Understanding Conversion Probability data and report in Google Analytics

Introduction to ‘Conversion Probability’

In the context of Google Analytics, ‘conversion probability’ can either refer to the ‘conversion probability report’ or ‘% conversion probability’ dimension.

Introduction to ‘Conversion Probability report’ in Google Analytics

By default the conversion probability data is available in the conversion probability report.

Follow the steps below to view the conversion probability report in Google Analytics:

Step-1: Login to your Google Analytics account and then navigate to the reporting view which has been collecting ecommerce data for the last 30 days or more, with 1k or more transactions recorded in the last 30 days.

If you are accessing the GA view for the first time, since the new ‘conversion probability’ report is available for your view then you will most likely see a guided tour:

Step-2: Click on the ‘SEE REPORT’ button to jump straight into the conversion probability report. Otherwise, navigate to Audience > Behavior > Conversion Probability:

Note (1): You will see the ‘conversion probability’ report in your GA view only when your reporting view meet the minimum requirement for calculating the ‘conversion probability’ data which is at least 1000 ecommerce transactions, recorded in the last 30 days.

Note (2): In the context of ‘conversion probability’ report, a conversion means only ‘ecommerce transaction’. Thought this definition is likely to change in the future.

Conversion Probability report is available in the Google Analytics Demo Account

If your GA reporting view is not eligible for ‘conversion probability’ reporting then you can navigate to the Google Analytics Demo account to see the ‘conversion probability’ report.

This report is available there.

Minimum requirement for calculating and reporting on the ‘conversion probability’ data.

Your GA reporting view must have recorded at least 1000 ecommerce transactions in the last 30 days for the ‘conversion probability’ data to be calculated and reported in your view.

Once your reporting view is eligible for ‘conversion probability’ data, you will see the ‘conversion probability’ report in your GA view.

Note (1): Once your GA view reach the initial threshold of 1000 ecommerce transactions, Google Analytics then needs 30 days of data to model. If the number of transactions in the reporting view falls below 1000 per month, then GA uses the last good model to generate data for the report.

Note (2): The ‘conversion probability’ data is delayed by 24 hours. So you won’t see ‘conversion probability’ data for today. In order to see ‘conversion probability’ data for today, you would need to wait till tomorrow (i.e. for next 24 hours)

Introduction to ‘Conversion Probability’ data in Google Analytics

The ‘conversion probability’ data in GA is made up of:

#1 ‘% Conversion Probability’ dimension

#2 Up to five ‘% Conversion Probability buckets’

#3 ‘Average % Conversion Probability’ metric

Google Analytics use machine learning to compute the ‘conversion probability’ data for your reporting view, provided your reporting view meet the minimum requirement for calculating the ‘conversion probability’ data.

You can see the ‘% Conversion Probability’ dimension, ‘Average % Conversion Probability’ metric and the five ‘% Conversion Probability buckets’ through the ‘Conversion Probability’ report.

Introduction to ‘% Conversion Probability’ dimension

‘% Conversion Probability’ is a dimension which is an estimate of, how likely, a user will make a purchase on your website in the next 30 days.

It is calculated for each user.

Its value can range from 1 to 100.

Higher the ‘conversion probability’ score for an individual user, higher is the probability, the user will make a purchase on your website in the next 30 days.

Conversely, lower the ‘conversion probability’ score for an individual user, lower is the probability, the user will make a purchase on your website in the next 30 days.

If the ‘conversion probability’ score for a user is close to 1 then it means the user has no probability of making a purchase on your website, in the next 30 days.

If the ‘conversion probability’ score for a user is close to 100 then it means, it is extremely likely, the user will make a purchase on your website, in the next 30 days.

If the ‘conversion probability’ score for a user is zero then it means GA could not calculate the ‘conversion probability’ score for the selected time period.

You can see the ‘% conversion probability’ dimension:

  • In the ‘conversion probability’ report
  • While creating a custom report
  • While creating a custom segment
  • While creating a remarketing audience

Introduction to ‘% Conversion Probability buckets’

The ‘‘conversion probability’ ‘ report show distribution of: ‘sessions’, ‘sessions with transactions’ and ‘sessions without transactions’ for each of the following five ‘conversion probability’ buckets:

0-1

2-5

6-20

21-50

51-100

Each ‘conversion probability’ bucket (except 0-1) is made up of GA sessions whose ‘conversion probability’ score fall in particular range.

For example the bucket ‘51-100’ is made up of GA sessions whose ‘conversion probability’ score is in the 51-100 range.

