October 12, 2015
Finding the path to more relevant offers
Speaking last week at the Shop.org 2015 Digital Summit in Philadelphia, Doug Mack, CEO of Fanatics, the largest e-tailer of fan merchandise, discussed the challenges around delivering relevant offers to consumers.
Beyond Fanatics.com and Fansedge.com, the company runs the fan gear websites for every major U.S. sports league, many professional teams such as the New York Giants and Chicago Bulls, several major media firms, and over 150 colleges.
For Fanatics, the first step is identifying relevant merchandise. Demand for jerseys and other fan gear can swing wildly year after year based on a team or a player’s performance. But the big win is getting an early read on "micro-moments" or when a surprising win or breakout performance creates a surge in demand for a team or player.
Once potentially strong merchandise is found, personalization is used to link relevant offers to consumers.
"Every day there are these great moments in sports but the question is how do you marry the right moment with the right fan with the right merchandise?" said Mr. Mack, who formerly ran One Kings Lane and co-founded Scene7.
Fanatics taps its database of tens of millions of customers to build profiles.
Doug Mack – Photo: NRF
"To begin to use that data to profile and build understanding of what given fans are interested in is a huge opportunity," Mr. Mack said. "And I know that is a pretty well-known discipline in digital retail. But getting that structure right and those categories right and the data collection right is super important."
Beyond its platform, social feeds can determine individual interests.
"Knowing where somebody lives, that’s a really big clue for our business," added Mr. Mack. Many of its partners simply ask customers about their favorite players.
"When you bring all that in you can begin to really micro-dissect who’s going to be interested in what," said Mr. Mack.
Asked in the Q&A session about the potential to be overwhelmed by data, Mr. Mack said to first step back to gauge how to use "structured data as a foundation" and explore the best way to classify. As data is gathered and used for profiling over a period of time, "you can then begin to get that sensitivity by different criteria, as to what is going to be the most needle moving and which areas are less important. But these are really hard problems to solve."
On the positive side, organizations that can gain scale and capitalize on the data first "are going to have a tremendous competitive advantage. Statistically, you’ll be more likely to make better decisions in every inch of the business. It’s going to be really cool."
BrainTrust
Dick Seesel
Principal, Retailing In Focus LLC
Recent Discussions
Discussion Questions
How can retailers avoid being overwhelmed by data in creating relevant offers for consumers? Where do you see the opportunities and pitfalls in the use of personal data?
Poll
Retailers need to determine which data is important to customers and which is superfluous. By focusing on the relevant retailers may miss some opportunities, but will get the majority right.
Secondly, retailers need to continually test messages and incentives. This is not expensive and can yield valuable information.
Finally, don’t be afraid to ask consumers what they want.
Retailers need only ask for the data they need to determine the demographics and the preferences of their customers. In some cases that could be very little data and requests for more data can be made as the customer becomes more comfortable with the retailer or the retail website.
Asking for a lot of data too soon would be a red flag to most customers and may turn many away from the retailer. The one major pitfall is when the retailer attempts to be your friend. They are not your friend but they should be your place of loyalty, your place of bargains, and your place where they know who you are and what you like.
It’s up to a fast-growing e-commerce player like Fanatics to figure out its data management requirements before it gets run over by too much information. Amazon is the obvious role model for using predictive technology to personalize shopper offers based on their purchase patterns and site visits. So in a sense Fanatics is not reinventing the wheel but needs to make sure it has the necessary resources (IT expertise and computing power) to manage its growth effectively.
Technology has advanced well beyond where it was in our industry even only six months ago. There are tools available that can virtually eliminate “being overwhelmed by data” and craft relevant, very personalized offers for shoppers regardless of channel through interpretive, predictive and prescriptive analytics.
The use of this data in these ways becomes a compelling incentive for shoppers due to its relevancy. The privacy issue rarely comes into play when the offers are welcomed as they are by shoppers.
Today there is no excuse for retailers being overwhelmed by data other than their inability to know what to do with it. The new technologies available for data management, insights, selection and segmentation are more accessible than ever, however retail as an industry has long lagged in being data-driven.
