Patrick Hoare, Vanguard Consulting
I like digital services. If I can do it online and it works, I will. My phone and computer remember my personal details and my card details, so it’s quick and easy. And many customers think like me; banking transactions on smart phones overtook those done on computers for the first time in 2015. What’s not to like?
Well, I don’t always want to use online channels. Sometimes I have questions and want to talk them through. Take the example of applying for a credit card. For some customers, maybe the majority, applying online can be good for them and good for the bottom line of the bank. It’s a win for both parties. However, look at the end-to-end customer experience in the following case study and consider the findings that emerge.
The logic of the bank in question was, ‘Online is cheaper, so let’s make customers apply digitally.’ So that’s what it did. Customers who telephoned were referred to the website. What happened? This is what we learned.
Over six months, spurred by a successful, if expensive, marketing campaign, 183,000 customers applied for an account online. Of those, 61,000 (33%) failed the credit-score process, which is not unusual and easy to rationalise – who wants risky customers? That left 122,000, who after filling in the form online had to do – guess what? Print the form, sign it and send it in by post… Of the ‘successful’ 122,000 applicants, 86,500 actually opened an account. On further review (the system couldn’t understand everything about the applicants at the point of application), a further 17,000 of the 122,000 became credit-score fails. The remaining 18,500 (10%) began the application process and were accepted, but for some reason never completed. Maybe the tortuous digital journey was a factor.
Lesson number 1
If your customers are happy to go online, design the channel to be clear, easy to use – and properly digital. In customer terms:
- Tell me clearly the information you need to make a decision
- Explain the product’s features
- Enable me to apply, and get verified, online
- Make an instant decision
- State my credit limit
- Allow me to use the account straight away.
The channel may have been ‘cheaper’, but we were beginning to understand that it was also unfriendly and poorly designed. It had ‘turned off’ 18,500 successful applicants – was it hiding other problems and opportunities? The short answer was yes.
As we saw, 78,000 people who wanted a card, or 43% of the total, were deemed a credit risk, either immediately or after a delay. For many that was just the cost of reducing risk. But the total warranted investigation. We found that the way the digital system was constructed meant that some customers in good credit standing were being rejected. Unaware that special student cards were available, for example, students were being rejected when they applied for a standard card – then rejected again on applying for a student card for having made ‘more than one application within six months’.
If there had been dialogue we’d have understood this earlier – along with other damage done by the confusing and rigid application form. All told, thousands of perfectly creditworthy customers were being turned down, while others dropped out of their own accord.
Lesson Number 2
Don’t be lured into believing that computers can replace humans.
I have yet to see a matrix of decision trees that can replace dialogue and understanding. In every instance when they make this mistake, businesses end up losing money.
Still, we had 86,500 customers a year making it through the application system on a cheap channel – good news, surely? Up to a point. What we found was that the non-digital digital application system turned customers off. Whether through lack of clarity or a deliberate marketing ploy, customers had to read the features and benefits very carefully. The payback option, defaulted to the minimum per month, was another problem. So was the time – on average 15.5 days – it took for customers to receive their cards, which had repercussions on card usage. Two months after dispatch, 26% of cards had not been used. Shockingly, 18 months after the start date 87% of accounts had been closed or had never been used.
Lesson Number 3
Don’t try to trick customers through ‘smart online marketing’.
Customers aren’t fooled – as we would quickly see if our measures related to purpose from their point of view. The lessons of mis-sold endowments, PPI and packaged bank accounts are crystal clear: we should have learned them by now.
We thought we had a cheap channel of acquisition that ticked all the boxes for ‘going digital’. The reality was that we had a complicated system that put off customers, drove in failure demand and did not lend itself to building long-term customer relationships.
When we quietly abandoned the digital-only policy and started talking to customers, the following things happened. An extra 21,661 applications were accepted, which was a good start. As customers and bank gained a better understanding of each other’s needs, applications were cleaner, with the result that the time taken from application to card issue halved to 7.4 days. With the application was still ‘warm’ in the customer’s mind, time taken to use the card also fell significantly, from 24.8 days to 16.2 days. And more customers used their cards: the proportion failing to use their card in the first two months fell to 12.2%. While it’s difficult to draw an empirical conclusion, the hypothesis is that customers had a better relationship with the bank and consequently had a higher propensity to use the card.
To summarise, we learned three important lessons when we studied what happened when the bank made customers apply digitally:
Lesson 1: If your customers are happy to go online, design the channel to be clear, easy to use – and properly digital.
Lesson 2 Don’t be lured into believing that computers can replace humans.
Lesson 3: Don’t try to trick customers through ‘smart online marketing’.
And most importantly, don’t fall for the ‘digital by default’ narrative. It can work for some customers in some circumstances – but don’t diss dialogue.