Corvidae’s Attribution Model
What attribution model does Corvidae use?
We partnered with Edinburgh University to develop a truly unique approach to attribution: Visit-Level Attribution.
This patented cookie-free approach uses a custom technique for each customer to understand how people really interact with your marketing and website, the decisions they make, and how those results in conversions and revenue for your business.
What is visit-level attribution?
As customer journeys become more complex, having the ability to unify data silos and go beyond channel attribution is crucial to understanding where the true value lies in your marketing efforts.
Visit-Level Attribution (VLA) is the only way to obtain the granular level of data needed to get the complete picture of an individual customer’s journey.
This allows you to:
- determine the most valuable interactions across the customer journey
- move budget from underperforming channel and campaigns
- acquire new customers for less
- drive growth and ROI
Find out more: A Complete Guide to Attribution Models
VLA allows you to go beyond existing attribution models to paint a far more accurate picture of the individuals interacting with your marketing activity and see their full conversion journey.
Benefits of VLA
Optimise your marketing budget
Understand the user journey at the most granular level
Improve the quality of your underlying data to allow greater accuracy of attribution
How does VLA work?
A technical overview
Before ingestion into our Neural Network, our proprietary Machine Learning model ingests data from each marketing channel and stitches sessions across devices and data silos.
By using probabilistic techniques, we can generate a view of individual sessions and unify a user’s activity across all touchpoints to give a clear view of where conversions are taking place – and the touchpoints that influence these conversions.
But Corvidae and VLA allows you to take this one step further.
By understanding where conversions are taking place in the customer journey, we can identify which channels and campaigns have opportunities for growth and begin to automate this process.
Allowing you to move marketing spend from areas that are underachieving and move it elsewhere – resulting in lower CPAs and increased ROI.
80% of your core analytics data is wrong.
Reaching beyond Google 360 and Adobe Analytics, Corvidae is the only attribution solution which completely rebuilds your marketing data.
How VLA boosts your attributable data
Complex customer journeys involving interactions with a range of devices and media – often over an extended period – present some very real challenges when it comes to collecting and effectively attributing data around these journeys.
VLA provides marketers with an accurate view of an individual’s probability to convert at every touchpoint.
It does this by effectively rebuilding your marketing data and laying online data alongside offline data to identify overlaps in time, location, and other identifiers.
For example, if a customer shows a heightened level of engagement with one touchpoint than is commonly seen by most customers, this is then given a greater weight when attributing the eventual revenue to that individual event.
Using the VLA deployed in Corvidae gives a significant boost in attributable data, often by 300% or more compared to Google or Adobe analytics packages.
Enhanced campaign-level attribution with VLA
One of the key benefits of VLA is the more granular campaign level view of attribution it provides. Channel attribution still exists but is aggregated.
This means that you can split up and view the data in different ways:
Paid Search
- By source (Google/Bing/etc.)
- By campaign, including brand vs. non-brand
- By competitor key phrases
Display/Native Ads
- By source (e.g. Google DV360, Amobee, Beeswax)
- By campaign
- By referrer
Affiliates
- By campaign
- By source
- By referrer (Note: referrers don’t usually allow impressions to be tracked via them.)
Corvidae attribution case study examples
Working with one of the UK’s largest retailers, and using our patented VLA approach, we were able to correctly attribute an additional £8 million of their spend to revenue – which hadn’t been attributed to any channel previously.
As a result, hugely affected the way they assessed the impact of their spend moving forward.
For another client, Corvidae reduced the cost of sale by 12% and eliminated £1 million of media spend from Google ads, and delivered the same amount of revenue in a 6 month period.
VLA provides marketers with an accurate view of an individual’s probability to convert at every touchpoint.
Attribution model FAQs
No.
Corvidae attributes revenue to the most granular level possible – the individual visit or impression, and then allows aggregation up to insightful segments.
Shapley and Markov models cannot handle such a large amount of data points in their operation and the underlying mathematics fail. Due to this, we have developed our own attribution modelling techniques to allow our own approach to be viable.
Deterministic data refers to data collected through marketing activities that is clear, unequivocal, and exact. With this data, you can be certain of the identity of an individual that has interacted with your brand. Examples of deterministic data include: age, gender, interests and previous purchasing history.
Probabilistic data relates to data you might receive from offline advertising or social media channels, like Facebook or Amazon. You know that your paid, earned, or shared channel exposure is likely to have reached a given number of people that fit within the profile of your target consumer. But it’s not possible to be certain that someone who takes the desired action on your website was one of those consumers.
Last click is heavily biased towards direct visits. Last non-direct click ignores all the preceding channels and first click rewards only the first touchpoint, meaning marketers can never understand the true value of their marketing mix. It is impossible for them to understand the impact each channel has on total revenue and the customer journey.
The machine learning that Corvidae uses enables us to ‘see’ the customer behaviour behind the generated clickstream data, meaning we can stitch together data points that represent a single user moving across multiple channels and devices.
With the user journey accurately mapped, Corvidae determines how valuable each event in the journey was in making someone transact. With that, Corvidae can split the overall revenue across many different events that can be grouped into campaign, creative, channel, etc groupings to evaluate overall performance”.
Our studies have shown that in comparison to Google 360 Last Click, we are able to attribute over double the amount of revenue.
Find out more in our blog: Corvidae vs. Last Click: Solving the Attribution Problem
Getting started is easy
Thanks to our first-party pixel deployment, getting started with Corvidae is easy. After just 30 days, you will begin receiving accurate and actionable attribution reporting.
Find out how Corvidae’s attribution model can transform your marketing measurement.
Corvidae allows you to focus on finding higher value, lower cost opportunities further up the funnel to boost your marketing effectiveness and increase ROAS.
Let’s talk about how we can put Corvidae to work for you.