EN
With technology slowly progressing, data is starting to be at the heart of every transaction, decision, and strategy within banking. Whether it’s recommending financial products to users or automating decision-making processes, banks rely on data to understand their customers and provide a seamless experience.
But what happens when the data is inaccurate? The cost is far greater than you might think. And that goes double for digital banking, where data inaccuracy can severely undermine the bank’s potential to grow and engage its clients.
Imagine if digital banking ran on guesswork. Without accurate data, every recommendation, every interaction becomes a shot in the dark. The more digitally native the bank the more it depends on clean, accurate data to function effectively. From transaction categorisation to user behavior analysis, accuracy matters. It’s not just about storing huge amounts of data; it’s about how precise that data is and how it informs everything from customer-facing services to backend operations.
One key area where data inaccuracies frequently surface is transaction histories. For instance, miscategorised transactions often lead to customers misinterpreting their spending patterns. A bank might label a €150 purchase at a local grocery store as a restaurant expense simply because the merchant’s payment processor uses an outdated Merchant Category Code (MCC).
This inaccuracy frustrates users, who initiate chargebacks because incomplete transaction details leave them confused, leading them to report purchases they simply don’t recognize. This process also generates additional operational costs for banks, as handling chargebacks can be resource-intensive and time-consuming.
This issue has even drawn the attention of Mastercard, which some time ago introduced a mandate which partly aims at improving transaction data quality to reduce these incidents. With digital banking apps thriving on customer engagement, these seemingly small mistakes carry big implications.
In the larger scope, inaccuracy directly limits a bank’s ability to leverage data for increasing revenue. Imagine the power of precise transaction data paired with AI-driven models that can make tailored offers. Banks can upsell financial products, cross-sell services, or offer relevant advice. But the foundation for all of this is accurate data. Inaccurate data leads to flawed recommendations, missed opportunities, and potential loss of customer trust.
Understanding client behavior is essential to building trust and driving engagement. The difference between a bank working with enriched data and one dealing with raw, unprocessed information is the difference between tailored precision and generic, outdated insights. Raw, non-enriched data generally includes only the most basic transaction details: date, amount, and a cryptic merchant description, often coupled with an outdated or incorrect Merchant Category Code (MCC). In contrast, enriched transaction data offers a more complete, accurate, and user-friendly view. Through enrichment, raw data is transformed into detailed, categorised insights that clarify transaction histories. This process involves cleansing and enriching records with relevant merchant names, precise locations, logos, accurate categorisation and much more.
With enhanced client financial profiles, banks can offer services like credit scoring based on smarter data, providing a more accurate picture of the customer’s financial health. Rather than just evaluating a user’s past transactions, enriched profiles consider broader behavioural patterns and demographics. This allows banks to fine-tune credit offers, ensuring they provide the right product to the right person at the right time.
Take a look at how non-enriched data can show a completely different lifestyle compared to enriched, accurate insights into payment history and spending habits of the customer.
Let’s look at several specific scenarios where data inaccuracy can have negative consequences for banks:
Cross-selling is a powerful tool for banks, allowing them to offer complementary products to existing customers. For instance, a user with high transaction volumes in travel might be the perfect candidate for a travel credit card. Some targeted campaigns can have a better conversion rate. However, when banks rely on inaccurate data, these campaigns fail. Imagine a user who primarily shops at local stores but is misclassified as an international shopper.
Suddenly, they receive irrelevant offers for international travel insurance or foreign exchange services, resulting in frustration and disengagement. The Dateio platform showcases how, when done right, targeted campaigns based on transaction data can increase engagement and drive higher conversion rates. But the prerequisite is data accuracy.
What does it mean?
Imagine you have 3 million users. Let's check the table to see how accurate data effect the outcome!
Each targeted campaign is expected to be relevant to 12% of users, or 360,000 people. However, with accurate data, you'd find that the campaign is actually relevant to 14% of users, or 420,000 people. If the campaign's conversion rate is 5%, then for the initial 12% relevance, 18,000 users would convert. With 14% relevance, 21,000 users would convert. If your revenue per user is €1, the additional 3,000 conversions from accurate data would lead to an extra €3,000 in revenue. Inaccurate data, on the other hand, would mean a missed opportunity for this additional revenue.
AI chatbots are increasingly handling a large share of customer interactions. However, without quality data to feed into these models, the entire system is pretty much useless. Chatbots require accurate customer information and context to provide meaningful solutions. If the data is inaccurate, customers might be directed to irrelevant FAQs or given faulty recommendations, which negatively affects trust in the service. As highlighted in our AI chatbots article, AI needs a steady stream of accurate data to function optimally. Without it, the efficiency and effectiveness of these AI tools are reduced.
What does it mean: AI chatbots rely heavily on accurate data to provide relevant assistance. Imagine your bank serves 500,000 users monthly through a chatbot, and 20% of these interactions are negatively affected by inaccurate data, leading to 100,000 unsatisfactory interactions. If each unsuccessful interaction represents a potential churn or loss in trust valued at €0.50 per interaction, then the total cost of inaccurate data in chatbot interactions could reach €50,000 per month. By using enriched, accurate data, banks can significantly reduce these costs and enhance customer satisfaction.
It's all about the details now, so banks are offering customers insights into how their spending affects the environment. For example, a carbon footprint. These initiatives allow users to make more eco-conscious decisions. However, when the data behind those insights is inaccurate, the value of such services goes down. Proper Carbon footprint insights are based on enriched data, allowing users to see an accurate reflection of their environmental impact. Accurate data makes these sustainability efforts meaningful and credible.
Model showcase - Based on the general EU model. Example of €50 spent: For instance, a €50 transaction under a general “transport” MCC might show a high carbon emission estimate, perhaps around 49.8 kg CO₂e. However, when the data is enriched to specify the merchant and transaction type, the real environmental impact becomes clearer. That €50 could be spent on a range of transportation modes: a ski resort visit with just 5.6 kg CO₂e, a bus ride at 25.5 kg CO₂e, or a bike-sharing trip, which adds only 2.3 kg CO₂e. This granular detail allows users to understand the ecological impact of their spending choices better - highlighting, for example, the significant difference between biking and taking a taxi.
Raw transaction data is often messy, incomplete, and difficult to interpret. Before you even begin worrying about accuracy, you need to ensure the data is usable. This is where data enrichment comes into play. Enhanced data provides clarity, context, and precision that raw data lacks. By processing and categorising payment histories accurately, banks can build a much clearer picture of customer behavior.
Through enriched transaction histories, banks can offer users comprehensive financial insights, eliminating the confusion and guesswork that raw data often brings. Open banking further amplifies this by connecting various data streams, but again, the key to leveraging these possibilities is ensuring that the data is clean and accurate.
At the end of the day, inaccurate data costs banks not only in terms of lost revenue but also lost trust and engagement. Leveraging enriched data ensures that your digital banking services meet the expectations of modern consumers while also enabling advanced technologies like AI to function at their full potential.
.webp)
Michal Maliarov
Senior insider
