Having spent a large amount of time developing an approach to keyword research, I have come to the conclusion that the following goals are applicable to most campaigns: <strong>- Find keyword themes to gear the website to for long term success</strong> <strong>- Find key phrases/pages where concentration can quickly, and cost effectively, improve sales/leads (called ‘low hanging fruit’).</strong> Whilst the former is incredibly important, this blog post is going to deal with the latter.
Having spent a large amount of time developing an approach to keyword research, I have come to the conclusion that the following goals are applicable to most campaigns: - Find keyword themes to gear the website to for long term success - Find key phrases/pages where concentration can quickly, and cost effectively, improve sales/leads (called ‘low hanging fruit’). Whilst the former is incredibly important, this blog post is going to deal with the latter.
Firstly, you need to put together a set of keywords. I’m not going to go into this step too deeply as there is some fantastic information out there about how to do this – particularly here and here. My extra layer of gloss on top of this is how to assess this keyword set to find ‘low hanging fruit.’ This means collecting metrics relevant to the following three criteria and finding the right balance between them.
Criterion 1: Will this phrase deliver a significant amount of extra traffic quickly?
The two metrics that we need to collect that are relevant to this are exact monthly search volume and current ranking. As you may be aware, the higher the ranking, the higher the percentage of clicks gained. Therefore a website is already ranking on the first page, then increasing by one place, which will have a bigger impact then increasing by x places, but remaining on page two or below. Therefore, any assessment of ranking with regards to this criterion must be heavily weighted in favour of phrases already ranking on the first page (but not first placed, as you can’t do much more!).
Criterion 2: Will it be cost effective to get the phrase to a position that delivers a significant amount of extra traffic?
The metric I use to cover this criterion is SEOmoz’s keyword difficulty. The upshot is that the lower the percentage, the lower the difficulty of ranking highly. A limitation of this metric (for the purpose in which I am using it) is that the keyword difficulty metric measures the difficulty of reaching the first page (top ten). This means if you are already on there for a key phrase, then it may not quite fit with measuring the difficulty of ranking even higher i.e. top five, three or first. However, I still tend to think of it as a good indicator as to how difficult it will be to advance the position of a key phrase in general.
Criterion 3: Will this traffic convert to leads/sales?
This is probably the most important. An obvious choice of metric to take a look at for this criterion would be conversion rate. However, this rests upon three assumptions:
- The page that is ranking is the high quality and relevant: this is not always the case. A key phrase could be worth concentrating on but has a low conversion rate because the wrong page is ranking.
- Key phrases with rankings beyond the first page attract enough traffic to gain conversions: this is doubtful. We don’t want to completely discount phrases just because they haven’t had a fair shot by being lower down in the rankings.
- You have access to conversion data: not everyone does, especially at the proposal stage of a project.
Points to note: - I have weighted the rankings using Chikita’s traffic percentages. - Difficulty weighting is skewed towardslow percentages. - I have tried to ensure that the formula detects if you enter a key phrase that is already ranking first (as long as you fill in the other cells correctly). In addition, by adding the URL ranking for each phrase, you can then use a pivot table to add up the scores associated with each to identify pages that are particularly strong (in terms of opportunity), as illustrated by the second tab (‘URL scores’). You can edit the weighting and grading tabs to fit your needs – for example you may want to tweak the grading bands if you are working on in an area with relatively low search volumes or relatively high difficulty. As with any method, there are limitations. I look forward to hearing ideas for improvements. Also just to note, the data I’ve put in the spreadsheet as an example is pretty random – derived from sites that I’ve had a look at in the past but not necessarily having worked on or become affiliated to in any way. Kind thanks goes to Rand Fishkin who made some helpful suggestions during the development of the spreadsheet. @richlawre