The Death of the Mail-in Rebate (a data insight driven transformation story)

A data insight driven transformation story

All of us at some time in our careers have instituted a policy, process or even strategy that at the time seemed to make good business sense but in the long run turned out to be both detrimental to the sustainable business value for your organization and caused undue effort and stress on your customer.

Depending on your geo demographics, which is a fancy way to say whether you are old like me, you may remember a time where pricing in consumer electronic stores and online were manipulated by the promise of a rebate. What you might not know is how the hated rebate started to meet its demise back in early 2002 eventually being completely phased out by 2005 by companies like BestBuy with other retailers following suit.

This is an old story filled with many characters and many witnesses. I will do my best to accurately tell this story based on the context that I have and the best recollection of events that happened many years ago. I apologize up front if I have missed something or misconstrued something. Please feel free to add or correct this story if you were a participant. Keep in mind that the goal of this story is not to disparage any individuals or brands but to show how data driven insights can be used to transform deeply imbedded activities that are cultural and business norms.

A seat with a view

I was lucky enough to have a front row seat or maybe a better way to explain it is that I actually was on the field playing in this rough and ready combat between the customer’s wallets and BestBuy’s profit. This is story about how data and unbiased analysis helped leaders see insights that were contrary to their beliefs and drove them to make better strategic decisions. These decisions ultimately led to more lifetime value from the customers and better retention but I will explain this a little later.

First, I need to give you a little context here. I was the VP of Enterprise Customer care from 2000-2003 at BestBuy. I had responsibility for over 100 million phone calls, emails and other interactions coming into the company for customer service, home delivery, extended warrant repairs, internet order status, pricing & availability, geek squad and yes, rebate customer service. I also had a team of attorneys, a data privacy team as I was also the Chief Privacy Officer and a membership in the CRM Executive Strategy team. I also had responsibility for Advertising Compliance which means I had to ensure that we were following legal guidelines for our advertising. As a hands on and ears on leader, I had the honor of listening to a lot of customers give me their interpretation of the experience they were having across every interaction with BestBuy and especially around rebates.

How rebates work

Some of you may be familiar with how a rebate works but it might make a little sense to explain it a little. The advertising or circular, as it was called, drove a large amount of the traffic to a store. The goal of your circular was to have items that were compelling to the market at a price that differentiated you from the competition and drove customers to go from the shopping / research mode to the buying mode. These ads were driven by the merchandising category leaders who were buoyed by the manufacturer with Market Development Funds (MDF) or sometimes known as Co-Op Funds for placement in these ads. The goal of the merchant leader was to maximize volume while still maintaining an appropriate profit margin in their category.

Rebates were used to show a lower price for an item with the net value being received once you got your money back from the manufacturer or rebate center. It set an expectation for the customer that they were paying x for an item once they received the rebate and drove thinking about immediate gratification with some incremental work associated with applying for a rebate.

The question you may already be asking yourself is why not just lower the price? But this is where the behavioral science of redemption and the attitudinal understanding of human expectation collide to create an advantage for the retailer. The technical term for this is called a Breakage Model. Based on the amount of the rebate, the purchase price of the item and the amount of effort required to get the rebate dictate a formula on what percentage of customers will actually apply for it and receive it. This calculation is figured into your profit margin forecast on an ads and that drives the overall strategy. Pretty smart use of historical behavior patterns to predict future activities, unless you are the customer.

The category leader was not the bad guy in this story

You might be thinking that this paints an ugly picture of the people running these categories and you couldn’t be any further from the truth. It is true that they used the breakage model in their calculations but this was a deeply embedded practice that had been around for decades that no one really understood or challenged. The key issue comes down to what I would call departmental measurement. What gets measured not only gets managed but also becomes the measurement for success. This group of people were measured on the volume of items sold from the circular and the gross profit margin for the product. Their ultimate job was to drive year over year revenue increases and grab both share of wallet and share of market. This was a weekly focus which automatically drove transactional thinking. The reports that they generated, the performance that was tracked and the results that were rewarded only cemented the continuation of this strategy.

Voice of the customer

I often joked about my job being like the guy who follows the animals in a parade. I can tell you the details about the dietary habits of these animals without ever stepping foot into the cage or being one of the handlers.

Hopefully I haven’t lost you in that description because I actually think it is a privilege to be in a place where you can spot anomalies and have access to the most important data that can help your circus keep it star attractions healthy and profitable. You have the opportunity to look at things from a different perspective and see things that others may not or may not want to see. You can actually bring incremental value to your organization if you know where to start.

The biggest mistake that customer orientated functions such as service can make is that they get caught up in their own departmental measurements. Yes, you do have a function to run with operational measurements but you have a bigger responsibility to also measure the attitudinal and behavioral data that eventually leads to overall financial results. I put a diagram down below to give you a landscape of what you should be looking to understand.

It all starts with a question

I was luckily raised by parents who taught me to question everything. This upbringing has provided me quite a lot of friction over the years but also given me the confidence to challenge the status quo. I was coached by some brilliant leaders over the years to make sure you had some facts to back up your opinions whenever possible to reduce some of this friction.

