SEO ROI: Create a Revenue Forecast

Get A Rough Estimate Of Potential SEO Traffic

In Short: Forecast SEO REVENUE Potential & Not Just Traffic Volume

Don’t just blindly choose keywords based on the generic volume numbers you extract from any old SEO Tool, spend some time to estimate your potential conversions into leads, how many leads turn into sales, how much revenue gained per sale and if it’s a recurring revenue project then how long do people usually stick around.

Want Me To Do A Keyword Forecast For You? Okay, But It’ll Cost you $5
(No Seriously. Give me a keyword, conversion rate, lead to sale rate, revenue per sale and I’ll uncover a decent sized set of related keywords and provide you with a forecast of potential SEO revenue.

Step By Step Process To Calculate SEO ROI

  • Do this process for each unique product(product type if they’re all super similar) or service
  • Calculate at what rate is of your traffic that turns into leads by filling out a form (Conversion rate)
  • Calculate how many leads turn into sales (Lead to Sale Rate)
  • Define how much revenue you make per sale. (Revenue per sale)
  • Optional: If this is a recurring service, multiply the revenue by the average lifespan of the client (how many times you will get to charge them)
  • Gather the list of relevant keywords that are most likely actually purchase the product or service
  • Get the total volume of those keywords
  • Multiply that volume by a PLAUSIBLE organic click through rate, this study by Advanced Web Rankings is pretty good.
  • Take your potential traffic number and multiply it by the conversion rate, lead to sale rate and revenue per sale. This will provide you with a realistic potential SEO Return on Investment and a solid number you can show your boss or client.

Organic Traffic Estimator: An SEO Forecasting Process

Process for attempting to evaluate traffic potential , I have taken the following, general approach:

  • I identify current device% breakout between mobile and desktop via analytics data
  • I use Rand/Jumpshots study of click death by device type (this may be controversial to some people, but I’ve noticed it models our actual data pretty decently)
  • desktop/mobile search volume from your tool of choice (e.g. SEMrush, Ahrefs, etc.)
  • averaged CTR for top five positions (generally what we’re shooting for re: rankings), pulled from top X keywords in GSC

So, for example:
I want to project what my outcomes may be if I rank in top 5 for any given set of terms, thereby projecting what my opportunity is if I try to invest. This can be used in conjunction with other metrics, such as average difficulty for a corpus of keywords, etc. to make decisions on re: do you invest your time in them or not.
Here are the things I work through –
Calculating target ranking position (target ranking position = top 5) average – I recommend doing this for desktop and mobile to use the relative CTR for each later on:

  • I pulled the top keywords from GSC into sheets for the last 6 months (capped out the pull at around 50k terms gathered)
  • I rounded the position data to nearest whole position and applied groupings to them (position 1, top 3, first page, etc.) for future analysis
  • I used averageif on CTR for all top 10 positions (averageif pos = 1, averageif pos =2, etc.)
  • Then, I average the top 5 position averages, giving me a relative average for if I ranked in the top 5 positions
  • this % gets used in final equation

Find device breakout for current audience:

  • from your analytics of record (Omniture, GA, etc.), find out the device breakout by mobile vs desktop
  • these % get used in final equation

Leverage click data from Rand/Jumpshot study:

  • While a recent study – and one which people may scoff at, or ignore – I find being mindful of loss-of-click to be an important element; if you trust the click potential data from Ahrefs or SEMrush, you could use that info on the keyword level instead of using this broad study
  • let’s say we don’t use Ahrefs or SEMrush click estimates though, I would use the 39% clicks on mobile (61% no clicks was the number referenced in the study), and 65% clicks on desktop (34.5% no clicks was number referenced in study) in my final equation

Get desktop AND mobile data for keywords, as available:

  • for each term I’m going to include in my corpus for this analysis, I will try to get both the desktop data as well as the mobile data; if mobile data is not available (or vice versa), then I will use whatever is available
  • the keyword data could be for existing rankings (current marketshare/footprint), and/or for new terms we want to go after (gap footprint) – these can be used to support different questions (e.g. should we invest in optimizing current content and what would outcomes potentially be if so)
  • data needed = search volume and current ranking position (if pulling for current footprint)

Based on all this data, we can now calculate traffic potential.
traffic potential = ((mobile sv*0.39)*mobile traffic %)*avg T5 mobile CTR + ((desktop sv*0.65)*desktop traffic %)*avg T5 desktop CTR
This equation is applied to every keyword we currently rank for, not in the top 5 (e.g. position 6-100). This should give us insights into answering the question of “if we improve our ranking position for this corpus of keywords, what might the traffic estimate look like”.
You can then use this in comparison/conjunction with other metrics, like average difficulty for a topical category (e.g. risk reward based on comp to traffic opp), etc.

Actionable Recommendation

 My recommendation for this is to not look at/forecast on a per-term basis, but instead to do it in groups or as a whole (e.g. all terms that make up a certain topical niche, or all terms that reflect the current footprint for a site). By grouping things together, you get a better understanding of topical opportunity and risk/rewards (when looking at KW difficulty, revenue opps, etc).


Posted

in

by

Tags: