How to Solve Legacy Content Issues with a Partial Migration

Most websites grow in size and develop organically. Initially, the core of landing pages on a specific niche is augmented by a few information pages including the contact, about us, mission statement, job openings, etc.

As time passes, more supplemental landing pages (which do not cover the core business activity but are somehow needed) germinate.

Public relations landing pages address media and journalist inquiries. Shareholder interests are covered with investor information pages.

Much larger parts of a website are created when you add an archive, community forum, technical support, one-off event, merchandise channels, and the almost inevitable corporate blog.

Occasionally, a subdomain hosted staging server duplicates the entire website, multiplying SEO problems.

While not an exhaustive list, each of the above scenarios makes it more difficult for a website to perform well in Google Search. They frequently dilute content signals and cause ranking visibility loss over time.

However, these SEO issues are avoidable and easy to fix with a partial migration to a new top-level domain (nTLD).

Determining What to Move

The reason that content signals are relevant and important for SEO is owed to the fact that rankings are all about ratios.

For example, if a website consists of 80% user forum, which isn’t monitored and consequently continuously abused by spammers, the remaining 20% of the website (even if of superior quality) will have a hard time ranking.

That is because, for the most part, the website consists of low-quality or downright spam pages.

Of course, not every subdomain-hosted forum has to become a spam hub. Yet many do over time, as priorities and time allocations shift and once-promising initiatives are quietly abandoned.

A similar scenario can be applied to any other of the potentially out-of-control and rapidly growing parts of a website that generate either duplicate, low quality, lean content, or spam pages.

However, for great rankings, signal consistency is the key. The best method of dealing with no longer needed content by far is to delete it and return a custom 404 response.

Sometimes that isn’t an option due to politics specific to the prevailing culture within the organization. Even so, the growing impact of that outdated, off-topic, and lean content on rankings means the issue must be dealt with.

This is when a partial migration becomes the next best viable option.

Examples of Partial Migration Candidates

The problem wasn’t as simple to address before functional TLDs became as abundantly available as they are now.

With TLDs, your outdated, off-topic, lean, or otherwise undesirable content — hosted on a subdomain or a directory — can be safely moved to a new location.

Here are a few practical examples:

staging.example.com -> example.dev

An unprotected staging server hosted on the main site is always a dangerous proposition and therefore should be moved off the site, ideally to a .dev domain.

That domain can (but does not have to) be representative of the brand. The staging server’s contents, however, must be protected with a secure login at all times.

blog.example.com -> example.blog

A legacy blog that does not serve the purpose to improve users’ experience and is not an essential part of the website is potentially best moved to a .blog domain.

authors.example.com -> example.expert

Next to an undesirable blog, there may l0nger lean content author pages which, unless they demonstrate expertise and strengthen the website authority, can be moved to a new .author domain.

newsroom.example.com -> example.news
pr.example.com -> example.press

Large brand websites frequently offer media packages along with PR materials for journalistic purposes. Their media landing pages tend to grow in numbers as press release archives swell with legacy content.

This is why they are best moved to a branded .press domain.

community.example.com -> example.community
status.example.com -> example.report
wiki.example.com -> example.wiki

User-generated content poses a challenge in that it is most desirable from a marketing perspective and can tremendously boost relevance in a given niche.

At the same time, maintaining communities requires resources and is frequently abandoned at some point.

Communities “in the wild,” with no supervision or oversight, almost always attract abusive behavior and ultimately undesirable content submissions. Therefore, communities are best moved to a branded .community domain.

The same applies to community-edited resource pages, which are best moved to a branded .wiki domain. Past incident reports, which require preservation, are ideally moved to a .report domain.

support.example.com -> example.support
help.example.com -> example.help

Help and FAQ pages naturally grow over time, as online services expand and become more complex.

They often consist of a large and constantly increasing volume of individual issues, briefly outlined along with a short description of the solution.

These usually are textbook examples of lean content pages., which is why it is best to move help and or support pages off the main website to a branded .help or .support domain.

In this instance, considering moving content off the main domain is even more important, because frequently third-party CMS software is being used, offering the website operator no control over SEO signals.

shop.example.com -> example.shop
sale.example.com -> example.sale

For websites offering merchandise as a side business, branded .shop (or alternatively .sale) domains may be an attractive alternative for moving stock to another domain and avoiding the dilution of content signals.

events.example.com -> example.events
tickets.example.com -> example.tickets

For one-off and periodic events, branded .events or .tickets TLDs are likely attractive alternatives for moving content off the main website.

The examples highlighted are among the most notorious for large website operators. They are not exhaustive, though.

Depending on the specific niche, content quality, and volume ratios, there may be other (and equally, if not more, promising) areas for improvement in cases where obsolete content can’t be deleted but can be moved to other branded TLDs.

For the majority of brands, many functional TLDs are still available and can be secured at a low cost.

Unavailable TLDs do not pose a roadblock, since there are plenty of alternatives to consider and choose from; TLDs like .org, .site, .net, .io or .co are among such generic, popular alternatives.

When to Migrate Legacy Content

The decision to secure alternative branded TLDs and selectively migrate content should be taken solely on the grounds of a thorough audit and investigation.

It is only possible to gauge the impact supplemental content may have on the entire website after:

• Reviewing both Bing Webmaster Tools and Google Search Console data.
• Analyzing server logs covering a representative period of time.
• Crawling the website with a state-of-the-art crawler such as Botify, DeepCrawl or Screaming Frog.

In the course of the analysis, only substantial parts of the website that no longer serve their initially intended purpose but cannot be deleted should be considered for migration to an nTLD.

This is why a few pages of contact information are unlikely candidates for moving to a .contact or .chat domain (not to mention that users may be confused about who they are contacting).

Any migration, even one that appears inconsequential, is not without its risks.

When the process is finalized, the more relevant content tends to rank better, even though there is less content overall.

Why Move Content to a New TLD?

For a variety of reasons, deleting content often poses roadblocks.

Political considerations may play a role, as may a genuine fear of angering users.

When substantial volumes of landing pages (which have the potential to hold back search engine rankings of the entire site) cannot be deleted, partial migration becomes the next best option.

That step alone can go a long way towards ensuring compliance with Google’s Webmaster Guidelines.

While there are associated risks — especially if canonicals are not set correctly — there are advantages including a positive impact on crawl prioritization.

With substantially fewer landing pages, chances are that the entire website can be crawled more frequently.

While hosting and page performance play a decisive role, in general, fewer yet more relevant and better-optimized landing pages will rank better for competitive queries.

Having said that, partial content migrations by no means address all possible crawl budget challenges, such as indexable filtering or crawler traps large websites experience.

These serious, complicated SEO handicaps warrant separate consideration and resolution.

Another advantage of partial migration is improved SEO resource prioritization.

Moving user-generated content often unburdens the outreach team from repetitive monitoring and policing duties. It can also allow more time for promising outreach operations including brand recognition initiatives, which can push CTRs up.

Most importantly, the main website can be disassociated with any low-quality content or spam.

At the same time, any advantages stemming from perceived high-quality backlinks can be retained by linking from the new website to the main site, which is legitimate as far as Google is concerned.

As a general rule, cross-linking — even using highly commercial anchor texts and PageRank passing links — is sanctioned by leading search engines, as long as it remains obvious that there’s a legitimate connection between the sites.

Alternatives to Partial Content Migration

Large, non-generic brands with considerable resources at their disposal, following a long-term strategy, can adapt the migration strategy above after successfully applying for their own brand TLD.

You can participate in the application process during specific ICANN granted opening periods. Even so, there is no guarantee of the success of that application.

The process is relatively costly, especially in comparison to purchasing and maintaining even a large portfolio of functional TLDs.

For these reasons, relatively few organizations have chosen to take that step.

Fewer yet have succeeded in doing so and some have had to reconsider, rolling back and dropping their branded TLDs.

For the vast majority of website operators, a brand TLD isn’t a feasible option, because the process is too challenging.

For some, their brand name is too generic. For others, their very specific brand name is simply too long to be readable as a TLD.

And for a few, it’s a luxury option that, if implemented, allows for total control of the brand TLD including all of the associated domains.

This entails almost limitless possibilities to bring order and consistency to content signals. A branded TLD is an overkill option for all but a select few organizations.

Tips for Monitoring the Success of Your Partial Migration

Once a partial migration is successfully implemented, consider a few more important steps.

First, record and retain the new website’s raw web server logs. If not, there is no way to recover lost data when it is needed for analysis.

At the very least, the new website should be added and verified in Google Search Console and Bing Webmaster Tools.

More ambitious website operators want to collect as much free search engine data as possible and may consider adding their websites to Seznam Webmaster Console and Yandex Webmaster Console, as well.

There are others, too, such as Baidu Webmaster Tools and Naver Webmaster Tools. These latter two pose serious challenges because of their strict verification procedures and the language barrier for non-native Chinese speakers.

Lastly, it is imperative to ensure ongoing monitoring of both the main and the new websites for some time after any migration and to counteract swiftly in case of any unintended consequences.

While doing so, it is important to verify Googlebot and Bingbot first.

Keep Focusing on Improving Signal Input

Organic rankings in search are a consequence of signal input. Search engines and their algorithms continuously collect relevant on- and off-page signals, read them, and attempt to understand their relevancy and quality.

They often succeed and generate relevant SERPs as a result. Search, though, is complex and at time, those search engine algorithms fail.

In competitive niches, trusting that the algorithms will get it right is not a successful strategy. Instead of wishful thinking, put your best SEO foot forward by optimizing those relevant signals wherever possible.

Website operators are in control of their websites’ on-page signals. Exerting that control makes all the difference, especially for commercial websites.

The strategy outlined above can bring rewards including on-topic, consistent content signals, fewer lean content landing pages, fewer policy violation risks, and fewer landing pages competing for both search engine crawl budget and user attention in the SERPs.

More Resources: