The Omnichannel Dilemma: How Retailers Can Get It Right



The biggest word in retail right now is omnichannel, which is the seamless integration of online and offline shopping. When omnichannel is done right, a customer can shop in her favorite store to find that perfect pair of winter boots while using her smartphone to read reviews and check whether the store across town has a black pair in her size. Omnichannel creates a frictionless world for demanding consumers. But for businesses, building that omnichannel network while also developing new products can be a daunting challenge.


Santiago Gallino, a professor of operations, information and decisions at Wharton, and Robert Rooderkerk, an operations management professor at Rotterdam School of Management at Erasmus University, have been researching this challenge and found some best practices based on the success of companies that are getting it right. Their study, “New Product Development in an Omnichannel World,” was published in California Management Review. They recently spoke with Knowledge@Wharton about their findings.


An edited transcript of the conversation follows.


Knowledge@Wharton: Your study is prescriptive for retailers. You’re not just telling them they need to be omnichannel in new product development, you’re handing them a research-based road map on how to get there. There are seven recommendations in your paper. Among them, you advise companies to get more input from customers earlier in the product development process and even make them co-creators of products. How is that different from the past?


Santiago Gallino: I would agree that many of these things are not completely new or things that companies were not doing before. The change is the extent to which this becomes available to companies; before, it was harder for them to get to those co-creation opportunities. In the article, we explore Lego Ideas and Nike By You. These are companies that have been keeping an eye on customers and the things they like for many years. But now, they have this opportunity that the digital platform has given them to invite customers very early on in the process, or even watch how customers can create new products or combine their existing products in different ways. This allows companies and brands to learn from actual customers. But they also learn from people who are not going to buy, understand why they’re not buying now, and learn what they can do to make them become a customer in the future.


Another key change is the volume and the speed of data that the companies can access from these customers, and how companies inform their creativity and new product development process based on this new volume of data. I think that these two aspects, the idea of learning not only from customers but from the non-customers, and the volume and speed at which the companies receive the data, are really critical in this new process.


Knowledge@Wharton: Does this co-creation go beyond the traditional focus group that is put together when companies are trying to bring something to market?


“We used to think about the relationship between manufacturer and retailer as something that has a competing nature. But going forward, we should expect to see more collaboration.”–Santiago Gallino

Robert Rooderkerk: Focus groups are very static. They are very expensive to run. You often have to run multiple ones. But customers are often quite intrinsically motivated to contribute to a company or its innovation process. So, it’s really about activating those customers.


Santiago already explained that getting access to customers has become easier using all sorts of digital platforms. But you’re also seeing companies connecting to customers in new ways, often forming online communities. A great example is Lego, the toy manufacturer that has been extremely successful in creating an online community and has leveraged that community for new product development ideas. They often hold contests, called Lego Ideas, in which people can contribute an idea for a new toy set. Not only do people contribute the ideas, the community also votes on these ideas. You instantaneously generate ideas and vet them with your potential customers — and these customers are often extremely creative. You’re using a sort of outside-in innovation. You’re almost outsourcing your innovation process. It’s very simple. You have to appeal to their intrinsic motivation. You can often also apply extrinsic motivation, such as in the case of Lego. They provide the winning entry with 1% of the revenue that’s made from the products. They also get credited on the packaging.


Online focus groups have gotten updated to what we call “online focus group 2.0,” where it’s like an online gated community with a moderator, and there are a lot of gamification components in there. For instance, consumers have to provide feedback on a new ice cream flavor for Unilever. The ones that come up with the best recommendations get to see the new product that Unilever is making at that moment. Then, instantaneously, Unilever gets feedback on that. So, focus groups themselves have become more like online communities.


Knowledge@Wharton: In your paper, you say retailers need to widen their scope and get more input from their trade partners. What kind of input?


Rooderkerk: Manufacturers typically have organized a new product development process by doing a lot of marketing research, focus groups, or they bring consumers in for a taste test or whatever. But it’s still kind of a push demand to innovation, where the manufacturer starts the innovation process, solicits feedback and screens interesting ideas based on that.


Trade partners often have much better access to the consumer. Selling to a retailer means you broaden your distribution, but it also means you give up some of the direct access to your customers. Now, a retailer can sit on the data and not share it, or actually use it for private label development. Let’s take Amazon as an example. Amazon has leveraged data obtained by selling big brands on its platform by actually developing its own brands. Amazon currently already has more than 100 brands, often under the radar.


But a retailer could also take the point of view of sharing data, or insights from that data, with manufacturers. A good example is Alibaba’s Tmall platform, which has started a marketing research branch called the Tmall Insight Center that shares data with leading manufacturers to jumpstart their innovation process. Alibaba was mining its data and saw that, for example, consumers who were looking for chocolate were often also buying spicy foods. They made a very simple suggestion to Snickers to export its avenue of combining chocolate with spice, and they came up with a spicy Snickers. That turned out to be one of the most successful CPG (consumer packaged goods) introductions in China. Leveraging data that a retailer’s sitting on and sharing it with the manufacturer could lead to much better innovations. Then the retailer has the first rights to sell that product.