Yes there are exceptions, namely Amazon, but in general retailers haven’t made investments in capabilities or technologies to be more relevant to customers. It starts with leadership and a c-level commitment like Fanatics — otherwise it’s business as usual for retailers.
There is no escaping the deluge of data which is growing exponentially at the same time that consumers are becoming more demanding in their expectations and impatient with blunders in misuse of personal data.
To make relevant offers requires knowing enough about the consumer who has opted in to share personal data. It also requires blending customer behavior in the physical and digital stores, online searches as well as understanding the context framing the offer.
There are pitfalls when loyalty card data is shared across a family and offers are based on this pseudo-profile. Also, when data is limited to a customer’s behavior in one banner or chain, you get an incomplete view of that customer. Another is limiting efforts to structured data which leaves out rich possibilities of publicly available social media data.
But all the data collection is of little value unless the retailer has the tools to glean actionable insights from the data or builds a data-driven digital enterprise for the 21st century.
Market research is undergoing a huge transformation. Most of the traditional ways of gathering data (surveys and active responses) are giving way to passive monitoring of behaviors. For example, once shoppers opt-in we can gather information passively from cellphones and computers, so that we know where people are, where they are clicking and what they are watching, among other things. We have access to more information, in more ways, than ever before.
This new kind of data gathering is more akin to surveillance — which means that without a plan, we could be drowned in information. So two tools are necessary to navigate through this new world. First, as Mr. Mack says, a structured plan to view data is useful. This means that (at least in part) we need to know what we want to know. That sounds comical, but it’s not so easy. This step means we need to think ahead of time about what we are seeking — including “what ifs.”
The second tool set we’ll get increasingly accustomed to is data visualization. While even the humble bar chart is a form of data visualization, we’ll need to employ more sophisticated expressions, as more information becomes available to us. Creating these visualizations or expressions is a special skill set and one which will be increasingly sought after. But beyond that, we’ll be able to view the information in context to other behaviors, which will give us new and sometimes unexpected views of the world.
Its seems too intuitive to say that it all begins with a strategy that makes sense to both retailer and customer, but I am surprised at how often that is not the case.
To that end, and as others have said, data no longer has to be overwhelming. Perhaps the variety of options of systems and software a retailer uses to mine, filter and analyze may be more overwhelming than the data itself.
Start with basic “blocking and tackling” targeting strategies and don’t over-think the process. Over time, the shopper’s responses (or lack thereof) to data-driven communication will provide the feedback needed to refine and become more sophisticated in the targeting approach.
Also, never forget that the more sophisticated the targeting you do, the more relevant content you need.
The train has left the station: For retailers, the need to transform to be customer focused versus product focused has only increased in the last several years. Developing more and deeper customer centric capabilities is now paramount in today’s retail world where the customer is in the driver’s seat.
Retailers now need to understand, predict and deliver what customers want. In recent years, disruptive technology has brought on many new challenges, but has also enabled new opportunities:
- Capturing new data sources
- Combining data to enable new insights and take new actions
- Implementing more sophisticated analytics to differentiate and grow
Those retailers that might still be struggling with “being overwhelmed by data” need to just open their eyes, take a look around, and find the support tools needed to properly manage/drive the customer data and the analytics. The technology enabled support tools — even for those “really hard problems to solve” — are right there for those retailers who need them to start better capitalizing on their data.
Every retailer should have a data strategy in place by now. There are so many tools and technologies to help them manage the data that there is no excuse.
The important thing about that strategy is that retailers must not only know their customers but they must also know which customers are the top influencers and most profitable. If they do not first isolate these key customers, yes, the data will be too overwhelming.
And that’s my 2 cents.
My experience as a retailer using extensive marketing personalization is that shopper identified purchase data is gold. While other information — demographics, social media activity, etc. — is certainly valuable, in fast moving consumer goods retail, nothing surpasses historical purchase data.
State of the art personalization has gone far beyond marketers doing the targeting. Today, incredibly sophisticated predictive algorithms combined with machine learning feeding off big data enable strategically driven personalization across all individual shoppers. The big data is not just purchase data, it is a massive number of calculated scores, refreshed with each purchase, that track brand loyalty, discount propensities, purchase frequency, and so on for each individual shopper.
The challenge for retailers is creating a large enough “offer pool” to support this kind of personalization. Experience shows in supermarket retailing a pool of 400+ offers is required for effectiveness; more is better. Many supermarket and consumer goods retailers tend to overlook the offers they already have in their weekly sale items and TPR items around the store. Retailers can successfully leverage this existing content to power up very effective personalized marketing programs without incurring additional markdown cost.
Marketing personalization done this way can be incredibly powerful.
Companies are slowly becoming aware of how little verification there is in email lists. This is largely due to the fact that most people own more than three active email accounts for seemingly valid reasons. Additionally it takes a lot of time and resources to keep off of general spam site lists that serve to shut down many corporate campaigns. The larger companies farm out e-mail campaigns for just this reason. The slower buildup of lists created by satisfied customers themselves is still the best method out there.
More Discussions
Should Grocers Pay Online Influencers?
What should the relationship between grocers and influencers look like?Will Google’s Pomelli Photoshoot Live up to the Hype for Retail SMBs?
Will Pomelli’s Photoshoot tool be a game-changer for retail SMBs?Mass Retailers, Including Walmart, Tie Grocers on Penetration For the First Time: What’s Next?
Walmart and other mass-channel retailers are reaching dominant penetration levels. What’s next?
Retailers need to determine which data is important to customers and which is superfluous. By focusing on the relevant retailers may miss some opportunities, but will get the majority right.
Secondly, retailers need to continually test messages and incentives. This is not expensive and can yield valuable information.
Finally, don’t be afraid to ask consumers what they want.
Retailers need only ask for the data they need to determine the demographics and the preferences of their customers. In some cases that could be very little data and requests for more data can be made as the customer becomes more comfortable with the retailer or the retail website.
Asking for a lot of data too soon would be a red flag to most customers and may turn many away from the retailer. The one major pitfall is when the retailer attempts to be your friend. They are not your friend but they should be your place of loyalty, your place of bargains, and your place where they know who you are and what you like.
It’s up to a fast-growing e-commerce player like Fanatics to figure out its data management requirements before it gets run over by too much information. Amazon is the obvious role model for using predictive technology to personalize shopper offers based on their purchase patterns and site visits. So in a sense Fanatics is not reinventing the wheel but needs to make sure it has the necessary resources (IT expertise and computing power) to manage its growth effectively.
Technology has advanced well beyond where it was in our industry even only six months ago. There are tools available that can virtually eliminate “being overwhelmed by data” and craft relevant, very personalized offers for shoppers regardless of channel through interpretive, predictive and prescriptive analytics.
The use of this data in these ways becomes a compelling incentive for shoppers due to its relevancy. The privacy issue rarely comes into play when the offers are welcomed as they are by shoppers.
Today there is no excuse for retailers being overwhelmed by data other than their inability to know what to do with it. The new technologies available for data management, insights, selection and segmentation are more accessible than ever, however retail as an industry has long lagged in being data-driven.
Yes there are exceptions, namely Amazon, but in general retailers haven’t made investments in capabilities or technologies to be more relevant to customers. It starts with leadership and a c-level commitment like Fanatics — otherwise it’s business as usual for retailers.
There is no escaping the deluge of data which is growing exponentially at the same time that consumers are becoming more demanding in their expectations and impatient with blunders in misuse of personal data.
To make relevant offers requires knowing enough about the consumer who has opted in to share personal data. It also requires blending customer behavior in the physical and digital stores, online searches as well as understanding the context framing the offer.
There are pitfalls when loyalty card data is shared across a family and offers are based on this pseudo-profile. Also, when data is limited to a customer’s behavior in one banner or chain, you get an incomplete view of that customer. Another is limiting efforts to structured data which leaves out rich possibilities of publicly available social media data.
