AI Case Study
Wall Street Journal improves visitor to subscriber conversion rates by optimising experience to match machine learning modelled propensity to subscribe
Faced with the media-wide shift from print to online, and with online advertising margins being squeezed, growing subscriber numbers is key to building a sustainable business. The team modelled propensity to respond to subscription offerings - offering different editorial packages (and limitations to free access) to maximise the take-up rate. This evolved over both time (of day, of month, of year) differences but also as a way to balance against advertising demands (e.g. the need to provide guaranteed eyeballs for advertisers) to ensure optimal commercial results.
Industry
Consumer Goods And Services
Media And Publishing
Project Overview
Non-subscribed visitors to WSJ.com each get a propensity score based on more than 60 signals or variables. These include frequency, recency, depth, favoured devices and preferred content types (e.g. whether the reader is visiting for the first time, the operating system they’re using, the device they’re reading on, what they chose to click on, and their location - the analysis will infer demographic data from that location).
The propensity score then decides how many free articles the reader will see creating a more flexible paywall that takes away guesswork around how many stories, or what kinds of stories, to let readers read for free.
According to The Drum: "Non-subscribed visitors fall into groups that can be roughly defined as hot, warm, or cold .... those with high scores above a certain threshold — indicating a high likelihood of subscribing — will hit a hard paywall. Those who score lower might get to browse stories for free in one session — and then hit the paywall. Or they may be offered guest passes to the site, in various time increments, in exchange for providing an email address (thus giving the Journal more signals to analyze). The passes are also offered based on a visitor’s score, aimed at people whose scores indicate they could be nudged into subscribing if tantalized with just a little bit more Journal content."
The WSJ has fixed subscription pricing. They plan to use these techniques to predict churn - which is the sort of area that one imagines could lead to more variable pricing. The WSJ employs a team of about 10 subscription analytics staff - one of their tasks is top provide colour on the propensity groups to help managers make informed decisions.
Reported Results
The WSJ is cagey around conversion rate numbers or attempting to quantify the direct impact of their machine learning amongst the broader mix of marketing tools. They do point to the organisation’s latest subscriber numbers - over 3m in total (of whom 1,389,000 were digital subscribers, according to News Corp’s latest earnings report in 2018, up from 1.08 million a year earlier).
Technology
Function
Marketing
Marketing Research Planning
Background
The WSJ, owner of premium branded business content, was one of the first publishers to go live with a paywall, aiming to drive subscription revenue online. Historically there were three paywall methods - freemium (much free but certain features charged), metered (a limited amount for free) and hard (everything is charged) - which all essentially offered standard propositions. The WSJ has built what it calls a dynamic paywall - offering users with different modelled behaviour different paywall experiences.
Benefits
Data