Another great example is here in the Netherlands, where Coolblue was analyzing its online data and noticed what consumers were looking at when they were searching for a laptop. Not only do they see what people buy, they also see what consumers search for and cannot find. Based on the filter behavior, they noticed that consumers were looking for certain combination of specifications that Apple was currently not providing. But these customers were very interested in an Apple. Coolblue went to Apple and said, “We would like you to make a new MacBook that’s made to order for us because we believe there’s a lot of demand.” Apple was very skeptical, but Coolblue backed up their claim by saying, “We’re going to order a lot right away, off the shelf.” It turned out to be a great success. Apple sent its vice president of sales to our very small country to figure out what these guys were doing here.


The second component is operational considerations. A retailer knows very well the difficulties involved in selling and shipping product. For instance, in Holland, a retailer knows very well what the size of a mailbox is. An international manufacturer doesn’t have that local market knowledge. By slightly adjusting the packaging, it could put through a door, which would lower transportation costs or delivery costs. Customer preferences and operational considerations are typically the data that retailers have and could share with manufacturers.


“Leveraging data that a retailer’s sitting on and sharing it with the manufacturer could lead to much better innovations.”–Robert Rooderkerk

Knowledge@Wharton: Are there downsides to manufacturers and retailers working together?


Gallino: The mindset in the past will suggest that these tend to be confrontational or competing relationships. I think what we are seeing today is that when these are seen as collaboration and opportunities to partnership, it’s when the customer gets the best of the two. This is even more evident when you see retailers that are platforms. The customer of the platform is both the end customer who is going to buy the product, but also those vendors that need to be attracted to place the products in the platform. This is a game-changer for the role of the traditional retailer now transforming to a platform that sees customers on both sides.


In the past we used to think about the relationship between manufacturer and retailer as something that has a competing nature. But going forward, we should expect to see more collaboration, more data sharing and an opportunity for both of these to grow.


Knowledge@Wharton: Your paper also recommends companies think outside of marketing and sales in new product development. They need to involve departments that don’t always get considered. Can you talk about that?


Gallino: One clear example of this is when we think about the packaging and the delivery process of the product. In the past, when we ignore the omnichannel component, all of this is going to happen from a distribution warehouse to a retail store to the customer home. But nowadays, this path is not as clean as it used to be. And in many cases, companies are delivering products directly to the customers.


This requires involving in the development process areas that traditionally might not have a lot to say. But to date, they have had a lot to say about the delivery of the product. What can be more efficient? What can be more convenient for the customer? What can be more cost-effective and safer in terms of making sure that the product arrives in good shape and in the condition that the customer expects? And there is the experience once the customer opens the product at home, which is a very different experience compared to the store. In the store, you get to meet the sales representative, you could talk with them, they package everything for you, and you take it home. There is a lot of emotional engagement and learning of the product in that process.


Nowadays, we receive a box at home, we open the box, and that’s when the magic happens. Companies are being more thoughtful and creative about that change in the omnichannel layout and are thinking about how they can create the wow factor at the moment that someone opens the box.


There is one example that I like a lot. Being from Argentina, grilling and barbecuing is part of our DNA. There is a company, Spark Grills, that developed new grills only designed for a digital experience in terms of finding and buying the product. They also redesigned the charcoal that you use. Instead of the traditional briquettes, this is like a big brick that is much more compact and easier to deliver. So, the company sells the grill and the newly designed briquette that you can use with that grill, and the delivery becomes a lot more efficient, cleaner for the customer to handle. I think that that is one example of companies that have thought beyond the product itself, but also the whole experience in a context where I’m trying to sell grills online without the need to visit the traditional store.


“We receive a box at home, we open the box, and that’s when the magic happens. Companies are being more thoughtful and creative about…how they can create the wow factor.”–Santiago Gallino

Knowledge@Wharton: Dr. Rooderkerk, there’s another example in the paper about the delivery of Heineken in the Netherlands. Can you talk about that as another way to get departments besides marketing and sales involved the last mile?


Rooderkerk: We have a completely digital native online grocery retailer called Picnic. They’re currently No. 2 in the Netherlands, and their model is centered on operational excellence. Instead of thinking, “We’re going to offer the largest assortment,” it’s all about efficiency.


