How La Presse automates, optimizes and personalizes marketing in real time with Snowplow

Background

Once a traditional daily broadsheet, La Presse is now a fully digital, French-language newspaper with almost 4 million active readers each month. The service is available on various platforms, online at lapresse.ca, through its mobile app on Android and iOS, and for tablets through La Presse+.  

The Canadian media company knew the shift from print to digital meant joining an ultra-competitive fight for online users’ attention. Success would require seizing new opportunities to personalize content and advertising.

To do this, La Presse needed a sophisticated data strategy—including the ability to:

• Collect real-time, event-level behavioral data from all platforms
• Automate marketing actions
• Personalize the reader experience

All while managing the total cost of ownership.

Challenge

La Presse faced a twofold challenge in discontinuing their print edition:

1. Going head-to-head with the Google-Apple-Facebook-Amazon (GAFA) online ad juggernaut.

Competing on a solely digital media platform and sharing the ad market with GAFA, the La Presse ad offering had to reflect truly real-time insight, based on tracking everything that happened on their multiple platforms at the event level. This advanced requirement prompted a move away from third-party analytics providers to a data analytics solution that would fit better into the new reality and their tech ecosystem (the Modern Data Stack).

2. Enhancing the customer experience through data-led marketing and automation.

Shifting from print to digital required a new set of user considerations and opportunities. These included the possibility to tailor content, messaging and advertising to drive readers down the engagement funnel and create a truly personalized user experience.

Undertaking these initiatives required collecting and using data on a far more granular level. Luckily, La Presse already had a seasoned data team in place in their Growth, Analytics and Data Science department, with three main areas of expertise:

• Growth/digital marketing
• Consumer insights (reader demographics, declarative/non-trackable data, etc.)
• Data science & integration

La Presse’s Data Science & Integration team needed to ensure that they had the right technical solutions in place to support its business objectives. Finding and implementing a real-time analytics solution was key. This needed to meet the demands of the four key internal stakeholders at La Presse—sales, marketing, product and newsroom.

Yet, having worked primarily with Localytics and Google Analytics (free), two roadblocks stood in the way of getting more flexibility and transparency from their data:

1. As they increased the number of data points they collected, the total cost of ownership (TCO) was rising

2. The ability to track everything with no latency, and then control the data collected and how it was used became business critical

So how could they meet their requirement for flexible, real-time data at the right price point?  

Solution

In searching for a new analytics solution, two main use cases (the ad offering and the marketing automation piece) formed the basis for their requirements. La Presse sought a customizable, self-hosted, real-time analytics solution.

With the ability to track more data points, integrate with AWS, and customize the data pipeline in real time, Snowplow Behavioral Data Platform emerged as the winner.

Taking advantage of a “good internal data team”, La Presse set up their new Snowplow-enhanced data ecosystem. On a human resources/team level, this involved a specific internal workflow:

1. Data analysts gather internal data needs

2. Data engineering team deliver needed trackers and resulting data to data lake and lake shores

3. Data analysts do QA and validation internally, and deliver reports and analysis

4. Data Scientists build models on top of the integrated data

Snowplow delivered a game changer. We can offer business proactivity in Growth Marketing and Sales, and deliver zero-latency use cases.”

HERVÉ MENSAH | DIRECTOR, DATA SCIENCE & INTEGRATION AT LA PRESSE

La Presse’s data sources—which are internal (e.g. ERP, internal CMS), external (e.g. ads, app, social media, etc.) and Snowplow-based—are funneled into a data lake. While La Presse still relies on data warehouses for reports and dashboards that reflect consistent, well-defined, enterprise-wide data, the data lake has been instrumental in:

• Decoupling data from silos and opening it up
• Reducing the length of data discovery phase
• Surfacing and leveraging data in immediate ways, such as feeding new features into their apps (e.g. customization of segments they are developing, or directing readers straight to relevant articles)

The data lake provides specific, real-time, event-level data that enables this kind of agility, while the data warehouse continues to provide descriptive reports of what has happened before.

In the La Presse setup, the data lake ingests the raw data, and La Presse created a second layer on top of the data lake. called ‘lake shores’.

These lake shores, which maintain event-level granularity, are where La Presse splits data into tables that make more sense in terms of business needs. The data team then filters, cleans and gathers the data.

From here, the processed data is fed horizontally into La Presse’s business models. In addition to feeding into output that informs models, patterns and behavioral insights, the data is also fed into out-of-the-box or custom-built tools.  These include the customer journey builder (a marketing automation tool) and a segment builder for ad campaigns.

