Searching for the Best Tool for Backlink Analysis Part 2: Evaluation

After the overview of the backlink tools we tested in the first part, the most important criteria that a tool for backlink analysis should meet are examined here. Which evaluations are essential, and what additional features are there?

Requirements for a backlink tool

While some features are a nice addition, the following points are essential for a meaningful backlink analysis. The comparison clearly shows that not all tools meet these minimum requirements. For all the tools examined, backlinks for any domain can be analyzed. SEOlytics generally allows the evaluation of any domain, but filtering and certain information are only available for a limited number of registered domains.

The number of backlinks – a quantitative difference

In general, it can be said that no tool can provide a complete list of all backlinks. Our tests have shown that practically every tool finds links missing from other evaluations. The quantitative number of backlinks can only provide a benchmark value, a scale particularly useful in direct comparison with other domains. This table shows how much the numbers of individual tools differ:

So what can be concluded from the different numbers? For English-speaking tools, the difference is easily explained, as they do not focus on German domains, from which a German website generally obtains most of its backlinks. The higher values of the German tools initially indicate a larger data basis compared to their competitors. However, a more nuanced view is worthwhile here: the currency of the links is also important here. In our comparison, we found backlinks from last year that were obviously not checked recently. Many of them can no longer be found. Here, the question arises: What does a backlink analysis say at all that provides outdated data and thus incorrect numbers?

Data currency

Some tools, like Xovi, Raven Tools, SEOlytics, and MajesticSEO, provide a date in their analyses for each link found. Here, at least, old and new links can be identified. However, it would be more helpful at this point to indicate when each link was last checked. This becomes problematic, especially when these time indications are missing, as is the case, for example, with SEOmoz or Searchmetrics Essentials. Users, therefore, receive no indication of the currency of the data. A mere promise that a powerful crawler is currently checking data and determining new backlinks does little good when outdated backlinks are actually found in evaluations.

Analysis of the domain, a specific page or a subdomain

By default, the entire domain is usually analyzed, but sometimes the backlink situation for a subdomain or a single URL should also be examined. With the exception of Raven Tools, all analyzed tools enable such filtering.

Backlinks by domains

The mere listing of a multitude of backlinks is often not very meaningful. As is well-known, 10,000 links from the same domain are of comparatively little help, while the same number of links from different domains is a pretty good result for many domains. All tools offer grouping by domains, but Sitefactor only shows the top domains. Raven Tools does not list the linking domains as a menu item; here, they can only be grouped by domains in the backlink table.

SEOlytics also does not offer domain analysis in the menu; however, backlinks can be filtered by domains for registered domains.

Backlinks by IP addresses

The number of different domains as a backlink source also shows little meaningfulness, which is why a look at the different IP addresses from which the backlinks originate is worthwhile. Domain networks can thus be quickly identified. Sistrix, Searchmetrics Essentials, and Xovi offer this evaluation, while MajesticSEO at least allows domains to be sorted by IP address, but the links are not summarized by IP addresses. Sitefactor only outputs the top IP addresses, here only the number of backlinks from the top 5,000 links that the tool displays. SEOmoz, SEOlytics, and Raven Tools lack IP address evaluation.

Backlinks by countries

Anyone who has purchased or had cheap ru-links purchased should be able to identify this in the backlink analysis. Such practices can also be quickly uncovered for the competition if the tool used offers this analysis. Sistrix and Searchmetrics Essentials provide an overview of backlinks by countries, but this analysis is missing from MajesticSEO, Xovi, Raven Tools, and SEOmoz. SEOlytics does show the respective country for individual backlinks and allows filtering by countries for registered domains, but an explicit evaluation by countries is not available.

Analyze link texts

Far more important in SEO work is the analysis of link texts. Is only the brand linked, or are there also keyword-optimized link texts? Sistrix, SEOlytics, SEOmoz, and Searchmetrics Essentials enable the analysis of link texts; at Xovi, this evaluation can only be obtained through a comprehensive export. Sitefactor shows the top link texts. Raven Tools does allow sorting by link texts, but grouping is not possible. Finally, MajesticSEO lacks any evaluation of link texts.

Type of backlink: nofollow-links and image links

It is widely known that not only the number of backlinks and the link source are important but also the type of link. Accordingly, all analyzed tools provide information on whether it is a text link, an image link, a follow or nofollow link for each backlink. However, this information is only found in the export table at Xovi. SEOmoz also lists redirects in the backlink table. MajesticSEO lists the number of image links, nofollow links, redirects, deleted links, and frames in the overview. Sistrix goes a step further in the export and shows canonical and meta-refresh alongside the redirections, image, and nofollow links.

Placement of the backlink

An indication that is certainly difficult to carry out in practice and is therefore not provided by any of the analyzed tools, is the evaluation of the link placement: is it a footer link, a content link, or is the backlink found in the sidebar? This analysis is likely not only complex but also prone to errors, so for now, this evaluation remains a pipe dream.

Identify most-linked pages

Another standard evaluation is the analysis of the linked pages. How often is the homepage linked, and which subpages have the most backlinks? With the exception of Raven Tools, all tools evaluate the linked pages. Sitefactor at least shows the top pages.

Additional comparison criteria

In addition to the above points, which should be a minimum requirement for a professional backlink tool (after all, all the solutions examined cost money!), there are some points that would be desirable and by which the various tools were also examined.

Analyze historical data

Generally, one wants to check the current status of the backlinks; however, retrieving historical data can also provide insights into which backlinks may have been lost and how the number of backlinks has generally developed. Out of the selection of backlink tools, only MajesticSEO and Sistrix show historical backlink data. According to their own statements, the data goes back several years and thus provides a comprehensive picture – whose completeness, of course, cannot be guaranteed any more than it can with current figures. SEOlytics allows at least the last three months to be analyzed.

Extent of the database

As mentioned above, the mere number of domains and URLs in the database is not yet a quality criterion; however, it stands to reason that a larger database potentially provides more meaningful and comprehensive results. The comparison between the individual tools is hardly satisfactory here, as not all providers provide an insight into their database. When considering the mere number of URLs and domains in the index, SEOmoz is clearly ahead; however, this says little about the number of backlinks found for a German domain, as shown in the comparison above: SEOmoz finds significantly fewer backlinks for spiegel.de than all the examined German competitors. Searchmetrics Essentials boasts an impressive database among the German tools, but a comparable figure for the other tools is lacking, so no assessment can be made here.

Sources of backlinks

Since the Yahoo! Site Explorer has stopped its service, the sources that the various tools use for their data have gained particular importance. However, the comparability is lacking here as well, as the sources are often not apparent without calling or writing to support. Sistrix, Searchmetrics, SEOmoz, and Majestic use their own crawlers, while Raven Tools claims to use data from MajesticSEO. The significant differences between the backlinks found by Raven Tools and MajesticSEO are quite incomprehensible in light of this claim.

Export options: XLS, CSV, and PDF

Analyzing backlinks online and filtering them by various criteria is a good and helpful thing. However, for evaluating and further processing a large number of backlinks, it is essential to be able to export the results of the analysis. The standard here is the CSV format. In itself, this is sensible, unfortunately, the exports of many tools are so buggy that the exported tables cannot be processed flawlessly. Certain characters, such as quotation marks and commas, cause problems and also lead to the tables not being converted into evaluable rows and columns without errors. This was the case, for example, with SEOmoz, RavenTools, and Sitefactor. Some tools, like SEOlytics, offer other formats, such as XLS, PDF, or HTML, beyond the CSV files. At Xovi, the export is limited to 5,000 records, SEOlytics provides a maximum of 2,000 records in the download.

Searchmetrics Essentials does not allow a tabular export of backlinks in the free beta version; only a short overview can be downloaded as a PDF. The Sistrix file is stored as a gz file, which must first be unpacked. For a domain with a large number of backlinks, the resulting unpacked CSV file can temporarily even bring a powerful computer to its knees. Nevertheless, the exported file shows the backlinks in correct columns.

Compare backlinks of different domains

The direct comparison of several domains via just the backlink exports proves to be rather uncomfortable. Therefore, some tools offer the possibility to compare domains directly. With Sitefactor and SEOmoz, up to 5 domains can be compared. Additionally, MajesticSEO offers the Clique Hunter, which allows finding domains that link to several of the analyzed domains.

SPS, OVI, ACRank, Majestic Million and PageAuthority?

In SEO tools, it almost seems to be a good tone: (Almost) every tool comes up with its own values, obscured by mysterious abbreviations. If you're lucky, at least the meaning of these homemade values is explained; sometimes, an explanation is even found that indicates how the displayed numbers are derived. At MajesticSEO, one reads of the MajesticMillion, at Sistrix, a visibility index is calculated for each domain (which is not found in the backlink module), Searchmetrics invents the SPS or Searchmetrics Page Strength, at Xovi there is the OVI, Raven Tools offers the ACRank, SEOlytics the SEOlytics Visibility Rank (SVR), SEOmoz the MozRank and MozTrust.

Only a professional at most can see through this variety of values (of which only a small selection has been listed here as examples), but even they will quickly become skeptical due to the opaque algorithms that lead to these values. In most cases, these values do not come up with a reference range; whether a value of, for example, 0.5 or 6,666 is good or bad can only be guessed at best in direct comparison with a meaningful control group. So do these values offer real added value? This can be doubted in many cases. If the user cannot evaluate these values and cannot comprehend the factors that lead to a particular value, he can do little more with them than to marvel at the extravagant evaluations the professionals have come up with. Decisions should not be made on this basis.

Costs and further features

The costs of the individual tools were already mentioned in the first part. It should be noted here that the tools differ so significantly in their scope and modules that a comparison is difficult. While SEOmoz charges $99 for all modules and tools and the smallest package, you get just one of the comprehensive modules from Sistrix for €100. A direct comparison cannot take place here: cheaper tools generally offer less scope. It can generally be said that unused features of a more expensive tool are just as much wasted money as a cheap tool that does not provide meaningful data.

The individual tools offer a multitude of further features that go beyond the general requirements. Here are a few functions to illustrate:

• MajesticSEO and SEOlytics show the number of links from educational and governmental domains; information not found in any other tool, though its usefulness can be debated.

  • Searchmetrics Essentials groups by news, forum, blog, directory, shop, and general and clusters the sectors from which the links come. How meaningful this is and whether all links can be correctly assigned is questionable. According to their own statement, the assignment is made based on rankings. How accurate this is depends on whether the respective keywords can be correctly assigned and whether the linking domains rank in their market segment.

  • Xovi, Sitefactor, and Searchmetrics show the page titles of the linking page for the found backlinks.

  • Searchmetrics and SEOmoz also provide the number of social media links (Facebook Likes and Shares, Tweets, and Google +1), an interesting size in the year 2012.

  • Another plus that must be highlighted and can significantly ease evaluation: In particular, Sistrix, SEOlytics, and SEOmoz offer extensive filtering options for the found backlinks.

Alternatives? Analyze backlinks for free!

Since each of the presented tools obviously has its strengths and weaknesses, the question arises about cheaper alternatives. For personal use and one's own domains, the backlink analysis in Google Webmastertools offers a selection of backlinks. For a quick overview of the most important link sources, linked pages, and link texts, this free evaluation is often sufficient. In addition, there are, of course, numerous free backlink tools, often financed through excessive advertising and usually with a comparatively small data base. However, for a one-time analysis without the claim to completeness, many of these tools can provide a starting point.

Conclusion: bugs, features, confusion, and disappointmentsAll the analyzed backlink tools are undoubtedly suitable for supporting link building: The backlinks found from competitors can be a valuable inspiration for building links for one's own domain. It is important to be aware that only a snippet is ever displayed, not a complete result.

The actual purpose of a backlink tool – to offer evaluation capabilities for one's own backlinks, identify broken links, and gain an overview of linked pages and link texts – is fulfilled by the presented tools to varying extents. It is clear: The perfect tool that fulfills all wishes and requirements and makes the search engine optimizer happy with its analysis does not yet exist. Whether one is therefore willing to pay a certain amount monthly for a tool that provides unsatisfactory backlink results or whether, for many purposes, the free evaluations in Google Webmastertools are sufficient is left to be seen.

The fact is that although large numbers, powerful databases, and adventurous comparative values are advertised, the weaknesses are not discussed. This is especially annoying when serious economic decisions are made based on the data from professional analysis tools, in the firm belief in the meaningfulness of various internal agency algorithms and values. A little more focus on the essential functions and data instead of ever new mysterious homegrown values would be desirable.