Guest Blogger

by Omer Artun, Chief Science Officer at Acquia

Few marketers today would disagree that one of the most important factors in achieving success in the field of customer experience is through the intelligent use of customer data. Gathering information from across the full range of channels to provide a unified view of customer behaviour creates a massively valuable repository of data which can be mined and analysed to inform enterprise marketing strategies.

Over the past year, customer behaviours have changed through force of circumstance, and if the pandemic has taught businesses anything, it is how to be innovative through technology. While in-person interactions have been severely curtailed, other channels have become heavily relied upon for customer engagement.

Every time a customer, or potential customer, interacts with a business or a brand, data is generated. Depending on the size of the business, this can equate to millions of pieces of information. How these are captured and treated can drive the success of a business: ensuring the efficient manipulation of these insights is the key to unlocking their potential. This is the role of the Customer Data Platform (CDP).

At its most basic, a CDP is used to transform this raw customer data from multiple and diverse channels into an organised unified view. A CDP must ensure that data entering the system is accurate and of consistent quality, in the correct format and that the data is complete with validated versions of the required pre-programmed elements – for example, an email or mailing address in a valid format.

Once this initial information is collated, cleaned and unified, the process of transformation and enhancement begins. Transformation might include adding tags to products that enable them to be identified as similar or cumulative purchases over time. Enhancement could include integrating relevant data from external sources, such as personal or business demographics or location information. This requires the CDP to have the capability to combine data from online and offline channels in the same database and build customer profiles that unify both data sets.

It is important to recognise the proliferation of channels. For online, these have extended from the usual website interactions, ecommerce, email, mobile apps and social media to include touchpoints via smart televisions and the increasing use of devices connected through the Internet of Things (IoT). These include, for example, fitness trackers as a source of health information. Depending on the business requirements, it is essential that the CDP can ingest data from all relevant sources: streaming data in particular may require specialised processing.  Offline channels include in-store purchases, promotional events, call centres, complaints and service calls. CDPs must be flexible and scalable in order to receive data from new channels as they emerge.

The CDP brings all this data together to form the Single Customer View, providing the foundations for targeted customer engagement campaigns and and equipping brands with a real-time understanding of their customers in order to create highly personalised content.

However, all this data is generated from actual transactions and actual interactions. The CDP sets out what the customer has done and how the customer has behaved – past tense. Can a CDP help brands to predict future actions?  Well, marketers are increasingly using machine learning and predictive analytics to anticipate customer behaviours, recommend actions and optimise customer choice.

CDPs that can support machine learning and predictive analytics give retailers and brands a huge advantage. By understanding a customer’s likelihood to buy, based on their profile and anticipated behaviours, marketing budgets can be used far more effectively to generate a higher return on investment in targeting, developing and executing campaigns.

Machine learning analyses existing data, automatically adapting its models over time as new information is added, and the processes happen with minimal human intervention. This is particularly valuable to businesses and brands with multiple products, segments, messages, campaigns and channels that are too complicated to evaluate and analyse through purely manual methods.

Predictive analytics utilise machine learning to provide brands with data that can inform their campaigns by highlighting what is likely to appeal to or resonate with particular sets of customers. Retailers can also use the information to identify and target individuals that share the characteristics of their most profitable customers, helping to drive revenue and increase profit margins.

CDPs can have built-in machine learning and predictive analytical capabilities, or these functions can be provided by external systems. The latter is often preferred by the personnel tasked with delivering the results of the analysis. If this is the case then a CDP should be chosen that allows connection to third party tools which can take the raw data from the CDPs for analysis.

A single unified view of customer behaviour – both past, present and predicted – is a powerful marketing tool. The CDP is the engine driving this, and it is critical that businesses and brands select the CDP with the capability to support their particular needs. Marketers need to set their goals and then define the functions required in the CDP to enable those goals to be met. This will ensure that expectations are set. For example, if customer data is always required to be up to date and ready to provide value in real-time if the campaign demands, then the CDP needs that capability. Data enters a CDP from many sources and undergoes several processes before it is formatted and accessible. The time taken to complete this can vary from a few seconds to a few days. The speed of the latter would not be suited to marketing based on real-time customer interactions such as presenting suggestions during a web browsing session for cross-selling and/or upselling.

Access to unified customer data is not just of benefit to the marketing teams. Operations and finance teams can use the information to evaluate which stores should be reopened once coronavirus restrictions are relaxed. Call centre staff and customer service advisors need to see a history of transactions and touchpoints to best support customer demands.

The power of the CDP is in enabling the business or brand to have access to a wealth of accurate, timely, clean and organised information on customer behaviours through a combination of factual and predicted actions. It is essential to choose the right CDP in order to maximise its potential for your business.

About the Author

Omer is the Chief Science Officer at Acquia. He officially joined Acquia in December 2019 when Acquia bought AgilOne, which Omer founded in 2006.  AgilOne is the Customer data and Engagement hub built for modern marketers as it allows them to analyse, understand and predict customer behaviour so that the customer receives personalised messages, offerings, pricing, emails to suit them. This is achieved through AI, Machine Learning and data management capabilities.

You may also like...

Keep Up To Date - Subscribe To Our Email Newsletter Today

Get the latest industry news direct to your inbox on all your devices.

We may use your information to send you details about goods and services which we feel may be of interest to you. We will process your data in accordance with our Privacy Policy as displayed on our parent website