How La Presse uses Snowplow

Once the data engineering team implemented and configured Snowplow to their requirements, the possibilities were virtually unlimited. La Presse was immediately able to surface insights and build on top of Snowplow, with the stated aims of letting their marketing team create scenarios to establish desired customer journeys and give advertisers better targeting and value.

To do this, La Presse collects events from across its platforms with the Snowplow, powered by AWS, data collection pipeline. Trackers are set up on their website and two mobile platforms. Enrichments are applied, good and bad data filtered, and all data is delivered to the data lake through the real-time pipeline (figure 2).

Marketing automation with real-time qualification

One of the primary ways La Presse uses Snowplow is in animating its marketing automation process. This is essentially a real-time qualification engine that forges a user profile based on the journey a user/customer takes when using La Presse properties.

Put simply, when a reader takes an action, La Presse proactively recalculates how the reader is consuming content (e.g. by time of day, topic/article, demographic information, etc). Specific actions are then triggered in near-real time per user, resulting in an increasingly customized experience.

Not only does this deliver the best possible user experience, but it also guides the best course of action on the business side. That is, should La Presse push personalized content to keep the user engaged? Should they send an email? Based on everything the user data has taught La Presse, they can tailor the necessary levels of engagement and personalization across the entire customer journey. This is directed by analysts who access the web interface that lets them control the different parts of the customer journey (figures 3 and 4).

For each stage of the customer journey, there is a suggested course of action. During the journey, the focus is on user interest. If the data suggests that a user likes cars, then they’ll receive increasingly tailored, car-related content. This level of personalization is thanks to Snowplow, and has led La Presse to develop even more sophisticated custom tools to enhance the user experience.

Enhancing sell-side advertising offer

Another way La Presse uses Snowplow is in its sell-side advertising offer. In the same way as data informs the customer journey builder, data fuels La Presse’s segment builder tool, helping advertisers target and reach the right audience for their ads.

Snowplow tracks user interest and activities on the La Presse website and smartphone/tablet apps. The real-time user data is then fed into the segment builder. La Presse uses this to offer customized audiences as well as its existing out-of-the-box audiences to advertisers. The number of ad impressions has increased, leading to greater value on the advertiser side.

Why did La Presse choose Snowplow

• Endless tracking possibilities – tracking as many events as desired
• Ease of use – tracking new events is easy to implement
• Real-time capabilities – game-changing instant data, no more two to three hour delays, zero-latency use cases, allowing business proactivity in marketing and sales
• Competitive total cost of ownership – not billed by data points, and cost of support and infrastructure are independent and predictable
• Not a black box – full visibility and control, retention of complete data ownership and being able to integrate with existing AWS account

Results

With the data created by Snowplow, the sales, advertising and marketing teams have a clear picture of the La Presse readership. This is used to deliver personalized, targeted content that every single reader wants.

Snowplow has made a difference to La Presse in two key ways.

First, in delivering marketing automation and insights to tackle targeting of both ads and content to very specific audiences. The level of segmentation is near individual level. This has led to:

• A reduction in unsubscribe rates by finding the appropriate thresholds in ‘send frequency’
• Open rates (email and notifications) increased by 25%
• CTR increased by 33%
• Snowplow data and insights contributed to generating 2.5 million CAD in donations (‘contributions volontaires’) within a four-month period, thanks to advertising retargeting and tailoring efforts based on Snowplow data.

Second, in improving La Presse’s internal core metrics. La Presse tracks three key metrics daily: active users (daily, weekly, monthly, yearly), time spent on each platform, and number of sessions. This can be broken down further by article, by article sections, by demographic data.

Snowplow has helped to identify important data predictively, almost before it’s needed, for example:

• Highlighting potential signs of churn before the churn happens
• Proactively changing segmentation based on real-time data
• Improving the content La Presse offers both in general and on a granular, individual level

Each of these patterns translates into an action and a business result, such as boosting revenue for sales by delivering more precisely targeted advertising for advertisers or increased loyalty with the La Presse brand through relevant use of marketing engagement tools.

Having Snowplow helps the business. For sales, it can be an increase in revenue because of the audience we are able to provide; for marketing, better insight into users and how they can become more engaged.”

HERVÉ MENSAH | DIRECTOR, DATA SCIENCE & INTEGRATION AT LA PRESSE

How you can get started with Snowplow

To learn more about how Snowplow can empower your organization with behavioral data creation, book in a chat with our team today. Alternatively, Try Snowplow is our free, easy-to-use version of our technology, which allows you to create your own behavioral data in under 30 mins.

‍