Market Segmentation and Analysis.
One of the Crown Jewels in the Krakken Vertical Online Market Analysis (VOMA) application is its ability to segment and analyze markets. This demonstration will show you exactly how this is done and the power behind it.
Let's start with a definition of a cluster: A cluster is a collection of themes. A cluster becomes interesting when the themes in a cluster represent a specific market segment. At that point the "cross cluster analysis" that is available on the "Themes Overview" screen can give you some very interesting information.
Let's take a closer look at why this is so.
To segment a market you will want to look at a specific aspect of that market, by itself. Generally, the smaller the niche of the market in which you are looking, the fewer meaningful ways there are to segment the market.
Examples of interesting market segmentation include degree of buying intent; educational or informational keywords, product class and geo location.
When drilling into themes in a cluster, you will find some terms are significantly broader or narrower than the norm (as defined by the numbers of competing pages). Often times, these terms actually represent a different segment of your market - either a larger market segment than the other terms in your cluster represent, or a smaller niche, respectively. At other times, you'll find terms where the cost per click is significantly higher or lower than the other terms. These terms also often represent a different market segment. In fact, keeping an eye on these values will allow you to find other segments that might not have occurred to you during your initial analysis.
When you've sifted the wheat from the chaff, and found a cluster of themes that create a meaningful segment in your market, you'll likely have the makings of a great silo for your website.
Let's look at a specific example to help make this clearer.
In this example I'm going to focus my market segment on a combination of product and buying intent, by keeping my focus on "beer kits" and limiting the range of cost per click.
Above you see some research I did for a website on "Home Brewing". I wanted to promote some affiliate beer products. I started by asking myself 'what are the different components of home brewing?' Answer: there are "Beer Kits" and "Home Brewing Equipment". So this lends itself nicely to two clusters (i.e. market segments) already.
As I drilled into "beer kits" I found different ways of expressing that idea, and those might end up being good synonyms for each other. I drilled into a number of them and found they all had about the same cost per click and were roughly in the same class of competing pages.
My next step was to evaluate which one of these terms I wanted to be the key term (parent term) in my theme (silo landing page). As you can see above, 'homebrewing' is a bit broader term (more numbers of competing pages) then my other themes and not as valuable (lower cost per click). Home brewing, as you'll recall was actually my "website market", a market that also includes at least one other silo (home brewing equipment). However, because the numbers of competing pages were so small for this niche, I had wanted to include a broader term. I was thinking that this would make a good "KGO" term (Keyword Generation Only). This means I won't actually try to rank for this term in my beer kit silo, but I'll use it to generate other good keywords for the silo. The numbers above do indeed indicate that it should be a KGO term and not actually a synonym. But more on that in a minute.
Of the terms left, "beer kits" is the most competitive (the white paper validated this information) and I decided that it would be my parent theme. I wanted all the rest of the terms to be synonyms, all the way down to "beer mixes".
Now let's take a look at the competition. In the above pie chart what you are looking at is the result of all of my themes put together. In this case, I only have one keyword in each theme (the parent terms listed above). This pie chart shows you the total 'Organic Market Value', and the 'Market Share' of the top domains.
Here you'll see that homebrewers.com has got the largest percentage of the marketshare. But what does this really mean?
This pie chart analyzes domains by their ranking, position and cost for each keyword.
At the risk of getting too technical here, the keyword cost is multiplied by the traffic and that value is multiplied by a number that represents how much of that traffic a site will get based on the ranking position. This is a site's Organic Rank Value, or ORV
All of a sites' ORV ranking values for this cluster are tallied up and displayed at the top of the pie chart. The top ORV domains are listed individually in the pie chart along with their pie slice indicating their percentage of the total ORV.
I can see at a glance who is dominating across this cluster. This is why you want to keep an eye on keywords with significantly higher or lower costs that might actually belong in another segment (cluster). The results of this pie chart can be skewed, especially by highly trafficked or highly valued themes.
