In Data Customers Trust

A key priority of digital enterprises is managing data-related challenges, such as integrating data from disparate sources, and managing data quality, privacy, and security. But with the rapid rise in AI adoption, leaders must elevate the pitch from ensuring data compliance to building data trust. This is critical, because when all organizations use AI-based automation to become equally efficient, ethical and responsible data practices will feed their sources of differentiation.   

A sound data and AI economy rests on five cornerstones of trust, ethics, privacy, compliance and security. Building data trust is about ensuring the right data attributes -- quality, accuracy, conformance with standards, clear sources and ownership, and so on. When an organization trusts its data, it can act upon its insights with confidence. But with AI in the picture, organizations need to take a broader view of data trust. Based on not just the attributes of data but the way it is applied. Considering not only its impact on the organization but also on customers, partners, and other stakeholders. Seeking not only to create data that can be trusted, but an organization that their customers will trust by underpinning their data foundation with ‘responsible by design’ principles. Effective data governance enables enterprises to reimagine their offerings to create unique experiences and value through domain data hubs and new business models. In doing so, data becomes the hub for creating new frontiers of efficiency and decision-making.

Here are three ways to go about it:

1. Show consumers that they can trust you with their data

First and foremost, enterprises must earn their customers’ trust to get them to share their data. Start with building data health, that is, ensuring your data is clean, consistent, complete, and free of error.  Establish data lineage that shows where a piece of data is at any given time, where it came from, and where it is going. Apply permissions and policies to control who can access or modify various types of data. The addition of metadata and business logic provides context and a clearer understanding of data and makes it easier to search and discover particular data, such as sensitive information.

Clearly communicate to customers how their personal information is kept safe and anonymized to protect privacy and confidentiality. A major reason why people do not share information is that they are apprehensive about losing control; organizations can address this by allowing consumers to customize cookies, delete their data from the company records, and opt out whenever they wish.  

2. Show them the value of sharing and have an open-door policy.

Knowing how their data is being used, and to what purpose, can improve consumers’ confidence in sharing information. Therefore, enterprises should make sure that they have an open-door policy when it comes to ingesting and using customer information and be upfront about why they need the data, and at what points it is being collected. The opt in-opt out mechanism will not only enable consent but will build trust and confidence that they still own and control the data.

Above all, they need to offer customers real value in exchange for their information. A recent PwC survey found that about 80% of US consumers would share some kind of personal data for a more personalized service or improved experience. Customers are more likely to trust a company with their information if they see that they can save money, spend less effort getting the job done, receive only relevant marketing messages, or enjoy entertaining, enriching experiences. Large organizations like ours delivers best practices and help clients build privacy practices to sustain compliance in the ever-changing data privacy regulatory landscape across the globe. For example, we helped an American multichannel video programming distributor define and implement a privacy by design process along with necessary controls to ensure data privacy compliance throughout the software development process integrating consent collection and proper processing of data based on available consent.

3. Do the right thing with data ethics for all

One major barrier of human trust to adopt data driven decisions lies in being able to understand the chain of facts or “reasoning” that led to the output.  A faulty training data set is usually what causes incomplete or skewed and biased outcomes or erroneous outputs. Such events can cause irreparable reputational damage to enterprises and cost them the trust of their customers. It only takes one breach to cost an enterprise the hard-won trust of their customers.

Companies that become relevant and sustainable this decade, will be the ones that ensures they employ policies and procedures to guarantee that their data assets are used for the good of humankind; that they are devoid of factual errors, privacy violations, and biased outcomes; that they have transparency and control for the customers who trust them with their data.

They should take a human-centric approach when using data. Ethical use of data happens when the enterprise never loses sight of the people it is serving.