SEO has always been a difficult channel to understand in terms of ROI; performance is driven through multiple tactics and trying to attribute sales to strategy is almost impossible.
It has left marketers in a position where they know they need to invest but are unsure how to quantify investment and more importantly forecast growth.
Attribution is made even more difficult due to the fact Google moved to secure search in 2013, resulting in marketers losing valuable keyword data.
Further challenges were introduced this year when Google started to use ‘close variants’ to display keyword search volumes, meaning you don’t have an accurate picture of the market size.
Fortunately, there are workarounds to the latter but we are still in a position where we don’t understand what is driving organic performance.
So, what do we have to try and solve the problem of forecasting and ROI?
Firstly, we can still get ranking data through Googles API, this means we can know with some accuracy where we are appearing on Google for specific search terms. In Google Search Console we can also see partial traffic stats for these keywords, along with impressions and more importantly click through rates.
At Branded3 we have used this data as a starting point for preparing a forecast for SEO. Below are the steps we go through to put this together.
The first step is pulling together some meaningful keyword data; it needs to be substantial enough to provide a decent snap shot of your market place, covering the main keywords as they relate to your products and services.
The amount of keywords will purely depend on the size of your industry, 500 keywords may be acceptable for a niche retail business, but some of our clients in the financial sector may use up to 10,000 keywords to map out an accurate forecast.
There are lot of tools for discovering keywords, Google Keyword Planner remains the favourite, but if you want more information about your keywords and competition, you could use a paid tool like Keyword Explorer from Moz.
You then need to use a rank checking tool to understand your visibility for the keyword set you’ve selected. I wouldn’t recommend using any ranking data from Google Search Console as it’s often not an accurate reflection.
We use Advanced Web Rankings and have found it to be a solid platform for understanding keyword rankings on both desktop and mobile search.
Once you have pulled this data you should be left with a spreadsheet containing all your keywords, search volumes and their relevant ranking position on Google.
We now need to apply a click through rate model to the keyword set in order to understand how much of that potential traffic we’re actually receiving.
This is where things become particularly difficult as there is no way to get to a CTR model that is 100% accurate. Until recently we used a generic model from SlingshotSEO which suggested an 18% click through in position 1 and 1% CTR in 10th position on Google.
However, we have recently had to change the model due to changes in the amount of ads on Google, pushing down organic results and increasing clicks to ads, and also we found a new model to be more accurate.
We now use the Netbooster CTR model which gives a 19% CTR to position 1 and also shows CTRs as far down as page 3 on Googles search results.
You can now apply the CTR model to your keyword rankings to understand how much of the potential volume you are achieving, or not.
You should end up with a list that looks something like this:
As you can see from the above, the maximum traffic potential in this market is 2,361 visitors based on a maximum click through rate of 19%. However, due to current rankings we’re only realising 305 visitors or circa 13% of available traffic.
So we’re now at a point where we have collected keyword data, volumes, rankings and click through rates. We also understand based on our current visibility we have a traffic share of around 13% of the 2,361 potential market.
We now need to make some estimates on growth based on competition and investment. If a market is extremely competitive 3-5% traffic share growth is pretty typical, if it’s less competitive or you’re investing significantly then 10-15% is feasible. It requires some guess work and honesty in terms of your investment in relation to the wider market.
When forecasting over a 12 month period you should take into account:
- Seasonal trends based on analytics
- Any product launches
- Brand growth & ATL spend
- Market trends
Once you have decided on the traffic share you want to achieve, you can then plot out how you expect the growth to be delivered over a 12 month period. If you want to strengthen your business case you could add estimated cost per click amounts from Google Adwords to show how much the traffic would cost to purchase each month.
There is no way to forecast SEO with 100% accuracy, however this short guide is good starting point to understand your market size and the potential for growth.