Now for the kicker. We have installed a setting that will switch you between looking at the values for just the "seed terms" and "interesting theme terms" ! As you can see in the graphic below, you can toggle between "synonymic set" or seed terms, and "vertical market" or interesting theme terms.
Until now we have been looking at the results for just the seed terms for each of the themes listed above. But in VOMA you can drill into these terms and each theme will bring back a selection of related terms.
These terms are not normal "keyword tool" keywords. The are specifically selected for Vertical Market Research. This means that we are going to show you the best of the best. These terms come in three different classes.
We use a genealogical expression as a metaphor for comparing the relationships of these three classes. First are the verticals of your seed term (these are called the ancestors). These terms have numbers of competing pages that are an entire magnitude (or generation) larger than your parent theme. The "ancestors" don't figure into any of the calculations, they would skew the numbers too much, those are shown to allow you to "scale" the vertical mountains of your market and get a birds eye view of the lay of the land.
Next are the "siblings" of your seed term. These are terms that have roughly the same numbers of competing pages as the seed and are either closely related or a bit tangent, but important either because of cost or traffic.
The third class are the "children". These are keywords that are a "generation" smaller in terms of numbers of competing pages. These terms also may be tangent, but are notable because of cost and or traffic.
The "best" of the siblings and children are further selected and called "interesting theme terms". "Best" is defined by PPC cost, PPC traffic and Natural traffic as well as "relevance" and "importance" to the seed term.
Let's look at how adding the "interesting terms" to our cluster overview changes the results for our market segment:
Notice now you see how many interesting keywords are contained in each theme. Example "homebrewing (0/25)" means that homebrewing has zero synonyms and 25 interesting keywords. You'll see now that the cost per click min/max/avg are different for each theme because it's no longer showing the information for just the seed keyword.
The competing pages are still reflecting the seed term because there are no synonyms for these themes. However the traffic, searches and ORV are composite values for all the interesting keywords, as well as the seed.
Notice what happened to the pie chart showing the competitors with the highest ORV, now including the interesting keywords for our theme.
Usually some general information site like wikipedia takes the top position when you start looking at this many terms. Nobody else has the ability to rank for so many diverse terms.
With this view, you start to see the big players ranging across all these terms in your market segment. Surprisingly, it looks like my beer market is lead mostly by recipe sites!
This is showing me that they way I have segmented my market is not cohesive from a current market standpoint. In other words, nobody else in the "beer kits" market is going after this diverse a collection of terms.
Now you can see how segmenting a market and analyzing that segment can give you some depth of knowledge.
Let's look at one more thing. Let's see what happens when I delete the "homebrewing" theme. Remember I said that it was a little to general to include as a synonym in my cluster, so I was going to make it a KGO term? Let's see how removing this term from the synonyms is going to effect my results.
As you can see at the top of this last pie chart, the Organic Market Value across the cluster dropped substantially. However, wikipedia is moved down into third place and my competitors look much more like brewers and less like a 'who's who' in a cook's favorites listing.
This demonstrates how you can hone your keyword list, refining them to closer reflect a market segment.
It doesn't mean that "homebrewing" is not an appropriate term for my website, my USP, or even this silo; however, it does not look like the market currently considers it a synonym for the other terms that I am targeting in my "beer kits" silo.
This further substantiates my original idea for making this term KGO; however, this is not always the case.
Typically, you will always want to make a term KGO when the term is one that you are targeting further up your silo structure. However if this is a term that you WANT to target at this level, and you feel that it belongs with these terms, and not in another silo or article, then you are breaking new ground.
Either your product or service is new on the market (for example, you are the first person to combine peanut butter with chocolate) or your USP has created a unique combination, or no one else in your market is targeting all these terms. If this is the case, then go for it! Add it as a synonym for your silo and redefine the market.
If this is not the case then you will want to determine which theme or themes are causing the lack of coherence, and consider moving those to another cluster to form a new silo or article in your website.
As far as I know, Krakken VOMA is the only tool on the market capable of performing Market Segmentation and Analysis for your website project. Subscribe now to 'get Krakken' on your next project.