The Crystal Ball – Exploring a Walmart POS Seasonal Forecast #CPGBI

One of the most common uses of machine learning in analytics is to forecast time based data. It’s the quintessential sales question – what will my sales look like next month, or next quarter, or next year even – the proverbial crystal ball, if only it were that simple. Something that we were fairly quickly put together using MicroStrategy’s visual insights and R-Integration is an “Ordinary Least Squares” regression algorithm to fit the best curve that captures the general trend and seasonal variability of Walmart POS data to predict future sales.

The formula is:

Y = bTrend*Trend + Σ (bSeason_i*Seasoni) + bIntercept

where

• Y is a numeric metric (called the Dependent Variable)
• Trend is a numeric metric that’s an arithmetic sequence of monotonically increasing values
• Seasoni is a binary indicator metric derived from Season, a numeric or string metric that represents each season. Binary indicators have a value of 1 for the i-th season and are 0 for all other seasons. For n seasons, there are n-1 XSeason_i variables
• bTrend, bSeason_i, and bIntercept are coefficients determined by the regression algorithm.

As sales drop in for the coming months, we should be able to gauge the accuracy of our prediction for the rest of the year. If this hold true, we could use it for some of our business decisions going forward. We could also look at just the latest complete months, so we would not see that monthly drop in month 201402. We could also look at this weekly by switching out just a couple of metrics.

Something else we could do is create a variance against actual POS sales, and if the variance exceeds some number, like 10% difference plus or minus, we could create an alert and send out warning emails to key people in our business so that they can plan for unanticipated high sales, or research a drop in sales.

Business Intelligence vs Analytics vs Big Data vs Data Mining #CPGBI

The business intelligence landscape is rapidly changing, and there is a lot of confusion on what the difference between BI, Analytics, Big Data, and Data Mining is. Whats more, you turn your head for just a minute and then there are whole new classes of terms that you’ve never even heard of before.

In the below article from Dennis Junk at Aptera’s blog, he breaks it down into four main categories to help you understand. As a CPG company supporting Walmart, I believe it is important to have a strategy for all of these concepts – especially in the wake of their new supplier terms and squeezing they are employing. If you don’t, we can help you with that, as we are a full service Business Intelligence company that can deliver Analytics from Big Data sources and use the R statistical package to mine the data for meaningful insights.

This is the broadest category and encompasses the other three terms here (at least as they’re used in a business IT context). BI is data-driven decision-making. It includes the generation, aggregation, analysis, and visualization of data to inform and facilitate business management and strategizing. All the other terms refer to some aspect of how information is gathered or crunched, while BI goes beyond the data to include what business leaders actually do with the insights they glean from it. BI therefore is not strictly technological; it involves the processes and procedures that support data collection, sharing, and reporting, all in the service of making better decisions. One of the trends in recent years has been away from systems that rely on IT staff to provide reports and graphs for decision-makers toward what’s called self-service BI—tools that allow business users to generate their own reports and visualizations to share with colleagues and help everyone choose what course to take.

Analytics:

This is all the ways you can break down the data, assess trends over time, and compare one sector or measurement to another. It can also include the various ways the data is visualized to make the trends and relationships intuitive at a glance. If BI is about making decisions, analytics is about asking questions: How did sales for the new model compare to sales for the old one last month? How did one salesperson do compared to another? Are certain products selling better in certain locations? You can even ask questions about the future with systems that perform Predictive Analytics. Some companies treat analytics and BI as synonymous—or simply rely on one to the exclusion of the other. But analytics is generally the data crunching, question-answering phase leading up to the decision-making phase in the overall Business Intelligence process.

Data Mining:

Finding answers you didn’t know you were looking for beforehand is what Data Mining is all about. With so much information available, you can never be sure you’re not overlooking some key fact pointing the way to better performance. Data Mining is the practice of sifting through all the evidence in search of previously unrecognized patterns. Some companies are even hiring Data Scientists, experts in statistics and computer science who know all the tricks for finding the signals hidden in the noise. Data Mining probably fits within the category of analytics, but most analytics is based on data from systems set up to track known KPIs—so it’s usually more measuring than mining.

Not everyone will agree on these terms, as Dennis points out in his article, but it’s a good start. As a core strategy I believe your BI should encompass all three: easy to use analytics that allows your users to ask their own questions, big data to capture MORE than just sales data, and data mining so that you can leverage all of your data for the best insights possible.

Source: http://blog.apterainc.com/

What is Cluster Analysis? And Why Use It?

Why would you want to use cluster analysis on your retail sales data? Well, cluster analysis helps you identify non-independence in your data. Here is an example to help illustrate the point. Lets say we want to ask loads of teachers from many different schools what they think of their principal. If you ask two different teachers from two different schools, you will get two completely different answers that will be independent. But, if you ask two teachers from the same school, the answers will not be completely independent and could be very similar – but not EXACTLY the same.  Now if your job was to take the raw data and try to predict which school each teacher came from based on their answer – then you have an application of clustering.

The same thing can be applied to Walmart store performance for a supplier. You have some data points for a store like how long that store has been open, how many competitors it has located in its vicinity, what was your products sales performance for that store, some demographics for that area like unemployment and population, possibly even some historic weather data. Now you use a clustering algorithm to group your stores that are most closely related. This could be the first step in identifying under performing stores and why. It could give you a viable store list for a product test based on more than sales performance. It might help you further identify your product identity and who your actual customers are using enough demographic data. You might not find anything you didn’t already know. The important thing is that you are diving into your data to truly understand it on a level you never have before, and uncovering one of these nuggets could be millions of dollars difference to your company.

