When apply­ing net­work analy­sis to the field of SEO, the objects are web­sites and the rela­tion­ships between them are links. By graph­ing these rela­tion­ships, insights can be gained about how a web­site is linked to and exists on the web.

There’s a num­ber of ways that net­work analy­sis can be applied to SEO:

• Visu­al­iz­ing inter­nal links on a website
• Visu­al­iz­ing inter­link­ing between var­i­ous domains
• Visu­al­iz­ing back­links to a domain or domains
• Com­pet­i­tive analy­sis of backlinks

As an exam­ple of how net­work analy­sis can be used to find back­links, below is a net­work graph show­ing how six dif­fer­ent web­sites are linked to. The dots rep­re­sent web­sites and the lines rep­re­sent links. Our client’s site is high­lighted in blue, while five of their major com­peti­tors’ web­sites are high­lighted in red. Exter­nal sites that link to either our client or their com­peti­tors are rep­re­sented by black dots.



When apply­ing net­work analy­sis to the field of SEO, the objects are web­sites and the rela­tion­ships between them are links. By graph­ing these rela­tion­ships, insights can be gained about how a web­site is linked to and exists on the web.

As you can see in the graph, each of the six core web­sites has their own par­tic­u­lar back­link pro­file – sites that link only to them. Some links are shared between the dif­fer­ent sites, and these lie at the cen­ter of the graph.

What we’ve high­lighted in this next net­work graph are web­sites that link to at least two of our client’s com­peti­tors, but who do not link to our client.



These web­sites are prime link­ing tar­gets. As they link to at least two com­peti­tors, they demon­strate that they have an inter­est in that par­tic­u­lar indus­try and would be likely have an inter­est in our client’s website.

Net­work analy­sis can be applied broadly to the field of SEO in ana­lyz­ing how web­sites link to each other, and in con­duct­ing com­pet­i­tive analy­sis. The field also has obvi­ous appli­ca­tions for social media in under­stand­ing how users are con­nected to each other in online net­works. Any online social net­work – whether it be on Twit­ter, Face­book, Youtube, Flickr, etc. – can be eas­ily mod­eled using a net­work graph. A graph could focus on how one user is linked to other users, which is called an ego net­work. Or a social net­work graph can focus on a par­tic­u­lar topic – com­ments about a Youtube video, a Twit­ter hash­tag, or a Face­book group – and look at how that group of users is inter­con­nected. The search engines have stated that the social web has a small influ­ence on rank­ings so there is also an SEO impli­ca­tion for this type of analysis.

As an exam­ple of social net­work analy­sis, here is a net­work graph of the Twit­ter net­work of 4FRNT skis, a ski man­u­fac­turer. This graph has been ana­lyzed to 1.5 degrees, mean­ing that we’ve not just looked at who fol­lows 4FRNT on Twit­ter, but how those peo­ple are con­nected to each other as well.



While 4FRNT has about 1,200 fol­low­ers, those fol­low­ers are highly inter­con­nected with approx­i­mately 30,000 con­nec­tions between them. When you have a dataset this com­plex it can be hard to make sense of what is going on. For 4FRNT, what we’ve done is break down their net­work into clus­ters – groups of highly con­nected Twit­ter users – and dis­cov­er­ing who is cen­tral in those clusters.

Dots in orange are a group of users focused around ESPN Skis. Dots in green are users focused around Armada Skis, another ski man­u­fac­turer serv­ing a sim­i­lar niche mar­ket as 4FRNT. The dots in pink focus around a Twit­ter user called The_Canyons, which is a ski resort in Utah (based in the same loca­tion as 4FRNT). The dots in blue focus mainly around 4FRNT itself. Instead of just view­ing the graph as a mass of dots and lines, we’ve been able to get some sense about who on the 4FRNT net­work of users is serv­ing as sub-areas of attention.

If you’re look­ing to do your own net­work analy­sis, there are a num­ber of soft­ware pro­grams you can use. Sev­eral of the soft­ware pro­grams require knowl­edge of pro­gram­ming lan­guages such as Python or the sta­tis­ti­cal pro­gram­ming lan­guage R. An easy to use, free and open source plu­gin for Excel is NodeXL. If you are look­ing for some­thing more advanced that does not require pro­gram­ming skills, UCINet is a nice option.

- NodeXL Free Down­load and Doc­u­men­ta­tion

- UCINet Free 60 Day Trial and Doc­u­men­ta­tion

Net­work analy­sis is in use in areas includ­ing the sci­ences, social sci­ences, and busi­ness. Here are a few TED talks show­ing net­work analy­sis in action in other fields:

Eric Berlow: How Com­plex­ity Leads to Simplicity




Nicholas Chris­takis: The Hid­den Influ­ence of Social Networks




Here is my pre­sen­ta­tion from SMX West

Net­work Analy­sis for SEO and Social Media


Biography / Resume : Charlotte is a key member of Mediative’s service delivery team, focusing specifically on organic search marketing. With an educational background in commerce, cultural studies and IT (as well as receiving SEO-specific training from Search Engine Academy West), Charlotte brings a unique, specialized portfolio of experience to the table. Her organizational and analytical skills, coupled with her commitment to learning, ensure she consistently helps the client achieve success. Charlotte thrives on the challenges and opportunities that come from working in a continually evolving field.