You can open a ‘‘conversion probability’ bucket’ by clicking on the ‘+’ button next to it:

There is a possibility that you do not see a particular ‘‘conversion probability’ bucket’ or a particular range in your ‘‘conversion probability’ report.

For example, you may not see the ‘51-100 ‘conversion probability bucket’ in your report, if no GA sessions were recorded for your website where the ‘conversion probability’ score was in the 51-100 range.

Similarly, you may not see the ‘91-100’ range within the ‘51-100 conversion probability bucket’ in your report, if no GA sessions were recorded for your website where the ‘conversion probability’ score was in the 91-100 range.

Other thing to note that the ‘conversion probability’ buckets are not evenly distributed.

For example the ‘2-5 conversion probability bucket’ contains 4 points of spread, whereas the 51-100 bucket contains 50.

You may glance at the report and not realize that these rows are not equal.

Introduction to ‘Average % Conversion Probability’ metric

The ‘Average % conversion probability’ is the average value of ‘conversion probability’ score calculated for all users related to a particular Google Analytics dimension in a given time period.

You can see the ‘Average % conversion probability’ metric:

  • In the ‘conversion probability’ report
  • While creating a custom report

Through ‘conversion probability’ report you can determine the ‘average conversion probability’ for all users related to a primary dimensions like:

Note: The ‘Average % conversion probability’ score of 0.0 means the ‘‘Average % conversion probability’ metric was not calculated for all the users, related to a particular dimension, in the selected data range.

Introduction to ‘Conversion Probability Segment’

A ‘conversion probability segment’ is a type of custom segment which meets certain range of ‘conversion probability’ score.

You can create this segment by clicking on the ‘conversion probability segment’ button which is located next to a ‘conversion probability bucket’ in the ‘conversion probability’ report:

You can use ‘conversion probability segment’ to determine:

  • What percentage of your website users are most likely to convert in the subsequent sessions.
  • How the website users who are most likely to convert in the subsequent sessions, differ from other website users, in terms of their website usage.
  • How the marketing campaigns that generate website users, who are most likely to convert in the subsequent sessions, differ from other marketing campaigns.
  • Which conversion paths are followed by the website users, who are most likely to convert in the subsequent sessions.
  • Which conversion paths are followed by the website users, who are least likely to convert in the subsequent sessions.

Remarketing audience based on conversion probability

You can create a remarketing audience from a ‘conversion probability segment’.

Just right click on the down arrow button next to the ‘conversion probability segment’ and then select ‘build audience’:

You can publish this audience to Google Adwords for running remarketing campaigns or publish the audience to Google Optimize for testing purpose.

Similarly between conversion probability and session quality reports

If you have used ‘session quality’ report before then you can now see, that is very much similar to the new ‘conversion probability’ report.

If you are eligible for ‘session quality’ reporting, you are most likely eligible for ‘conversion probability’ reporting too.

If you do not see the ‘conversion probability’ report in your reporting view, despite of being able to access the ‘session quality report’ then most likely, GA has not computed the ‘conversion probability’ data for your view so far.

When that computation is complete, the ‘conversion probability’ report should be available in your reporting view.

My take on the ‘Conversion probability’ data

Though Google claims in its documentation that the ‘session quality’ report determine users’ proximity to transacting, the report itself talk about sessions proximity to transacting.

As the name suggest, the session quality data is actually based on sessions and not users, and hence the report is named ‘session quality’ and not ‘users quality’.

This is not the case with ‘conversion probability’ report where conversion probability score is calculated for each individual user.

So Google’s focus on optimizing for users rather than ‘random sessions’, is a step in the right direction.

However, Google is not transparent about its conversion probability estimates.

It conveniently uses the word ‘machine learning’, wherever it is supposed to explain, how the conversion probability estimates are actually calculated.

And this is now becoming a sort of trend, for Google.

Just throw the word ‘machine learning’ somewhere in the documentation and then no one will question us.

“Oh, it is machine learning, must be accurate.”

But marketers often forget that high value of data maturity is required, for a machine learning algorithm to actually work for your business and most organizations lack that.

If a company has a high level of data maturity then it means it has immediate access to high volumes of high quality data from multiple data sources and it has people and processes in place to capitalize on actionable insight in a timely manner.

Majority of companies, big or small, do not have the level of data maturity required to benefit from machine learning.

Yet they still use and rely on machine learning tools, in a hope that by some magic, it will improve their website conversions.

This could have dire consequences.

