a lot of preparatory work before we start the content strategy as that will include a lot of video). I want to gather every bit of link data I can, then weigh up all the pros and cons of each course of action.

One further question: with your three-tier disavow strategy, can you give some numbers on the proportions/ percentages of links at each level?
i.e. YIKES! 20% / HMMM... 30% / FINE 50%

What would you consider low-risk proportions, and high-risk proportions?

[Shai]

Well, looking back at our last example, we only disavowed the very worst links in stage 1 (3 stages in total). These were spammy comments and anchor text heavy links. We were worried about the high proportion of high comp/volume anchor text links in the link profile of the site which placed it in Penguin territory. The % was quite small at 7%. I would not go over 20% for the first disavow if you are worried. Saying that, we sometimes do a 100% disavow and see no reduction in rankings so its just me being extra careful.

It really depends on how bad the actual links are. For example, if you have a large proportion of scraper type links and decide to disavow those, they chances of any ranking movement is minimal. However, if you have some high quality bought links then these can contribute to the site's rankings and therefore disavowing them will mean a drop in rankings. We find that disavowing comments, directories, forum and scraper links hardly ever have any measurable effect on rankings once disavowed. Removing blog network (ones that have not been caught yet), paid or any other do follow links from more 'potent' sites is more likely to have an effect on rankings.

[FranticFish]

Thanks again.