It is no surprise that the Indian economy is more of a cash-driven market, especially in the e-commerce sector. Unlike the e-commerce industry in USA/Europe, COD is the forerunner in the Indian market. 

COD suits the Indian mindset and can make up to 70 percent of Indian e-commerce businesses. With smaller players, customer credibility is also in question, which can further accelerate the need to introduce COD to their business.

However, there’s a bigger problem in question when it comes to the Indian e-commerce industry– Return to Origin (RTO) costs. These RTO costs can be especially high in the case of COD orders.

What is RTO?

RTO is when orders cannot be delivered and have to be shipped back to the warehouse. This puts a significant cost burden on e-commerce firms as they lose a lot of money in shipping it back and forth.

Here’s how e-commerce companies lose money in these orders:

  • Forward & reverse logistics
  • Blocked Inventory (Items stuck in transit)
  • Physical quality check and re-packaging of returned items
  • Increased probability of damage to fragile items, and hence more money spent in shipping them
  • Operations cost in processing this order

We took time out to see what the actual numbers of RTO orders and what their share was. Here’s what we found– in case of COD orders, the percentage of RTO orders can be as high as 40 percent!

This means that at least one out of three orders were failed orders and were returned. When one-third of your orders have the potential to damage your bottom line, rather than adding value to it, it’s no doubt that the situation is worrisome.

Is there a pattern to these cancelled orders?

We took a closer look to see if there were any patterns to these orders, and if these patterns followed a Modus Operandi and we discovered a few interesting insights. Here’s what we found out:

  • Customer error (Intent is there but incomplete address, phone number, etc)
  • Orders from transitory addresses (hotels, friend’s place, etc)
  • Price-sensitive intent (Reorder because of drop in price)
  • Impulse buy but without paying (there is no downside to refusing delivery)
  • Intent to fraud (Habitual fraudsters)
  • Placing an order without any genuine intent

So, what is the solution to this?

Companies often perceive these costs as “mandatory” since there’s no proper solution set in place. Companies have little choice and fewer tools to prevent RTO — they just take it as a ‘cost of doing business’.

Some businesses also resort to static, generic solutions like the following:

  • Blocking all transactions on International credit cards
  • Not delivering to certain pin codes or cities
  • Capping the order size

But, what’s wrong with static solutions?

Well, sometimes, static solutions can do more harm than good as many genuine orders are lost in the process, not to mention customer dissatisfaction when they hit a dead end on a static solution. This can even affect customer relationships on a long-term basis. 

Solving the RTO problem by manually scanning every order does not work either due to the sheer scale of the problem and evolving nature of fraud techniques. 

With the Indian e-commerce market becoming hyper-competitive, firms need better solutions as they cannot afford to lose customers and orders. 

The way forward

Machine Learning technology offers an attractive solution as it addresses all the challenges in preventing fraud — scale, complexity and changing patterns.

  • Employing Machine Learning for fraud detection

Catching digital frauds requires us to first gather the ‘Forensic Evidence’. Every user interaction leaves behind a subtle digital forensics trail like proxy IP, device ID, email address, time to order, etc. 

Machine learning models combine hundreds of such innocuous parameters, which are seemingly unrelated, to identify the patterns that indicate fraud. These patterns are later used to zero down on customers who perform a fraud across different websites and make it to the blacklist.

  • Enriching the data

Machine learning and natural language processing are used to differentiate between real and fake address. This is only the beginning. Transaction and user data can be enriched by adding context to it.

For example, by adding the price of the user’s phone device or categorizing an address as five stars or one star, we turn meaningless data (phone model) into actionable information that increases the accuracy of the red or green flag that the machine learning models generate for every transaction.

  • Observing the user

Fraudsters are habitual in nature. They leave similar footprints on multiple sites. Network effects can be harnessed by pooling in anonymized data to predict and prevent fraudulent behaviour. This de-incentivises and penalises fraudulent behaviour across the ecosystem.

Moreover, e-commerce firms will truly know their customers so that goods are delivered to a person not merely to an address.

Most importantly, RTO will no longer be just the “cost of doing business”.

Curious to know more about how we’re solving this for merchants? Get in touch with a Thirdwatch expert today!