I have always been a hands on guy that dives into the details of the areas where I have responsibility. I am always combing the attitudinal, behavioral, operational and financial reports and raw data to try to better understand and to make adjustments. The rebate team that I managed met with me periodically for us to review the data and to determine how to be more efficient and effective at what we do. This was not always a fun meeting as the team was frustrated by the lack of urgency to pay the rebates, the amount of frustrated calls they received and the toll the anger from customers had on morale and overall employee engagement. I think we easily surmised that rebates were bad for customer experience but we had zero ammunition to challenge the program. So we started asking ourselves questions.

·        What happens to customers who don’t even apply for the rebate?

·        What happens to customers who are denied the rebate after submission?

·        What happens to customers who eventually get the rebate?

The questions start to get you to determine what information you need in order to answer those questions. We were a pretty sophisticated company when it came to data collection and were starting to get traction on some analysis so there were resources to use. When you have 87 million customers in your data base with 7 years of transaction and interaction information, you have a good haystack to start with. Not saying that all the data was as hygienic as you would like but it is a great place to start. I am absolutely jealous of today’s capabilities and often think back to what I could have done with today’s analytics and data science. Anyways, it gave us a running start.

From transaction to customer lifecycle

The questions that we were asking were not transaction orientated but instead asking about how a customer behaved over time. It meant that we had to put a customer centric lens on the data vs. a product category lens. We had to understand the journey of a customer before the rebate, during the rebate and after the rebate to understand the overall effect the rebate had on the lifetime value of a customer. We had to able to isolate specific customers and follow their buying behaviors over a long time in order to see specific patterns and then determine if there was a volume factor that could substantiate a hypothesis or were these just one offs. The questions start to morph at this point.

·        Do customers that buy a product with a rebate but do not apply for the rebate, buy additional products with rebates in the future?

·        Do customers who are denied a rebate buy any products in the future?

·        Do customers who get a rebate eventually buy more products with a rebate?

You even start to get into more detail.

·        What are the amounts of the rebates and what effect does that have on the behaviors?

·        What is the rebate effort (certain manufacturers more complicated) have to do with redemption, payment and overall retention?

Time to play with the data

I am not one to go on wild goose chases or just hope that if I throw a bunch of algorithms at a mound of data that it will drive an eventual answer. I would rather put a data dictionary together of the key data points and let the professionals use their big brains and big computers to run some models to determine if a hypothesis is right , misguided or completely opposite. I do not want to lead the witness or throw any of my personal biases at it. Having a hypothesis is just a way to go from boiling the ocean to microwaving a cup of water. It starts to isolate the work.

This isn’t a technical description of the models that we ran or a detailed account of how we did it. I think we all know that the technical capability that is available today far exceeds what I had at my disposal at the time. It can be done quicker and more effectively but it doesn’t replace the strategic value of using good design techniques to make sure you have framed the problem you are trying to solve or the question you are trying to answer. It teaches strategic thinking and places data driven insights into the organizational DNA.

The Insights

 Rebate fatigue

One of the first insights that we gleamed from the analysis is that people who didn’t bother to apply for the rebates were not likely to buy products in the future that had a rebate. So while the breakage model data looked good for these customers, the overall effects of rebates diminished on this group causing our ads to have less of an impact on them. This not only reduced our ROI on advertising spend but resulted in our lack of competitive differentiation in the market place.

 Rebate effectiveness based on amount

Overtime, you could see that the amount of rebate needed to drive effectiveness continually rise to overcome customer’s previous experiences of applying, reapplying, getting denied etc. People associated their expectations of the effort onto their next interaction with the circular. I often say that “the juice has to be worth the squeeze” and customers were now opting out of lower dollar amount rebated products.

Rebate caused attrition

The most damaging insight though was that customers who were denied their rebate or had to interact with Customer Care to lodge a complaint turned out to produce a reduction in share of wallet. These were customers who felt they were not given money they deserved and therefore took their business elsewhere.

Conclusion

Now we could go to senior leadership in the Merchandising Division with some answers. We had been telling them that customers were upset about the rebate process. We had been showing them the ratio of complaints and all the other operational data and I can’t blame them if it wasn’t compelling enough to drive change. Their reports and their biased view of rebates showed that it was driving successful margins. They were willing to put up with a little bit of friction as they could not see the overall damage to the brand and to the sustainable finances.

We began to show them the data and calculate the overall loss to Revenue and the reduction of ROI over a longer period than a quarter. Can I tell you that the initial findings drove immediate change? The answer is no. This should be something that you expect whenever you challenge status quo. I like to use the acronym S.A.R.A.H which is Shock, Anger, Resentment or Retaliation, Acceptance and Hope to describe the process that people go through for change or transformation. The leadership went through this as the magnitude of this change was significant and risky. We just continued to bring the data and to also circulate this data to leaders outside of the Merchandising group. It eventually caught on as small changes such as less rebates perc circular, instant rebates and automated rebates at the POS system occurred until they eventually killed the whole program. You have to stay the course if you want real change.

My final recommendation

In a lot of cases, you are aware of things that need to change but you may lack the courage or conviction to drive change. That is normal human behavior and there should be no shame in this. I don’t advocate battles of opinions or conjecture for the sake of debate. The person with the strongest voice or the most power will often win those arguments. But data driven insights; that is a different story. You can let the unbiased insights do the talking which removes the egos and focuses on facts. Eventually, people can’t overcome facts and the more people that know the facts eventually helps it become the tipping point.

I wish all of you access to the data and insights that empowers you to drive transformation.

Brian