How KLM activates data to deliver hyper-personalized customer journeys

As the flag carrier airline of the Netherlands, the vision of KLM’s marketing program is to humanize advertising through personalization and relevance at scale. Supported by a data-driven approach, KLM wanted to address the proliferation of fragmented channels and develop smarter, more effective media buying models. By combining data to connect every touchpoint of the customer journey, their aim was to serve more relevant advertising and improve media spend.

Laying the foundation for the future - a connected business architecture

The following three KLM journeys require a smart marketing hub to seamlessly connect all customer channels and internal business systems — present and future. They chose Supermetrics (formerly Relay42), which addressed the prerequisites for unifying technology.

1. Flexible fit, plug-and-play capabilities
The platform needed to be a seamless fit with all of their varied and existing best-in-class systems and channels. This ensured complete compatibility by design, allowing them to remain disruptive in their marketing without compromising their IT framework.

2. Faultless accuracy in knowing the customers
To sustain one-to-one conversations at scale, the platform must identify people — and their data trails — across the KLM marketing landscape, regardless of context, to keep their communications 100% relevant.

3. Actionable data in real time to automate dialogue
It follows that KLM needed to add actions to its existing analytics. The platform must orchestrate data in real-life moments rather than in bi-weekly batch data exports. This makes their journeys frictionless and keeps their customers happy.

4. Future-proof, for zero channel lock-in
In hacking their future responsibly, KLM wants to be able to try new ways of reaching their customers without a complete system overhaul or having to wait for development. This way, the marketing team can see what works while it still matters, not only to remain ahead of the game but also to give ourselves the room to change the rules completely.

The marketing ecosystem

Optimizing the media spend

Using a data management solution for smarter decision-making across paid and owned channels. KLM, wanted to achieve the best of both worlds by delivering incremental business value alongside serving relevant content. They started by using a single customer profile to map movements across biddable media and owned or internal systems and channels, and then created cross-channel rules and logic to alter bidding and targeting per customer, depending on their likelihood to book a flight.

Case A) If a customer converts on one channel or their departure date has passed, stop retargeting across all platforms
Not only is KLM able to orchestrate bidding across the digital landscape, but vice versa. If a passenger converts offline (e.g., call center), they can stop focusing their resources on sending them messages across every channel. Real-time automation can also be extended to flight dates: if a potential customer has searched for a destination through KLM but their departure date has passed, they know the ads and one-to-one messages are irrelevant, so they stop targeting.

Case B) Alter bidding resource allocation based on customer value
KLM utilizes value parameters, such as time to departure or search history for higher-value destinations, to control and shape its bidding strategy. This means they can increase or decrease bidding on channels, such as search, per individual based on their previous interactions with us and their likelihood to book via other channels, like their website or email. This includes using their Revenue Management System as a medium to feed Google Search with relevant data. By using the plane fill rate in combination with customer value, they concentrate their resources on passengers with higher intent.

For customers? This means fewer incessant ads and more content at an appropriate frequency, depending on their intent or urgency to book.

Case C) Personalize display and search content to maximize paid conversions
KLM's team can also use search behavior and interactions from their own channels to show relevant content across paid media, and vice versa. Say a customer is searching for flights to Spain on the website. If they have clicked through to a promotional offer via an email, they don’t only alter their bid based on key terms or relevant publishers – they alter the content they show, too. By dynamically updating content across display banners and search ads to reflect Spain and pre-fill dates, the team significantly increases their relevance and, hence, their conversions via paid channels.

Now, passengers are not only seeing relevant ads at the right time, but ads containing exactly the right content for them.

Case study results:

• Significant reduction of media waste
• 5% conversion rate increase

Orchestrating 1-to-1 marketing to many

Using a marketing platform to trigger messages that make sense by controlling sequencing and frequency

In digital advertising, accuracy and individual relevance are highly beneficial. When marketing via one-to-one channels, accuracy is crucial. At KLM, they use trigger-based logic to determine the best place (channel) to deliver a sequence of messages (based on which action, and in what volume?) to a specific segment of customers (in what context?). Then consider elements of value and how you can use them to determine what content you serve to each customer, when, and how.

Case D) Journey orchestration
Segment X of KLM customers have all opted in for email communications, have downloaded the KLM app, and have expressed an interest in a flight departing within the next month. These terms qualify them for a campaign for retargeting via email and push message.

This is how KLM could orchestrate their journey, along with the logic they might use.

Case E) Innovating for customer convenience with ‘Mail my Search’
When KLM's team considers innovation in marketing, they often leap to the need to add something new, purely for the sake of inventiveness or novelty. So instead of over-analyzing the reasons why a customer sale gets interrupted, they created a flexible and functional shortcut so passengers can browse, consider, compare — pause — then continue their purchase where they left off. We added service, and in doing so, we integrated our mobile and web channels.

Imagine a potential customer is about to leave KLM.com, having searched for a destination. The Data Activation platform triggers a pop-up when the customer’s email is not already in the KLM database. The customer is emailed their search, which then anchors their progress in the customer journey across devices so they can continue their booking where they left off.

