MMT DoubleBlack - A Loyalty Story

What follows is a description of the genesis and launch of MMT DoubleBlack – the subscription program from MMT that got lots of love in the few years it was around.

We needed new ideas to improve Flights NPS…

In mid-2017, our NPS for the domestic flights business had reached pretty high levels.?

Booking experience had improved big time – payment success rates were high, and failed/lost payments had reduced dramatically. Although Our fare/availability failures were still too high for our comfort, work was under way to get them down 80% within a couple of months. Fulfilment had also seen significant improvements. Almost 100% of our tickets were delivered to the customer inbox in close to real-time. Support Experience had seen a transformation as well – call center contacts were down 40-50%, escalation rates had reduced 70-80%, and CSAT had improved as much as 30-40 points. Instant Refunds, an industry first, was a done deal. Proactive refunds was in the works.?

We were running out of issues (reasons for detraction) to fix. (The only substantial one left was related to airline cancellation charges. But we largely ignored that- these were airline charges, not ours. We would have to live with them.) (To add context, airline cancellation fees were about Rs 3500, when the average fare was just about Rs. 5000)

The ‘easy’ NPS journey was over. We now needed to focus on increasing promoters – move neutrals to promoters.

(If you're not familiar with NPS, here's quick intro: Wha'ts NPS?)

The first set of ideas were relatively easier – let’s look at the customers’ travel journey in depth, find gaps, and see if we could add value. We found quite a few gaps between the time of booking and actual travel. The good thing was that these were not gaps that customers would naturally want their travel agent to fill, so if we could fix those, we would get some enduring customer love.

These were things like a seamless web check-in experience, and pre booking of seats and meals. We also did some work on alerts to travel to the airport (google maps started doing it soon after), and destination weather. There were lots of other such interventions we were thinking about– tracking flights and alerting customers to delays, pre booking meals at the airport, etc.

These would all help, but were each individually small. Some of these were already available with airline-direct channels. And other OTAs would catch-up.

We needed something bigger.

MMT Promise was one.?

But we wanted more. Something that was difficult for others to emulate.?

But how? Creating delight wasn’t easy, specially for a commodity business. There were only 3-4 major airlines. Domestic flights were largely point-to-point. So you couldn’t do much in the core product.?

Commissions were extremely low or zero. So there was no scope to use price as a differentiator. Even if there was, that’s not really sustainable, because the cost side levers were pretty much the same for everyone.

… Our share of wallet was low…

Separately, we had been wanting to show a linkage between NPS and core business/financial metrics. All the experience work was useless unless it resulted in improving some financial metric ultimately, right? Experience, in theory, would do that through two levers – higher share of wallet bringing more business, thus increasing customer LTV, and customer advocacy bringing in more customers at lower cost.

As part of the work we were doing on that count, we discovered that our share of wallet for our domestic flights customers was pretty low. (Share of wallet - MMT's share of their total spend on flights) Despite having a very high market share! This meant that a huge chunk of the travellers were using us, but were also booking elsewhere. It did look like customers would sometimes be tempted to book elsewhere based on limited time discounts run by other OTAs. It also meant that we were always under threat by some new deep-pocketed entrant coming in and spending money to take customers away. While we had already built substantial experience-led levers, clearly, we needed more.

Something that would get customers to make all their bookings with us. Incremental stuff wouldn’t do. We needed something dramatic. And something that wasn’t easy for others to copy.

So we looked at how we were different/what our competitive advantages were? Could be?

• Strong brand
• Great end-to-end transactional experience
• Fantastic support, also backed by the MMT Promise guarantee

These were already in place.

What others that we had, but were not leveraging in any manner at all?

• Our scale and high share of the total market
• The huge amount of data we had access to.

Surely, we could do something with this, without engaging in anti-competitive behavior?

Starting to build the outlines of a program…

Because the product was a commodity, the business was prone to competition using discounts as a means to take market share away, at least over short periods of time. It was clear that whatever we did, we needed to move from a transactional model to a relationship model with our customers. When you have a relationship, transactional level interventions by competition are less likely to have an impact.

So we started looking at a membership model – customers who would become members would get certain additional benefits, but they would get them for ALL their bookings with us.

And we had to look for things that would be higher value to them but lower cost to us. In theory, there were all sorts of possibilities. We could predict demand for airline paid meals, get into some sort of arrangement with airlines where we bought in bulk at a discount, and then passed on these discounts to our customers. We would be taking inventory risk, but good data analytics should cover us for that risk.

We came up with a list of such benefits (airline meals, premium seats, various hotel benefits-? free upgrade, discounted meals, early check-in etc…).?

As we started socialising our thoughts, we got the feedback that this wasn’t good enough. This was almost like a coupon book – basket of benefits. That’s a bit weak- difficult to sell. We needed a headline offering – something that is game changing, and catches people’s eyes.

…And it seemed we already had something in our stable!?

Cancellation !!! While we hadn’t really taken the cancellation fee detraction very seriously, we had been trying to solve that problem anyway – from a different lens, that’s all.

We had been experimenting with and selling a cancellation cover product called ‘free cancellation’ – for each flight you booked, you could pay a slight premium to cover the cancellation fees. It had reasonable success, but had fundamental challenges – the product carried greater risk and therefore higher price- (people would buy the insurance only if they already knew their plans might need to be cancelled).

Was it possible to create a program around this concept of cancellation cover? If we could spread the risk of the ‘free cancellation’ product over ALL of a customer’s flight bookings, the risk would be lower, and therefore we could reduce the price.

