The Ultimate Ecommerce SEO Strategy: Leveraging Product Level Data

When working on large ecommerce sites focus can be a key issue, especially where budgets may be limited. Selling in budget increases can also be extremely hard to do when faced with clients, rightfully, expecting great returns on their money.

It’s an issue our agency comes across a lot and so we have invested a huge amount of time and brainpower in working out the most effective way of measuring where the lowest-hanging fruit may be and to back up any suggestions of campaign budget increases. And we’re not talking ‘finger-in-the-air’ stuff or simple SERP analysis here. We’re talking ‘get your hands-dirty-knee-deep-in-Analytics-data’, quantifiable stuff.

While there are a plethora of ways to get statistically more involved, we now use ecommerce and funnel data to steer our tactical work, and I wanted to share one particularly effective way of working with the Moz community today.

The Three-Tier Approach

The process involves breaking your target keywords into three distinct sectors. This helps us segment what can be a huge task into three manageable chunks and also highlight three very different types of keyword in a structured way. Those three ‘tiers’ are as follows:

1. ‘Vanity Keywords’ – The terms we know don’t convert amazingly well but look great and bring big chunks of traffic.
2. ‘Brand Level Keywords’ – Think ‘Abercrombie Clothing’ – small traffic wins but big ROI.
3. ‘Easy Reach, High Conversion Keywords’ – Low competition, three word phrase plus terms sitting within the first three pages that convert at crazily high levels.

We then have three separate on- and off-page strategies to achieve top positions for each of the three campaigns, and each one is approached very differently.

Clearly those Tier One terms need the most muscle and it’s the tried and tested process of content creation, syndication and links that will get you there, but this post is not focused around how and where to do that.

It is the second and third tiers that we want to drill down into today as they offer the best ROI. But that isn’t to say that winning in this space is easy, as tackling such a broad spectrum of terms requires a lot of planning. To do it well you simply have to ensure that your data sets are incredibly well organized and thought out.

So what do we call it and how do we go about focusing in on these terms? Around the office we call the concept Product Level Action, and it’s basically about smoking out the site’s hottest product prospects to allow us to quickly improve those with the best ROI.

Here’s how we make it work:

Using Analytics we create a simple Custom Report that captures keywords, visits, revenue, and average revenue per visit. We then sort it so that it surfaces the keywords with the best Average Revenue per Visit and extract to CSV.

Next we add in a column recording where we rank for that term and then, using the Google Adwords tool, extract the EXACT MATCH monthly audience volume for each keyword on our list. You end up with something that looks like this:

So far so good. And here’s where it starts to get really interesting.

With the data in front of you it is now possible to very easily build up a ‘real’ picture of the potential revenues that COULD be generated if you were in, say, first, second, or third position for each of them.

To map this we need one more piece of data and that is average click through rates based on rank. There is clearly a lot of debate around how accurate an average could be but the much-respected SlingShot SEO just produced this awesome CTR study and many are already using it as a benchmark.

For the data below we have used a slightly older study by Quadzilla but the process is the same irrespective of which data you go with.

Here’s how you do that, step-by-step:

1. Create a new table that extracts Keyword and Average Revenue per Visit from the first table you created.
2. Add in three further columns for first position, second, and third. We will then fill these with the projected revenues based on search share.
3. Take the overall Exact Match volume for the first keyword you want to analyse (in this example we will use Blue Widget), so for our first term it is 1000.
4. Work out what 25.1% (average volume share for first positioned sites) of 1000 is, as this will give you the expected number of visits if you were in first position for that term.
5. Now times this number (251) by the Average Revenue per Visit (2.8 rounded down) – 251 x 2.8 = 702.80.
6. Therefore the projected revenue from being in first position for Blue Widget would be £702.80.
7. Repeat for second and third positions using their respective % share of volume percentages.

You end up with something that looks a little like this:

What a great sales story that is for your client! You can now tell them, with reasonable accuracy, what a top three position is worth and therefore what their ROI would be based on whatever extra spend you think is necessary to get them there.

And by aggregating all of these terms together you can now extrapolate out search spend V revenue.

Armed with this kind of insight and some extra data around the competitive set it’s then a relatively easily sell to suggest to the client that for X amount more time/spend you could increase their earnings considerably. And the beauty is the data is there to back up the claims.

This piece of analysis is just one piece of our insight and action jigsaw but in my eyes it is also the stage that gets the client most excited and engaged.

Simon Penson is founder of content-led SEO agency Zazzle Media. You can follow him @simonpenson.