Bonus : This significantly increased the capture of relevant contacts for KLM within the test market.

Tip : Balancing structural experiments with broader innovation strategy. Every experiment KLM conducts needs to be measurable. That’s why, at KLM, they always benchmark their tests against control groups within the Data Activation Platform. The results of these tests are then pushed to their chosen analytics platform through a server-to-server connection in real-time.

Case study result:

• 25% increase in revenue from the "Mail my search" initiative (customers vs control group).

Increasing customer engagement through hyper-relevant targeting

How can KLM’s marketing team qualify intent and interest to ensure what is offered truly makes sense to them? Where KLM identifies an overlap between entire third-party audiences (such as people who are in the market for travel) and individuals who have interacted with KLM before, the magic happens in terms of relevant outreach. KLM’s marketing team can leverage technology with precision identity management capabilities to turn the winning combination of intent and historical behavior into actions, results, and real-world impact.

A third-party audience of people interested in flying to Paris is compared with a segment of existing KLM customers, and an overlap of individuals is identified.

This valuable overlap segment is then used as the basis for a personalized email campaign. The email content is tailored according to each individual’s travel intent for Paris and their chosen dates.

Here, KLM can deliver hyper-personalized outreach based on technical matchmaking and simple, effective logic. By cross-pollinating data between trusted partners and sources across the KLM landscape, this journey delivers a triple win in terms of business value and customer service: the right message, to the right person, at the right time, and in the right context. This contextual intelligence enables smarter budget allocation and establishes a preferred method of contact from an airline the recipient already knows and trusts.

Case study results:

• +800% above average value per sent email
• +200% email open rate ("Mail my search")
• +170% increase in click-through rate
  

Case F) Activating data in real time with Google

KLM wanted to address the proliferation of fragmented channels and develop smarter, more effective media buying models. By combining data to connect every touchpoint of the customer journey, their aim was to serve more relevant advertising and improve media spend.

Goals:

• Develop smarter, more effective media buying models through data
• Drive relevant advertising
• Scale predictive modeling across all touchpoints in the customer journey

Approach:

• Combined contextual data to create a predictive model with granular layers
• Activated data in real time

Relay42 and KLM created a data flow setup with the Relay42 platform at the core. Thanks to Relay42’s tag management system (TMS), all customer interactions on KLM’s website and app, as well as relevant indicators from other channels and data sources such as email, social, CRM, call center, and affiliates, can be tracked and synced in Google Analytics 360.

Data gathered via Relay42’s own tags and DoubleClick Floodlight tags can also be sent from Relay42 to the DoubleClick platform so that relevant ads can be targeted to appropriate audiences.

In order to perform deep-dive analyses, KLM simply exports raw-level data (from DoubleClick using Data Transfer and from Google Analytics 360 through seamless integration) to BigQuery. BigQuery can then correlate site behavior with ad impressions, which enables Relay42 to activate the data and inform the build of predictive models.

A Predictive Model to Improve Display Remarketing

KLM decided to test a new approach to remarketing. “Our goal was to get rid of irrelevant ads, as they are simply annoying,” explains Kevin Duijndam, the airline’s Program Manager for Personalization. “Our assumption was that people who fly with us often already know us, so it would be irrelevant to tell them about flying with us again. However, we were wondering when exactly someone becomes a ‘frequent flyer.’ The more we thought about it, the more complex the set of business rules became, so in the end, we realized we couldn’t just focus on frequent flyers, but instead should use machine learning to understand when ads are irrelevant.”

KLM developed a real-time buying setup to include predictive modeling. In this setup, website and app interactions are tracked in the platform thanks to the Relay42 tag management system. Relevant consumer behavior can be streamed in real time to a prediction engine developed by KLM on the Google Cloud Platform, with the outputs then streamed straight back to the Relay42 platform. From here, Relay42 can activate rule-based segments based on the prediction outcome. And by syncing this with DoubleClick, ads can be served and targeted to maximize relevance.

The team tested their new predictive model to assess any gains in performance. KLM deliberately chose to measure the results in an A/B setup within a defined period rather than measuring the differences year over year or month over month. Such comparisons are less reliable due to rapid changes relating to seasonality, internal capacities and external factors caused by competitors.

Business Gains and Customer Experience Wins

In the test, KLM linked the Relay42 customer interaction data to their predictive model to predict how relevant their ads would be. The setup enabled a decision to be made in real time whether or not to serve a specific ad to a user.

Case study results:

Through the A/B tests, it became clear that the new model indeed generated a significant uplift in bookings. With the cost per booking 40% lower during the test period , KLM was able to achieve more than twice as many bookings at the same spend.

The test produced wins in terms of customer experience, too. The click-through rate for the test group was more than 1.4 times higher than for the control group, indicating that the new model was successfully reaching users with messages they found to be relevant rather than annoying.

The success goes beyond improving KLM’s display remarketing efforts: “Even more importantly, we’ve laid the IT data flow foundation in such a way that KLM is now able to execute on our data through all of our digital marketing channels and apply our prediction models at scale,” Kevin says. “So we can be flexible to plug in other models, but can also now scale through other online media channels like search or video.”