Skip to main content
blog-1-3-750x300.png

Taking control of AI personalization with more meaningful and actionable metrics

November 6, 2019
Anne Claire Bellec Kameleoon
Anne-Claire Bellec
Anne-Claire Bellec is Chief Marketing Officer (CMO) at Kameleoon, in charge of the company's marketing strategy. She regularly shares her thoughts on digital on the blog, particularly focussing on the subjects of optimization and personalization and how they can increase online conversions.

1 Bringing transparency to AI personalization metrics

Whatever sector you are in, the number one aim of digital marketers is to increase conversions. These could be:

  • Buying a product
  • Subscribing to a service
  • Engaging with a brand (such as by reading a story on an online media site)
  • Taking the next step in the customer journey, for example signing up for a demo or a newsletter

Making changes to the user experience, such as through special offers or personalized content, is proven to have a major positive impact on optimizing conversion rates. Our in-depth conversion rate optimization (CRO) guide blog post explains this in more detail.

However, simply increasing conversion rates doesn’t necessarily drive value or greater profit. For example, simply offering everyone who visits your site free shipping will increase sales, but would many of those customers have bought anyway? How much money have you wasted unnecessarily? Alternatively, these free shipping pop-ins might impact the user experience of those that have no intention of buying, damaging your brand reputation and driving future shoppers away.

Digital marketers therefore need to take a more intelligent approach to optimization and conversions. Deploying AI personalization is one way to do this by helping to deliver an individual experience that will increase conversions, maximizing revenue and sales. However, it brings other challenges. Often AI vendors adopt a black box approach, meaning that marketers have no idea what course of action was taken and why it was successful. This recent interview with Kameleoon’s CTO, Jean-Noël Rivasseau, discusses this in more detail.

Building on that, in this blog I’m going to explain how targeting and scoring works transparently within Kameleoon and the benefits it brings.

2 AI personalization targeting – how it works

Put simply, AI personalization algorithms learn which customer segments will best respond to a particular offer, based on hot data around their behavior and characteristics. It then provides a personalized experience and offers with the aim of achieving the goals you have set around conversions.

However, in many other cases this black box approach means marketers don’t know what is actually happening when it comes to targeting. Are the AI algorithms targeting the right segments to maximize ROI? Are they actually delivering the best possible results for your brand and marketing campaigns?

Kameleoon is very different – we put transparency at the heart of what we do. While we follow the standard AI personalization approach of creating segments and learning about their attributes, our platform then compares visitors, in real-time, to these segments. It provides a personalized experience and offers to them when they are identified as matching the criteria of a specific segment. This is delivered in a much more open, understandable way – we want to provide marketers with a metric that they can analyze and report on with confidence.

3 Introducing the Kameleoon Conversion Score™

Like most AI systems, we create an instant raw score for every user at every moment of their customer journey on your website. This measures their probability of converting, but is website/condition dependent. So it means that one raw score may vary from one conversion goal to another, or depending on the company and industry. It therefore doesn’t provide a consistent metric that can be used in planning and analysis.

Instead, we use the raw score to create the Kameleoon Conversion Score™ (KCS™). This is an actionable, visible metric rather than an abstract concept. It provides a clear KPI that sits along other digital marketing metrics in your reporting dashboard. It hides the complexity of the underlying machine learning algorithms and heuristics.

It works by calculating a score for every visitor in real-time, based on a scale of 1 to 100:

  • A KCS of 1 means the visitor has the lowest chance to convert compared to other visitors.
  • A KCS of 100 means that they have the highest chance of converting compared to other visitors.
  • A KCS of 30 means they belong to the 30th percentile

This is calculated for each visitor at a specific Key Moment in their website journey, defined by the digital marketers themselves. The Key Moment can be defined using a standard rule, such as calculating after a specific session duration. Alternatively it can be made more advanced by using dynamic behavioral rules – such as whether it is a new or returning visitor or a visitor landing on a product page coming from Google Ads for example..

It is possible to define multiple Key Moments, which results in multiple KCS profiles. This provides the ability to target visitors at different parts of their journey, and thus avoid them slipping through the net when it comes to personalization.

How the KCS works and helps marketers to identify their next best action 

imageblog

 

The KCS is displayed as a conversion graph, color-coded from blue (low/cold) to red (high/hot), as this example shows.

It is simple to understand, analyze and act upon. You can select a specific range and drill-down into it, and then use this to create a new segment to target.

For example, the 50-70 range is normally made up of moderately interested but hesitant consumers. Would an incentive (such as a discount), help drive them to convert? By selecting the range you can then create and deploy a personalization campaign targeted at this specific segment - and then measure the result.

The KCS clearly provides a lot of data, but also helps marketers to understand it in business terms. Kameleoon automatically suggests the ranges with the greatest potential, marked with flags, identifying the biggest opportunities. These may not be the highest ranges (i.e. close to 100), which will convert anyway, but could be lower ranges that will deliver the greatest ROI. In the case of the graph above there is a major opportunity in the 40-45 range.

Essentially, the AI algorithm shows the opportunities, but you are in control of all actions, which you choose to pursue, and how you follow-up with specific segments.

segment

Turning the KCS into action 

Insight alone doesn’t deliver results. Consequently, Kameleoon makes it very straightforward to create actions and assign them, so that they will be launched automatically for visitors within a specific segment and range.

Simply zoom in, see what the potential is and then you can create a specific personalization for that segment. You can then trigger actions using a straightforward WYSIWYG editor without needing specialist programming skills. You can create offers, urgency messages and other actions, and select whether they are seen by all, or specifically target at new or returning visitors. And, to ensure that you hit your goals and maximize ROI you can cap the display of actions when you hit a specific target, such as a certain number of conversions. You remain in control, and can see exactly what is happening with every action. There is no AI black box preventing you understanding the impact of your campaigns.

Putting marketers in control of AI personalization 

Digital marketers want to maximize conversions, but also have specific ROI targets – after all, it is counter-productive to increase revenues if the free shipping you provide to each shopper means you make a loss on every sale.

They therefore need to be in control and to understand the process, working with clear marketing metrics that link to business KPIs. Marketers have grown used to this with many existing solutions, from email management to CRM, but being able to view metrics is much more difficult with many AI personalization solutions that operate as black boxes. With the KCS, Kameleoon provides a clear, understandable and actionable metric that puts marketers in control. This allows them to score and target effectively and consequently drive greater digital marketing success.

Topics covered by this article
Anne Claire Bellec Kameleoon
Anne-Claire Bellec
Anne-Claire Bellec is Chief Marketing Officer (CMO) at Kameleoon, in charge of the company's marketing strategy. She regularly shares her thoughts on digital on the blog, particularly focussing on the subjects of optimization and personalization and how they can increase online conversions.