And it would act as a strong loyalty factor- why would a customer book elsewhere if each MMT booking was protected for cancellation charges?

Could this be our hero offering?

Modelling the P&L and risk

There were two revenue levers to cover the risk – a small annual subscription fee, and the increased revenue through higher share of wallet.

Our modelling suggested that if members bring about half of their non-MMT business to MMT, and cancellation rates don't rise too much, the program could break even at a very small annual fee.

The model was highly sensitive to the cancellation rates. Luckily, our per-transaction free cancellation product had established the ceiling for us. Cancellation rates on a subscription product would be definitely lower than that.?

That gave us huge comfort.??

But could this change customer behavior in ways we hadn’t thought about and create more risk? More worrisome – would the higher risk customer cohorts end up buying it, resulting in huge losses for us?

Adding to that, it would take a while before we could figure out the P&L of this program. Travel is not a high frequency category, so we would need to wait about a year.

It was clear that we would need to move cautiously.

But it was also clear that there were enough possibilities that we needed to take this to customers and see what happens. This was a completely novel idea, and there was no way to validate assumptions looking at past data.

Opening up an incredible possibility…

If it worked, it could have very significant implications on how we viewed our business and our relationship with our customers.

At an extreme, our commercial relationship could move from a transactional fee model to an annuity model. The main revenue mechanism would shift to the subscription fees – the money we would make off the transactions would get balanced with the cost of covering the cancellations. This would have huge implications on how we thought about our business and our customers.

We would have converted from a transactional ecommerce business to a relationship business like telecom.?

Launch and early results… and some a-ha moments!

Because of the huge risks and unknowns, we decided to make it an invite-only program, and focus on customers with a low share of wallet, but medium travel frequency. High frequency travellers would pose much higher risk, so we didn’t want to take a risk at the early stage. (Later we added them too, and limited the coverage to the first 10 bookings in the year.)

What were the results?

Early results were very very encouraging:

• Member NPS moved from 50 to 70. (and NPS of members who actually did a cancellation went up to 85!)
• Member transactions on domestic flights went up to 3x.?
• Most surprising to us was that member transactions increased in other businesses as well. International flights and hotels were not part of the program, but we got growth here as well!?

The program was creating a secondary effect- customers liked what they saw, and decided they wanted to trust us with all (or at least more) of their travel!

We were generating true brand loyalty!

As we talked to members to gain more insight, we got a huge a-ha moment.

For customers, the big game changer was NOT that they would save cancellation fees on bookings. It was that it freed them up to make bookings even when plans were tentative, thereby booking early and saving money!

This was worrisome for us. Because this was a change in behavior we had not foreseen. This would naturally result in much higher cancellation rates, and could be even higher than the free cancellation product. Luckily we were on a slow, measured rollout. So total risk was limited.

As more time went by, and more bookings and cancellations happened, cancellation rates indeed became higher.

But not as high as we had thought.

A side-note: the commitment impact of making a booking...

This didn’t make sense. Bookings had definitely increased. And booking windows had also increased – so data was in-synch with the anecdotal input of tentative bookings. Then why were cancellation rates not shooting up? As we talked more to the members, another major a-ha moment hit.

To some extent, the act of making the booking itself acted as a commitment to travel. Despite the fact that there was no cost to cancel. It became the stake in the ground, so to speak, that resulted in customers then converting the tentative plan to a confirmed plan, resulting in keeping the cancellation rates lower.?

Which means that the product ended up increasing travel itself! We had unwittingly, with no intent to do so, induced people to travel more!?

...End of side note

Coming back to the P&L- the higher cancellation rates could mean that the program was not sustainable. Luckily, the significant increase in bookings across all businesses (domestic and international flights and hotels) more than covered up for this. And therefore became even more of a strategic play- for someone else to replicate this, they had to have significant strength in all these businesses, therefore raising the bar even higher!

By Q3 2020, Double Black had slowly grown to about 150k members. While the numbers were small, they added disproportionate value. And fitted well into our overall loyalty strategy. (But that’s a topic for another day.) But then Covid happened, and Double Black was one of the casualties.

Postscript: Some interesting P&L possibilities

As we thought hard about how to reduce the cost of cancellations, we discovered some interesting possibilities...

As the cancellation rates were much higher than usual, airlines were making more money off these cancellations.

This threw up interesting possibilities – with more money in the ecosystem being generated by us, we could take a share of the additional money- either for ourselves to reduce risk, or share with customers to provide even stronger value.

We could, for instance, show the increased cancellation data to the airlines, and get them to share some of the benefits with us. It would take some doing, but at least the possibility existed.

There were other interesting possibilities too – based on the data we already had, we could model and find other opportunities to create value – some of the arbitrage kind, and some others too. Can’t speak here about some of the ideas we came up with, but they were pretty radical.?Let me just say that the thought was on how to avoid paying cancellation fees alltogther. Rest left as an exercise for the reader :-)

And some broad learnings...

Just in case they were not obvious :-)

• Listen to your users. Very closely. And talk to them. Regularly.
• Often, things that seem out of your control are not. You just need to think hard, identify the constraints, and question them.
• If a business practice doesn't seem fair, it's probably ripe for disruption.
• Launch, and iterate. Specially for radical products. The results will always surprise you, and if nothing else, help you develop significant insights about your business and customers.

Before you go...

Some other articles that might interest you...

Fixing Cancellations in the Cab Market

On Product Management and Problem Solving

The Power Of Service Guarantees

Customer Experience: A few learnings