Data is often called the ‘new oil’ and enterprises around the world have been scampering around for innovative ways to collect data. With technologies like IoT, Blockchain, Machine Learning and Artificial Intelligence advancing rapidly, data collection, storage and analyses is a cake-walk. Companies can obtain a lot of data unobtrusively and without actually asking you for it. A lot of companies also record your data with your consensus through means like feedback forms. Many companies also purchase customer data from sources like third party data vendors. However, it is necessary to not use this information for any kind of malpractice and not violate anyone’s privacy. Customers feel their privacy is violated when the information is used out of context and without their consent, raising the question of privacy and data security measures taken by organizations.
Why is customer privacy important?
Privacy not only matters to the customer but also to the brand, organization and the industry. Any malpractice is this aspect can tarnish the image of the brand and hurt the customers’ trust. This would not only affect the brands’ standing in the market but also make the customer suspicious about the other industry players.This would further lead to businesses and the industry losing out on the existing reputation, market and revenue streams. Hence, ensuring consumer privacy not only helps businesses grow, but also creates room for better revenue streams, innovation and development in the industry.
The latest Facebook-Cambridge Analytica was one such incident where the violation of customer privacy not only affected the general masses and the organizations involved, but also raised questions about one of the most powerful governments in the world. The personal data of up to 87 million Facebook users, mostly in the U.S., was obtained by the analytics firm, Cambridge Analytica. This data has reportedly helped elect President Donald Trump. While the reports on this incident might or might not be completely true, it has certainly created skepticism in the minds of the people all around the world.
Current approaches to data security and its drawbacks
One of the most common approaches of protecting user data is by pseudonymization i.e. processing it in a way that the personal data can no longer be mapped to a specific individual unless additional information is provided. Cases where this can fail include mapping of data from multiple sources. Customer activities recorded at specific points of time in a particular location, combined with the personal details like age, address and choices in products and services make a unique combination that belong to a specific data subject. This not only violates the security of an individual but also makes it difficult to hold one enterprise accountable since the current laws do not check for data leakages due to anonymizing of data.
Additionally, corporates fail when their data storage systems are hacked into and tampered with. Since technology is advancing by leaps and bounds, there are innovative ways to gain access to any network despite strong security protocols. This needs to be countered with stronger security systems and network protocols.
How do we protect consumer data?
One way of ensuring data security of consumer information is by introducing synthetic data along with the actual data. This can help an enterprise release their data without being concerned about jeopardizing their customers’ privacy. Several methods of synthesizing data are:
- Adding random noise in the actual data: Companies can add random records and dummy observations in their data so as to keep the aggregated numbers equal to the actual numbers and still protect the granular records
- Rounding, Top Coding and Aggregating: Data can be broken into deciles or quantiles and can then be rounded, averaged to the closest actual data. Top coding is a way of manipulating data by capping it at an approximate maximum. For instance, any observation with sales of greater than 500 units can be capped to 500 in a set of sales that ranges from 10 to 600. This method just presumes that any kind of approximation of data will be reflective of the actual numbers. Rounding and aggregation can be done within the same customer demographic set or the same market.
- Creating synthetic data: The statistical measures of the actual data can be computed and data can be created keeping the distribution of the data intact. The measures will be keep the same and will be reflective of the actual scenario. This way, the actual consumer data is not let out and the numbers are still available for industry or brand review.
- Swapping data records: While keeping the details intact, the actual numbers can be swapped within the records. This way the numbers will still be the same, but will not be used to track the details down to a specific individual.
All these measures can help an organization ensure data security and still be a competitive entity in the industry. Additionally, with the cutting edge technologies springing up every moment, enterprises need to adopt competitive encryption algorithms and security protocols to protect their consumers’ data. Blockchain technology is being adopted across industries for decentralized and distributed data storage. Blockchain can also enable organizations to limit data access to others by devising single sign-on interfaces for people who want to use the data. It was also eliminate scope of tampering or losing data since the network can store data permanently.
While all these methods can help companies in ensuring data security, a few ethical and behavioral changes are required to make it completely safe for their customers. Companies should pull data that is necessary and not everything else. It is necessary to ensure that data that is no longer necessary is destroyed and not made available to any other entity without the customers’ consent. Customers should be informed that the data extracted from them is safe and will not be mishandled.
All these practices together can create a better consumer-producer relationship. It is necessary that such good practices are standardized and adopted across industries.