They drive around in these little electric milk carts [for delivery]. Those carts have a lot of advantages. One is that they can navigate narrow streets, cobblestone streets here in the Netherlands, so they can quickly enter, for instance, the Amsterdam canals. That means that they can do about six deliveries an hour, compared to about three or four of their main competitors. There’s one downside. The carts are very small, very narrow. Certain big, bulky products— and those are the products that people typically like to order online— cannot be carried in those carts. One of those products is a crate of 24 bottles of Heineken beer. Picnic reached out to Heineken and said, “Can you help us think about how we can redesign or about an alternative that would fit in our carts?” They came up with something called a quarter crate — six bottles that are held together by a piece of cardboard. Now, that’s not new. But cardboard used to be quite thick because you had to carry it all the way from the store to home. Now, you only have to carry it from that small electric vehicle that stops in front of your home to your kitchen counter, so the cardboard can be much thinner. They can actually fit four of those in one tray, and that comes right to 24 bottles of beer.


This is a product development that was driven by operational considerations, and I think that’s what omnichannel is about. It’s about realizing that every different channel comes with its specific operational challenges, and that has profound implications for packaging. In a store, packaging has to have some kind of selling and advertising factor. That’s much less important when it’s online delivery, when it’s more about making transportation efficient. These kinds of considerations should be at the beginning of a product development process, not at the end. Where operations people used to come in at the end, now the marketing and innovation guys have designed the products. Now, it’s up to you to make sure it can be transported. Now, we’re going to factor it in from the very beginning.


Increasingly, trade partners demand these kinds of operational efficiencies. Santiago mentioned some great points about the wow factor with consumers, but also these platforms and the retailers to which you sell are pushing manufacturers to think along with them on operational efficiency. Scoring higher in that also means you get a more preferred placement on the website, and it directly translates to higher sales.


Knowledge@Wharton: In the paper, you discuss the importance of sharing data, saying it “should not be sequestered.” Can you talk about that?


Rooderkerk: I think for many companies, it’s kind of a knee-jerk reaction to sit on top of their data, to treat it as proprietary. It’s called a competitive advantage. But data is only a competitive advantage, if you’re able to mine it. In Dutch we have a saying that translates to, “You can only really grow something when you’re able to multiply.” And the multiplication comes in sharing it with your chain partners.


Still, a lot of product innovations fail, and that’s very costly to the whole chain. To the retailer, you have all of this shelf space that you’re not efficiently using. You have to phase out the products, return it to the manufacturer or sell it against reduced prices. So, sharing data with a chain partner means they get better insights into customer preferences. That translates directly into better innovations that are better for the retailer. They also have a better fit with the customers. It’s not a one-size-fits-all anymore. Retailers are not willing to just adopt a product and put it throughout their whole chain. But they’re more focused and are going to launch that in certain stores where they believe the product has the most potential. Only when it’s proven there, they’ll roll it out to the larger chain.


“You have to think about how to work together to grow the pie, and growing the pie is sharing data. Obviously, that comes with trust.”–Robert Rooderkerk

It’s kind of odd. There are still some sectors—and grocery retail is one of them— where the large players do demand manufacturers take a data-driven approach. But at the same time, they don’t really share the full data. They will share it at a very high aggregation level. They mask it in such a way that you’re basically giving your chain partners suboptimal instruments, but you demand optimal innovation. And that’s odd. You have to think about how to work together to grow the pie, and growing the pie is sharing data. Obviously, that comes with trust.


Gallino: In my experience, a good approach is incrementality. You don’t need to start with the biggest, most ambitious project the first day. There is a lot of low-hanging fruit; small sharing opportunities that can prove the trust and the value in contexts where it’s relatively easy to find a win. I will recommend that path if you want all relationships transitioned into more productive relationships, by building this trust in an incremental way.


Knowledge@Wharton: What’s next for this line of research?


Gallino: I think that the digital transformation in retail, omnichannel in particular, has been fantastic to follow and study. I think that we’re going to see more innovation going forward, more refined issues like the ones we are discussing today. Like, how data sharing is going to happen, how the companies are going to rethink the delivery of products, how they can make these communities of customers that are sharing their knowledge in creating new products. I’m extremely excited to learn from managers in the field doing these things and hopefully be able to contribute with some of my research along these lines.


Rooderkerk: I fully share Santiago’s excitement. We’re typically never short of things to talk about on this topic. I think we’re in the middle of a huge influx of new technology in retail. With the current pandemic, if there’s any silver lining to it, you see a lot of innovation, which I think is going to be very much on the intersection of store design and effectively using technologies to help consumers.


But you also see that even during a pandemic, we still want to visit the store. Even with a [face covering] on, there’s still enough to carry us to the store because stores are great places. They offer us a lot of inspiration, experience, etc. At the same time, we don’t have to make things needlessly complicated, so efficiency is also very important to consumers. How are we going to effectively use technology to help consumers find the product that provides the best fit with them, and doing that in a way that contributes to their day? All of these new technologies are leading to a huge abundance of new data, and they’re often unstructured. I think they provide a lot of interesting opportunities for us to improve retail operations, so I’m very excited to explore all these avenues in the coming years.

© 2021 TMG Search. All rights reserved.