Instead of improving sales their conversion rate and sales could tank.

So can you trust the ‘conversion probability’ score?

I leave it to you, to make an informed decision about it, through your own testing.

I won’t trust it.

I would use the more reliable method of creating remarketing audiences based on enhanced ecommerce segments.

You can learn more about creating remarketing audiences based on enhanced ecommerce segments from this article: Using Enhanced E-Commerce segments for Remarketing in Google Analytics

However, the ‘conversion probability’ report is still a nice addition to the Google Analytics arsenal.

 

Do you know the difference between Digital Analytics and Google Analytics?


99.99% of course creators themselves don’t know the difference between Digital analytics, Google Analytics (GA) and Google Tag Manager (GTM).

So they are teaching GA and GTM in the name of teaching Digital analytics.

They just copy each other. Monkey see, monkey do.

But Digital analytics is not about GA, GTM.

It is about analyzing and interpreting data, setting up goals, strategies and KPIs.

It’s about creating strategic roadmap for your business.


Digital Analytics is the core skill. Google Analytics is just a tool used to implement ‘Digital Analytics’.

You can also implement ‘Digital analytics’ via other tools like ‘adobe analytics’, ‘kissmetrics’ etc.

Using Google Analytics without the good understanding of ‘Digital analytics’ is like driving around in a car, in a big city without understanding the traffic rules and road signs.

You are either likely to end up somewhere other than your destination or you get involved in an accident.


You learn data analysis and interpretation from Digital analytics and not from Google Analytics.

The direction in which your analysis will move, will determine the direction in which your marketing campaigns and eventually your company will move to get the highest possible return on investment.

You get that direction from ‘Digital analytics’ and not from ‘Google Analytics’.


You learn to set up KPIs, strategies and measurement framework for your business from ‘Digital analytics’ and not from ‘Google Analytics’.

So if you are taking a course only on 'Google Analytics’, you are learning to use one of the tools of ‘Digital analytics’. You are not learning the ‘Digital analytics’ itself.

Since any person can learn to use Google Analytics in couple of weeks, you do no get any competitive advantage in the marketplace just by knowing GA.

You need to know lot more than GA in order to work in digital analytics and marketing field.


So what I have done, if you are interested, is I have put together a completely free training that will teach you exactly how I have been able to leverage digital analytics to generate floods of news sales and customers and how you can literally copy what I have done to get similar results.

Here what You'll Learn On This FREE Web Class!


1) The number 1 reason why most marketers and business owners are not able to scale their advertising and maximise sales.

2) Why you won’t get any competitive advantage in the marketplace just by knowing Google Analytics.

3) The number 1 reason why conversion optimization is not working for your business.

4) How to advertise on any marketing platform for FREE with an unlimited budget.

5) How to learn and master digital analytics in record time.

 
 

My best selling books on Digital Analytics and Conversion Optimization

Maths and Stats for Web Analytics and Conversion Optimization
This expert guide will teach you how to leverage the knowledge of maths and statistics in order to accurately interpret data and take actions, which can quickly improve the bottom-line of your online business.

Master the Essentials of Email Marketing Analytics
This book focuses solely on the ‘analytics’ that power your email marketing optimization program and will help you dramatically reduce your cost per acquisition and increase marketing ROI by tracking the performance of the various KPIs and metrics used for email marketing.

Attribution Modelling in Google Analytics and Beyond
Attribution modelling is the process of determining the most effective marketing channels for investment. This book has been written to help you implement attribution modelling. It will teach you how to leverage the knowledge of attribution modelling in order to allocate marketing budget and understand buying behaviour.

Attribution Modelling in Google Ads and Facebook
This book has been written to help you implement attribution modelling in Google Ads (Google AdWords) and Facebook. It will teach you, how to leverage the knowledge of attribution modelling in order to understand the customer purchasing journey and determine the most effective marketing channels for investment.

Himanshu Sharma

Digital Marketing Consultant and Founder of Optimizesmart.com

Himanshu helps business owners and marketing professionals in generating more sales and ROI by fixing their website tracking issues, helping them understand their true customers purchase journey and helping them determine the most effective marketing channels for investment.

He has over 12 years experience in digital analytics and digital marketing.

He was nominated for the Digital Analytics Association's Awards for Excellence.

The Digital Analytics Association is a world renowned not-for-profit association which helps organisations overcome the challenges of data acquisition and application.

He is the author of four best-selling books on analytics and conversion optimization:

error: Alert: Content is protected !!