How to analyze the performance of multi-domain SEO programs

Is having multiple domains helpful or harmful to your SEO?

Let’s go to find selected clips! The short answer is that having multiple domains will definitely cause problems for SEO performance due to cannibalization.

However, this need not be the case.

There are many strategies for how multiple domains complement each other for SEO, and I will point out some of them in the next process.

But for this article, let us focus on the problem (because let us face it, many people may read this article because you suspect that there is a problem with your multi-site SEO strategy!).

In this article, you will learn about the following:

• Why companies may use multiple domains for SEO.
• Common SEO issues for businesses with multiple domains.
• How to analyze the SEO performance of multiple domains.
• Key natural traffic patterns/issues to be aware of.
• Ideas on how to solve the homogenization problem caused by having multiple websites based on similar topics.

Why companies might use multiple domains for SEO

There are many valid reasons for companies to have multiple websites. Common reasons include:

Companies with multiple brands.

For example, SC Johnson owns Windex, Pledge, and many other brands, all of which deserve their own domain names. This is usually a good setting when each brand communicates with a different audience.

For businesses in multiple countries/regions.

For example, OLX operates similar classified advertising businesses in Portugal, India, and Brazil, and each business has its own ccTLD.Have an SEO advantage Country specific domain And the brand advantage.

Companies that grow through acquisitions.

When you buy a company, the website usually comes with it. This can provide opportunities for growth as it can expand your reach or SEO challenges that may come from cannibalization and/or confusion from search engines.

Multi-location business.

A national brand owns Separate website for each physical location Or territory. This usually leads to SERP competition between these local sites and national brand sites.

Companies that want to dominate SERPs.

Whether from acquisitions or local websites, many companies try to make multiple websites they control appear in the same search query.

As we will see, this can work well-until it doesn’t work.

Common SEO issues for businesses with multiple domains

For companies with multiple websites, we often see two main challenges in SEO:

Supporting multiple sites is a royal PITA and can be expensive.

Each site requires front-end and back-end work, continuous maintenance, design and UX resources, and then invest in SEO (content, backlink acquisition, new features, etc.).

With the increase in investment, it is clear that the demand for ROI and the demand for the employees who support it is also increasing.

We often recommend that customers who ask about a multi-domain strategy have much less effort/investment in getting a single domain ranking than getting multiple domains to rank well in a similar number of search queries.

So if you are going to take a multi-stop route, please be realistic Investment level And the effort required to make SEO work.

Cannibalize.

Usually, when we work for companies with multiple websites, the organic traffic and revenue of these websites are declining, and it is clear that these websites are having a negative impact on each other’s SEO performance.

There are many reasons for this, but the most common situation we see is that the sites have the same data/content. Sooner or later, this kind of thing seems to suffocate Google.

How to analyze the SEO performance of multiple domains

1. Aggregate your data

In order to analyze the SEO performance of multiple domains, it is helpful to be able to view all the data in one place first.In order to do this, you may need to build a data warehouse to ingest Google Search Console Provide data for each domain through the GSC API so that you can view aggregated data.

We do this through various Google Data Studio reports, such as this one, which shows the organic traffic of each domain over time.

This is an example of the six domains we looked at, where the sites are all directories to find specific types of local businesses, where there is a lot of overlap between the lists for each domain.

2. Segment domains by page type

Just looking at the overall domain can give you an idea of ​​the problem, but page-level analysis is very important to solve this problem.

Author screenshot, November 2021.

At this point, you need to determine the types of pages in each domain and divide them into multiple segments so that we can see which segments are underperforming.

3. Build a robust keyword set

After setting up the page/site data, you will need to be able to view these domains at the keyword level. The number of keywords depends on the size of the domain and the size of the target market.

For the recent 6-domain project for overlapping search queries in major cities in the United States, we used a 2 million-word keyword corpus.

This allows us to understand the performance of the localized pages of these sites in all of their markets, while leaving a large number of other keywords to determine the performance of the editorial page for “country” queries.

In order to export such a large keyword list, we usually start with a seed list.

In this case, we exported all non-branded keywords for each domain from Google Search Console through GSC API, and then applied them to various popular local queries (e.g. city + keywords, keywords in cities, near me Keywords-browser settings to user location, etc.).

If you cannot get enough words from GSC, then you need to have a process to expand your keyword corpus. You can use various ready-made tools (such as SEMRush or Keywordtool.io) to do this.

However, if you want to do a lot of this type of analysis, ideally you need to use your own Keyword expansion method Because licensing data from these tools can become expensive, and you will be limited by the tool’s perspective and your own strategy.

4. Categorize keywords by subject

Just like we grouped page types in each domain into segments, once you have a keyword list, you will want Group these terms into topics.

When you process any set of search queries on a large scale, the performance of a single keyword is usually not as important/actionable as the performance of your page for a specific topic in each domain.

For example, a page about “multi-domain SEO” can rank hundreds of different queries related to the topic.

Using a large keyword set for this type of classification can be a challenge. As with developing an initial keyword list, off-the-shelf SEO tools can help you achieve this goal.

An easy way is to run keywords through the n-gram tool and observe which phrases appear most frequently. However, when the subject is not obvious from the keywords, this will not help.

For example, just by looking at the text, you would not know that the Ford F-150 is a truck. This is where I strongly recommend that you invest in a machine learning method that can help you classify relevant keywords, but this is another article.

How to analyze SEO data for multiple domains

Now that you have aggregated the data into one view, your page types have been subdivided, and your huge keyword corpus has been categorized, you can begin to study how these domains interact with each other over time and how they interact with Google.

Here are some of the different types of analysis you can perform:

The observation domain serves as a ranking model for a group.

For example, in this case, we see that most of these domains are ranked in the bottom half of the first page or the top half of the second page. No strong domain name can enter the top-level results.

This is a sign that there may be a problem with the entire set of domains, not just one or two domains.

Author screenshot, November 2021.

Cross-domain analysis of the performance of different page types.

The main page types for these domains are city pages, location details pages (for example, a list of specific points of interest in the city), and edit pages. In this case, we see a distinctly different pattern for each page type.

City pages are declining across all domains, but one domain (the largest domain in the group in the long run) has most of the organic traffic that leads to queries on these pages.

The location details pages for all domains are also declining, but none of them dominates.

In the case of queries that lead to these pages, the largest sites have almost no content on their pages, while the smaller domains do have content on these pages.

This type of pattern sends a strong signal that smaller domains may cannibalize larger domains. If the larger domain has content, it may dominate these queries.

The edit page is probably the most interesting.

Among the six domains, one of the smaller sites is the only one with User-generated content (Comments and Q&A). This is the only domain that performs fairly well in “country” queries, which are usually non-location-specific information queries.

When we looked at the top SEO competitors and local data aggregators for these domains, none of them performed particularly well in these types of searches.

In these cases, authoritative content sites that focus on the topic tend to rank best. This indicates that customers may need to consider starting a domain that is primarily editorial in nature.

How to solve cannibalism

In the end, the recommendations for this specific set of domains boil down to the following points:

• For themes where one domain dominates, consider removing or de-emphasizing competing content from other domains, and double the theme investment in the winning domain.
• Use cross-links between domains to signal to Google that you want which domain to rank higher than others on a particular topic.
• In the case of possible cannibalization, consider merging the weaker domain into the stronger domain.
• Monitor the performance of all domains in a single report. This will help you spot trends in all domains as early as possible and help you understand how the SEO performance of one domain affects other domains.

If you want to get the best SEO performance from a multi-domain setup, it’s important to create a process that effectively analyzes how these domains interact with Google.

More resources:

Featured image: Shutterstock/nmedia