Google has just announced the future release of Google Knowledge Graph less than 24 hours ago as they have posted about it on the Google Blog. I immediately did a search query and definitely has not yet seen the knowledge graph implemented on Google from my view as how they presented it in the introduction video, but it would probably propagate soon.
By simply looking at the video alone, this can already tell you what opportunities you can maximize and take advantage of in using Google Knowledge Graph. Below are the 3 specific opportunity tactics that are existing good SEO practices even before the Knowledge Graph existed. But when Google Knowledge Graph rolls out, further emphasis on these existing practices can be part of your future everyday SEO strategy:
- Always think of LSI in Keyword Research and Content Optimization
- Stretching the Relevance for Easier Link Building
- Selling Higher in the Conversion Funnel and Tracking Analytics Attribution
Let’s look into these into more detail below:
1. Always consider LSI: Make Sure Google Understands The Context of Your Words
LSI or Latent Semantic Indexing as defined on Wikipedia is as follow:
Latent Semantic Indexing (LSI) is an indexing and retrieval method that uses a mathematical technique called Singular value decomposition (SVD) to identify patterns in the relationships between the terms and concepts contained in an unstructured collection of text. LSI is based on the principle that words that are used in the same contexts tend to have similar meanings. A key feature of LSI is its ability to extract the conceptual content of a body of text by establishing associations between those terms that occur in similar contexts.
Ok, so this is geek speak, what does this really mean? Meanings of words actually change depending on the relationship with other words. It’s context changes depending on how you use it. From a computer algorithm point of view, the context changes depending on the other words in close proximity to other words within the same html page. Extending to a more offpage SEO point of view, you can use the same concept at looking at the relevancy of pages linking to each other and the sets of words that make up the context. This was not some overnight phenomenon that Google just came up with, they have been playing around with LSI in the early 2000, that they decided to buy one of the leading companies in 2003, Applied Semantics.
Now I am going to steal a few examples from the slides I saw in a presentation of Kim Tyrone Agapito, one of the presentations I like at the Iloilo SEO Conference:
- Word: date
- I want to go on a date with you
- What’s the date today?
- Word: notebook
- The kid turned her notebook into a coloring book.
- My notebook runs a Intel Core i7 MSI
- Word: types
- Website types and the primary keywords you’re going to target are inter-related.
- The guy types fast because his paper is due today.
- Word: apocalypse
- marvel, apocalypse, thanos, x-men, avengers, iron man, fantastic four
- walking dead, apocalypse, zombie, left 4 dead, resident evil
From an SEO tactic standpoint, if I would try to boil this down into simple steps. I would say: Consider LSI when doing keyword research and writing content. Pay attention to related keywords, similar keywords and synonyms and do not even stare at search volume and amount of competition and KEI that much. As long as one of the keywords have high search volume and the synonyms, similar and related words do not, does not mean you should not target them.
Considering how Google Knowledge Graph works, Google should get a good idea of the main context of the page so that it properly associates it with the targeted related content. The more related words you use in the right context, the more I would assume Google would know how to associated to relevant content that can drive you additional traffic from Google Graph.
2. More Traffic Opportunities by Stretching the Relevance
SEO professionals in the past always concentrated on keywords that give traffic and keywords that don’t. Looking at keywords with high search volume are more important and keywords that are directly related to the product by being more specific. Long tail keywords often lead to the better conversions. But the problem is from a link building point of view, if you have a small niche product that only attracts a small amount of people, often it is more difficult to get many links naturally.
Using some creative writing, awesome images or making viral videos that are related to the product or service for link bait purposes would usually do a better job at getting these links. One thing I believe some SEOs have to get over is always thinking that whatever the page that gets link naturally should always have the keyword of the product or service you are targeting. Well I believe one mention is good enough, a link going to the product or service page in not always required in a blog post especially if the link bait is hosted on the site selling the product or service. The natural navigation of the site should be able to handle that and avoid any perception of over optimization. Unless the content is really talking about the product and service and does not appear as it was added in by force. So the key here is looking at opportunities to “stretch the relevance” (a term I first heard from Todd Malicoat, which I am currently borrowing the phrase. ) If you can come up with a relevant piece of content that has some link bait properties and can still tie it into the product or service, then do it for link building purposes to gain your links naturally. This is where basic marketing skills can come into play in the SEO game. These types of links you gain may not be targeting a specific keyword that is related to the product or service of the company you are promoting, but just because of the natural links gained through this method it would help in the over all trust and authority of the site that helps in the ranking of all keywords of the site overall.
From the Google Knowledge Graph point of view, stretching the relevance can help push your site into more topics that can possibly relate to your product and service, again gaining more opportunities to get organic search traffic from Google Knowledge Graph.
3. Selling Higher in the Conversion Funnel and Tracking Analytics Attribution Accordingly
Now when you stretch the relevance, and concentrate on LSI, you will be gaining more traffic that is not 100% related to your product or service, but still somewhat related. People that may be landing on your site may not be in the buying mode since they are less specific of what they want to search. The traffic you may gain may not necessarily give you immediate conversions. But with the continuous exposure, it is helping influence visitors to consider buying your product or service one day, when the time does arise when they need it.
The critical thing here is how will you monitor this? And this is where attribution analytics comes into play. When you do get a conversion, and you are tracking your traffic properly with the right attribution metrics, you should be able to see how these related searches may be the initial entry points to your site and you will also see that this traffic did not immediately convert into a sale or lead. It may take some time of repeated visits in various ways, redirect hits, referring traffic, other search engines, social media, organic search or pay per click search. In the process you should also be able to see how the keywords may be changing over time and eventually being more specific near the end of the conversion funnel.
Thinking of LSI in keyword research and content writing and stretching the relevance of topics give you false notions of the value of organic search traffic without proper attribution tracking. Usually related stories may not sell a product or service right away, but may have also forever influence someone that may be part of longer buying cycle. Although Google Analytics has some form of Attribution tracking, it only saves data for 30 days. Longer buying cycles and deep conversion funnels cannot be tracked properly in the current attribution tracking features of Google Analytics. Not really knowing if your optimization efforts for Google Knowledge Graph may be beneficial or not..
To summarize everything above, when looking into optimizing sites for Google Knowledge Graph in some simple steps:
- Always consider synonyms, similar words and related words in keyword research and content writing. (What I am not saying here is stuffing content with a bunch of words. Google can detect stuffed words with LSI technology. Especially if the group of words are normally not be grouped together on a single sentence, a single paragraph, or even a single webpage.)
- Come up with good content for the words. Remember it is not keyword stuffing. Other words with context to what a word is about and Google understands this with it’s LSI technology.
- Stretch the relevance, not only would it be beneficial for Google Knowledge Graph, it makes content also more compelling for gaining natural links.
- To track success properly, consider looking into various attribution analytics vendors, although usually these do not come cheap, or free. So probably this is more for company websites than personal blogs or small companies.