There are many benefits gained by brands when they implement a customer decisioning solution to drive personalised engagement with each customer across all channels at every interaction.
A previous article (‘Customer Decisioning: Lifting the Bonnet to Explore how a Decisioning Engine Works’) explored the technical aspects of customer decisioning and stepped through an example to really delve into what’s going on ‘under the bonnet’ of the technology. With so much to be gained from leveraging customer decisioning to improve customer experience, it’s important to consider the common technical challenges, considerations and conundrums that may prevent brands from realising decisioning’s benefits in line with planned timescales and budget.
Clarify technology roles and responsibilities
Regardless of which business area leads the charge in implementing customer decisioning, it is imperative to have alignment across the business on the roles and responsibilities of the various elements of the martech stack. Are all business areas clear on which technology capability should do what, and why?
Successful customer decisioning implementation will bring significant benefit to digital, telephony operations, marketing, etc. to create significantly improved, personal customer experiences at scale. To succeed, however, it will require review and some flexibility from each business area on the methods through which their current customer-facing activities are delivered (e.g. content management, e-mail construction, mobile and web banner messaging, telephony prompts, etc.)
Too often business areas prefer to stick with their current practices, rather than agreeing the exact role for each element of the technology stack across the business in designing, building and delivering more sophisticated, personal customer journeys.
Gain alignment on data governance
Another area where alignment is key is between the data function, legal team, and data protection office on key areas such as preference management, the use of customer data, and clarification on the difference between marketing and service content. Having the legal team on board is key to being able to move forward at pace. It’s in the nature of legal teams to be risk averse and cautious with a data-led transformation programme – they’ll rightly want to avoid adverse customer sentiment, customer complaints, or challenge from a regulator.
Support your legal and data protection teams in understanding decisioning’s approach
Devote appropriate focus to data governance. Part of that process will likely include supporting the legal team and the data protection office in becoming comfortable with market-leading cloud technology solutions. This tech is already utilised by governments, the military, and leading financial services companies, and so the security level it provides is often far greater than other current internal applications. Nonetheless, it can take time for internal teams to get comfortable with the approach, and to collectively agree how and where personal identifiable information (PII) data is handled.
Think before you customise
Think carefully about whether to customise the decisioning platform – are there good reasons to be deviating from the out-of-the-box capabilities? You wouldn’t buy a sports car and then spend a significant amount of time retrofitting it to perform the function of a standard mid-range car. Maintenance and support can become more expensive over time as solutions become increasingly tailored.
Ensure consistency of definitions
A further watch-out relating to the technology includes ensuring there’s consistency on definitions – is there a clear understanding across the business of key functional terms? Is there clarity of language on the business requirements, such that there’s traceability for the functional build and subsequent testing of capabilities? Have the correct stakeholders reviewed, understood, agreed and signed-off requirements?
Carefully consider your data science and artificial intelligence needs
Real-time customer decisioning is decisioning at the individual customer level, utilising sophisticated logic working against the most up-to-date available data with artificial intelligence (AI) and machine learning techniques. Different types of models are commonly deployed, with the central decisioning brain continuously updating models and calculations using adaptive techniques, ensuring AI-driven propensity scores are utilised to create the very best customer experience.
With so much happening within the platform, brands often face a challenge around oversight and monitoring. These questions often cover model performance; consistency, sensitivity and removal of bias; the need to ensure everything is fit-for-purpose; and the dispelling of the business myth that the decisioning platform is some form of black-box.
Other considerations include alignment with internal standards of the data science team, demonstration of broader treating customer fairly (TCF) principles, plus adherence to local market regulations e.g. General Data Protection Regulation (GDPR), California Consumer Privacy Act (CCPA), etc. Be upfront and proactive in addressing these likely concerns before they arise.
Use the data you have to rapidly make a difference
One of the biggest dependencies will be on data, including availability of data items, data governance, data quality and data latency. For example, is real-time data required for decisioning to support the initial Minimum Loveable Product (MLP) releases, or is overnight batch update sufficient? Analysts can get fixated on data gaps and the need to have hundreds of data items available in the underlying data model. However, the truth is that the majority of data items will not be required to deliver the early MLP releases or initial customer journeys. As the American author Zig Ziglar said, though: “It’s not what you’ve got, it’s what you use that makes a difference,” so focus on sourcing the core data required to deliver value quickly.
Data is the fuel that drives the decisioning engine, but often senior business stakeholders will not differentiate between ‘delivery of data’ and ‘delivery of decisioning’. It’s imperative to communicate upwards that a delay in providing the data model will cause a delay to decisioning MLP delivery, which dents the reputation of the transformation programme. Decisioning is dependent on the provision of correct data to the required standards, and so this dependency must be managed effectively – having exec level support, plus alignment across the IT and data divisions will greatly help!
Our contribution to customer-centric excellence
In a previous article ('Customer Decisioning: Why it’s Crucial to an Unbound Experience'), we highlighted the significance of customer decisioning in enhancing the customer experience. By embracing a 'segment of one' approach and data-driven insights, we empower precise, timely, and personal customer interactions that build meaningful moments and experiences. With our help, businesses can deliver exceptional customer engagement in this competitive landscape, fostering enduring relationships.
Want to read more? If you missed our last Decisioning Congress in October, you can access all the valuable content on-demand from across the two days. Explore the five stages of the Decisioning Life Cycle and gain actionable insights to become a world-class decisioning business.
If you’re considering a complete transformation of your customer experience, get in touch with Merkle today to drive real, valuable results.