Over the past decade, two technology solutions have transformed data-driven digital marketing: data management platforms (DMP) and more sophisticated customer data platforms (CDP). CDPs have proven extremely popular and effective because they are capable of ingesting data from multiple online and offline sources and finding common identifiers to create single-client views. There have been peaks and valleys in the success rates of CDPs in delivering on the promise of a single customer view – some have only been able to ingest data online, others have not been able to cope with real-time data. Many of the failures have occurred because when setting the goal of creating a single customer view, companies have not considered what else is required outside of the CDP implementation. In 2019, Gartner even predicted that the CDP will hit the “depression of disappointment” for the next 5 years, until 2024.
Think big, think business
The first question to ask yourself when considering taking on the challenge of creating a single customer view is “what else do we need?” This question is twofold: What else should we, as the CDP commissioning team, be able to do once we have a single vision of the customer? And what else does the business as a whole need that can be achieved through the CDP?
For example, if you want to measure the success of marketing triggers in terms of customer retention, you will need to ensure that you have ongoing customer data flowing into the CDP, which may require the involvement of your IT or customer service team. .
Another reason that CDPs don’t meet the goal of creating a single client view is because a goal is set, as in the example above, but there is no use case attached, that is, “what’s next” has not been defined. “. This creates a situation where the technology that has been selected and implemented may not be able to support data activation in a way that offers tangible value.
Standard use cases to get started
Here are some relatively basic use case examples for a CDP that enable brands to improve their marketing ROI.
Deletion of existing clients
When setting up a new marketing campaign, a CDP can be used to ensure that existing customers are not targeted. This is accomplished by using deterministic identifiers that identify existing customers online and ensure that they are not exposed to certain campaigns. This means that the entire campaign budget is used to attract new customers. We worked with a brand to implement this strategy, which resulted in a savings of £ 3 million in 12 months.
With a CDP, marketers can identify customers who have not been active for a specified period of time, for example, two weeks, three months, a year. These users can then be targeted and targeted through new campaigns, across multiple channels, specifically designed to bring those customers back. This can be an especially useful tactic as it targets consumers who you know are already interested in your brand or products, they just need to remind them!
This involves using different targeting rules based on customer behavior across multiple channels, leveraging real-time feedback loops to create personalized messages based on customer engagement and segment. The key benefit of this approach is that, like a musical orchestra, when all the channels work together, you get the best result. In marketing terms, this translates to higher customer engagement and improved lifetime value.
This is where we start looking for broader business applications. The ability to stream data, in real time, from the CDP to business intelligence (BI) tools enables multiple teams to access customized and automated reports. The benefits of this approach are multifaceted: marketers can optimize campaigns in real time against business goals (not just marketing goals) and BI teams can identify areas to improve other aspects of the business, for example purchasing and order fulfillment.
Advanced use cases for you to think about
Second-party data alliances
Building partnerships with other brands and associations allows you to target common audiences, as well as identify new audiences that share similar characteristics with your top performing audiences. Additionally, data security and brand security are not compromised as all partner data is stored in multiple data warehouses. This is a great way to earn incremental customers and revenue, as well as gather new insights for future campaigns.
Anti-abandonment leverage machine learning
Adding machine learning (ML) to a CDP enables the integration of custom ML algorithms and functions that analyze the characteristics of customers who churn, score, and use the learnings to predict potential new churn. Armed with this information, marketers can activate campaigns specifically designed to retain customers most at risk of churn. This, in turn, provides more information on the best messages and channels through which to avoid rotation between different audiences. We have seen a reduction of up to 20% in turnover with this approach.
Additionally, ML can be used to identify when someone is likely to buy a product and that information can be used to send a marketing message designed to get them to go overboard and make a purchase.
One of the main benefits of a CDP is the ability to create AB tests on site, which means multiple combinations of channel, creativity, time, demographics, etc. can be tested in real time. This rapid delivery of insights means that marketers can optimize campaigns within hours of launch, significantly improving performance and reducing operating costs.
When all marketing and customer data is centralized in a CDP, it is possible to unlock information about consumer behavior and interaction with advertising at each stage of the funnel, in real time. This enables marketers to react quickly to changes and optimize an individual’s advertising experience to maximize the likelihood of conversion. On a large scale, this translates to increased conversion rates and revenue.
What’s next for you?
There is no denying that CDP has enormous potential, not just through the use cases in this article, but many, many more. If you already have a CDP and are looking to expand the functions you perform today, or if you are thinking of implementing a CDP, the most important thing is to start at the end … what do you want to be able to do? Don’t get caught in the trap of saying or thinking “I want a unique view of the customer.” Once you have this clearly defined, you can work backwards and identify what you need to do to make it happen.