Background

In the broadly commoditized mobile telephony sector, customers have increasingly come to expect more from their providers than simply an efficient and reliable network. Platforms such as Orange Wednesdays and O2 Priority have encouraged customers to demand rewards and recognition as standard from their providers.

T-Mobile has a reputation for great value contracts and deals, but is not known for offering its customers loyalty benefits.

We knew from research that T-Mobile's customers are savvy shoppers, but that they did not feel valued. It was this that we needed to address whether we were going to change attitudes and behaviour and improve T-Mobile's retention rate.

We were initially briefed at the end of January 2012 and our online loyalty programme went live in May 2012. In the 9 months to the end of 2012, we ran the programme as a pilot and, in that time, we were able to prove a positive return on investment.

Objectives

Our primary objective was to reduce churn among T-Mobile's Pay As You Go (PAYG) and Pay Monthly customers. However, we also had a number of other tasks: to increase customer value through product adoption (cross-sell and up-sell) and product usage and to capture customer data.

We were responsible for translating the overall business objectives into a quantifiable set of targets and key performance indicators (KPIs) that provided focus for the programme.

Challenges

Understandably, with a project of this scale, there were a number of internal and external challenges to overcome. One of these was that, for the majority (nearly 90 per cent) of PAYG customers, we had no contact details other than their mobile numbers. We needed to find a solution that engaged as many customers as possible to drive retention and value growth, but also one that encouraged data capture.

Another challenge was a requirement to integrate our digital loyalty platform with T-Mobile's established online account management service (My T-Mobile). This created significant technical and user experience issues that needed to be worked though.

But possibly, the most significant challenge of all was the limited available budget and timeframe for the pilot — and we needed to be able to prove the success of the programme before it could be rolled out. Therefore, we had to minimize our set-up costs and the rewards offered had to be, in the main, self-funding. We worked in partnership with T-Mobile and reward agencies to create a suite of relevant and cost-effective rewards.

Developing the strategy

Our strategic development began with an immersion phase that included a rigorous ‘deep-dive’ into T-Mobile's insight bank. We recognize that many brands have a wealth of useful and usable knowledge and data stored away, and that not every project needs to begin with a round of new research. The deep-dive unlocked some valuable insights about customer value cycles, the ebb and flow of sentiment and the crunch points in fostering loyalty and it was towards these points that we directed our activity.

For the pilot, it made sense for us to select customers for inclusion in the programme based on their length of tenure and their value to T-Mobile. We learned that it would be most beneficial to reward early tenure customers — it becomes significantly easier to retain people later in their lifecycle — and those customers nearing their contract renewal date.

Our solution: A digital loyalty platform that offers customers rewards based on their value and length of tenure. Knowing that we wanted to reward and recognize customers at critical lifecycle stages, we devised a programme that offers rewards regularly, but, as far as customers are concerned, on a seemingly random basis.

Creative strategy

Our creative solution brilliantly conveys the ‘surprise and delight’ programme strategy. Just like an old-fashioned ice cream van, the ‘Treat Van’ only ‘visits’ for a limited period of time and customers are invited to go and choose their reward. It is designed to create a flurry of excitement and a little instant gratification.

We created a colourful, light-hearted Treat Van experience that makes choosing a reward almost as fun as receiving the reward itself — something that was especially important, given the limited budget for rewards (Figure 1).

Figure 1
figure 1

 T-Mobile Treat Van ‘visit’ message

Full size image

Data — making the complex simple

The number of current and historic handsets, tariffs, tenures, products, interactions, usage behaviour and digital and social profiles is vast. Every customer is unique and should be treated as an individual. But we needed to find a way to simplify the targeting and data selection process.

Core to the programme was a decision-making engine built on a complex set of algorithms, which matches rewards and communications to each customer. The decision engine converts our objectives, data inputs (such as T-Mobile's churn propensity models, cross-sell and lifetime value models, handset and operating system, tariffs and product usage), rules and strategy to determine:

  • Which five to ten rewards to assign to each customer from a pot of approximately 25;

  • How rewards should be prioritized when displayed;

  • When in the customer's lifecycle they should be contacted;

  • Outbound communications channel (SMS, MMS, email, direct mail) and the optimum channel mix;

  • Which rewards to ‘hero’ in the communications and on the website.

Contact strategy

Every month, certain customers are invited by SMS, MMS or email to claim a reward. The communication they receive depends on the available contact data for each customer and their handset type. This communication is also personalized based on the customer's lifecycle stage.

In each communication, the customer is offered three types of rewards to choose from. The bank of rewards was developed based on the insight about the lifestyles and interests of our core target audience segments. But again, the selection of these is personalized to each individual:

  • Telephony rewards — for example, free picture messages or international calls;

  • Multimedia rewards — for example, apps, games and music for their smartphone;

  • Affinity or third-party rewards — for example, Amazon and Lastminute.com.

Evaluation

Evaluating the communications and the rewards themselves has been key to the success of the programme. This included measuring responses to communications, online behaviour tracking, My-T-Mobile registrations, reward redemptions and then looking at the longer-term impact on customer value and churn. Control cells have been utilized throughout to show incremental uplift of all our activities.

We have used a number of tools as part of our measurement and evaluation plan:

Online tracking

Real-time tracking allows us to understand how the programme is performing so that tests and changes can be carried out on an ongoing basis. Every stage of the programme is tracked, allowing us to understand responses, conversions and drop-offs across the entire customer journey and this data is fed back in to the decision-making engine. This includes communications, website user journeys, My-T-Mobile registrations and log-ins and reward redemption.

Monitoring reward redemptions in real time has also helped us to identify quickly their popularity and effectiveness and allowed us to make changes to the rewards portfolio, swapping out the less-effective rewards and boosting response.

Testing

We have carried out split tests in each wave to optimize the invitation communications, for example, subject lines, creative variants, chaser communications and the reward selections.

Quantitative research

An online survey was conducted using Fuel's bespoke research tool, Fuel Gauge, in order to understand and explain the results from the programme response data and the online tracking. Customers were contacted via SMS or email and invited to complete the survey.

This helped us to identify why people had not responded to their Treat Van invitation and the reasons for non-redemptions among responders (eg registration process, types of rewards available). As a result of the findings, we were able to make enhancements to the communications, the online programme and the reward selections.

User experience research

User experience research gave us further valuable, qualitative insight about customers’ interactions with the Treat Van. We worked with What Users Do to recruit Treat Van reward-eligible customers to participate in the study. Each respondent recorded a video and voice commentary of their Treat Van experience. The results supported the broader themes shown in the quantitative research, but also uncovered opportunities to improve detailed elements of the user experience.

Social media monitoring

We have been running social media monitoring to gain insight about the effects of the Treat Van on customer attitudes, both positive and negative, and to ensure that any customer service issues are flagged and managed appropriately.

Summary

The T-Mobile Treat Van was a supremely focused, agile and sophisticated loyalty programme delivered against very pressing deadlines. The programme strategy, communications plan, data decision engine, creative and website were all built within a 4-month period.

The pilot has run successfully with continuous testing in place, enabling us to refine and improve the programme. However, most importantly, we have seen positive results against our communication objectives of reducing churn and increasing customer value. We have been able to prove that investment in the programme is justified. This means that we are now in the process of expanding and enhancing the programme.

The project won Gold for Best Use of Data in a Digital Campaign at the DMA Awards 2012.