But all the data collection is of little value unless the retailer has the tools to glean actionable insights from the data or builds a data-driven digital enterprise for the 21st century.
Market research is undergoing a huge transformation. Most of the traditional ways of gathering data (surveys and active responses) are giving way to passive monitoring of behaviors. For example, once shoppers opt-in we can gather information passively from cellphones and computers, so that we know where people are, where they are clicking and what they are watching, among other things. We have access to more information, in more ways, than ever before.
This new kind of data gathering is more akin to surveillance — which means that without a plan, we could be drowned in information. So two tools are necessary to navigate through this new world. First, as Mr. Mack says, a structured plan to view data is useful. This means that (at least in part) we need to know what we want to know. That sounds comical, but it’s not so easy. This step means we need to think ahead of time about what we are seeking — including “what ifs.”
The second tool set we’ll get increasingly accustomed to is data visualization. While even the humble bar chart is a form of data visualization, we’ll need to employ more sophisticated expressions, as more information becomes available to us. Creating these visualizations or expressions is a special skill set and one which will be increasingly sought after. But beyond that, we’ll be able to view the information in context to other behaviors, which will give us new and sometimes unexpected views of the world.
Its seems too intuitive to say that it all begins with a strategy that makes sense to both retailer and customer, but I am surprised at how often that is not the case.
To that end, and as others have said, data no longer has to be overwhelming. Perhaps the variety of options of systems and software a retailer uses to mine, filter and analyze may be more overwhelming than the data itself.
Start with basic “blocking and tackling” targeting strategies and don’t over-think the process. Over time, the shopper’s responses (or lack thereof) to data-driven communication will provide the feedback needed to refine and become more sophisticated in the targeting approach.
Also, never forget that the more sophisticated the targeting you do, the more relevant content you need.
The train has left the station: For retailers, the need to transform to be customer focused versus product focused has only increased in the last several years. Developing more and deeper customer centric capabilities is now paramount in today’s retail world where the customer is in the driver’s seat.
Retailers now need to understand, predict and deliver what customers want. In recent years, disruptive technology has brought on many new challenges, but has also enabled new opportunities:
Those retailers that might still be struggling with “being overwhelmed by data” need to just open their eyes, take a look around, and find the support tools needed to properly manage/drive the customer data and the analytics. The technology enabled support tools — even for those “really hard problems to solve” — are right there for those retailers who need them to start better capitalizing on their data.
Every retailer should have a data strategy in place by now. There are so many tools and technologies to help them manage the data that there is no excuse.
The important thing about that strategy is that retailers must not only know their customers but they must also know which customers are the top influencers and most profitable. If they do not first isolate these key customers, yes, the data will be too overwhelming.
And that’s my 2 cents.
My experience as a retailer using extensive marketing personalization is that shopper identified purchase data is gold. While other information — demographics, social media activity, etc. — is certainly valuable, in fast moving consumer goods retail, nothing surpasses historical purchase data.
State of the art personalization has gone far beyond marketers doing the targeting. Today, incredibly sophisticated predictive algorithms combined with machine learning feeding off big data enable strategically driven personalization across all individual shoppers. The big data is not just purchase data, it is a massive number of calculated scores, refreshed with each purchase, that track brand loyalty, discount propensities, purchase frequency, and so on for each individual shopper.
The challenge for retailers is creating a large enough “offer pool” to support this kind of personalization. Experience shows in supermarket retailing a pool of 400+ offers is required for effectiveness; more is better. Many supermarket and consumer goods retailers tend to overlook the offers they already have in their weekly sale items and TPR items around the store. Retailers can successfully leverage this existing content to power up very effective personalized marketing programs without incurring additional markdown cost.
Marketing personalization done this way can be incredibly powerful.
Companies are slowly becoming aware of how little verification there is in email lists. This is largely due to the fact that most people own more than three active email accounts for seemingly valid reasons. Additionally it takes a lot of time and resources to keep off of general spam site lists that serve to shut down many corporate campaigns. The larger companies farm out e-mail campaigns for just this reason. The slower buildup of lists created by satisfied customers themselves is still the best method out there.