Once you’ve built your base analysis, and in our case we built our report that you see above, turned it into an in-memory cube, and then built a MicroStrategy dashboard on top of it – we can then explore slicing and dicing our data along the different data points to help identify if any of the metrics in our analysis are a key contributor to a cluster alignment. This way we can determine what factor affects sales the most. Could it be store age? or store square footage? or unemployment? Ethnic breakdown? What of these are driving markdowns?

The great thing about using this analysis as a MicroStrategy dashboard is that it is pretty easy to tweak to look for your top performing stores, and refreshing the data source is very easy. In fact, this report could be automated each week and emailed to you. There might even be an application to look for cluster changes and have something like that generate an alert so you only need to be bothered if anything changes.

#ArkansasRazorbacks Set To Face Toledo In Little Rock #WooPig #Uncommon

No. 18 Arkansas will take on Toledo, a team picked to win its conference, on Saturday in Little Rock. The game kicks off at 3 p.m. on SEC Network Alternate

Game 2: Toledo| Saturday, Sept. 12 | War Memorial Stadium

Kickoff: 3 p.m. CT
TV: SEC Network Alternate

Find Out More . . .

#WooPig
#Uncommon
#NeverYield
#ArkansasRazorbacks
#GoHogs

Leveling the Playing Field: #MobileLocationAnalytics | #RetailCustomerAnalytics

This is an interesting article on how Mobile Location Analytics (aka Beacon Technology) is helping brick-and-mortar retailers compete effectively with online retailers by capturing customer behavior near and in the store.

You can read the article below, however, the question for you is how will you incorporate this new data source once it is provided to you by the retailer?

Will your rigid, difficult to modify, DSR incorporate this data stream in a timely manner – or the usual months or years that data model changes sometimes take in a data warehouse environment.

#MobileLocationAnalytics
#RetailCustomerAnalytics
#CPG
#CPGMarketing

How #PredictiveAnalytics is Changing the Retail Industry | #CPG Take Note

Ideally, a retailer’s customer data reflects the company’s success in reaching and nurturing its customers. Retailers built reports summarizing customer behavior using metrics such as conversion rate, average order value, recency of purchase and total amount spent in recent transactions. These measurements provided general insight into the behavioral tendencies of customers.

However, reports summarizing average behavior don’t provide the useful insights needed to determine how individual customers are likely to behave because general behavior tendencies are simply too broad. In order for retailers to create a meaningful dialogue with customers that honors the shopper’s preferred level and mode of engagement, it takes more than summarized reports, which is why customer intelligence and predictive analytics provide the opportunity to significantly change the retail marketing industry.

Customer intelligence is the practice of determining and delivering data-driven insights into past and predicted future customer behavior. To be effective, customer intelligence must combine raw transactional and behavioral data to generate derived measures. The process can best be described using the saying, “It’s not the data that is collected, it’s the data that is created.” Put into a predictive modeler’s perspective, the team not only collects a large amount of data, but also contextualizes that data by building derived attributes that provide additional insight into customer intent.

But how do data scientists and predictive modelers determine which derived attributes are relevant? Usually data scientists lack the deep domain expertise needed to clarify and prioritize their efforts. Therefore, a collaboration with domain experts is essential. This collaboration is like a three-legged stool. Each leg is critical to the stool remaining stable and fulfilling its intended purpose. When it comes to generating customer intelligence, the three legs of the stool are retail experts, data geeks and coders, and predictive modelers or data scientists.

Retail experts have domain expertise and can best frame the problem customer intelligence is aiming to solve. They suggest derived attributes that will provide value to both the brand and the company’s marketing campaign. Data geeks are needed to program these ideas and store them in a suitable database, which can often lead to greatly increased data storage requirements for the retailer. However, if the data can only be used to create solutions or make key marketing decisions if it’s properly stored and accessed. Inaccessible data means useless data and a wasted opportunity.

Predictive modelers and data scientists are then needed to use the stored data to build models that achieve those business objectives originally set by the retail expert. Predictive models find relationships between historic data and subsequent outcomes so that near-term and long-term customer behavior can be predicted. This leg of the stool aims to answer problems such as the likelihood of when a shopper will make their next purchase and what the value of that purchase will be. Sometimes, these relationships are so complex that only machine learning techniques will find them.

In a real world example, consider a retailer that would like to appropriately message high-valued, loyal shoppers who appear to be disengaging from the brand. A predictive model built from stored data could identify which shoppers are likely to purchase again with seven days, allowing the retailer to let them be the loyal customers they truly are. The predictive model can also show if certain shoppers are unlikely to purchase within seven days but have a high average order value. For these shoppers, the retailer could provide an incentive to bring the shoppers back to the brand. In either case, predicting what shoppers are likely to do is critical to understanding how best to complete the dialogue with them.

Moving forward, retailers will need to big data augment marketing decisions using insights gained from customer intelligence and predictive analytics. Each retailer’s data team must bring in elements from all aspects of the business, including retail experts, data geeks and predictive modelers. These key elements will set retailers up for success as we move forward into the era of big data.

#AnalyticsInRetail
#CategoryManagers
#CPG
#CPGMarketing
#NorthwestArkansas
#PredictiveAnalytics
#PredictiveAnalyticsRetail
#RetailingInNorthwestArkansas

#PredictiveAnalytics market will be worth \$5.24 billion by 2018 illuminated by new report

Interesting introductory article announcing a predictive analytics report.

The report “Predictive Analytics Market [(Fraud, Risk, Marketing, Operations), Verticals (BFSI, Healthcare, Environment, Government, Retail, Energy, Manufacturing, Transportation, Travel, Telecom, Sports)]: Worldwide Market Forecasts and Analysis (2013 – 2018)”, defines and segments the predictive analytics software market into various sub-segments with in-depth analysis and forecasting of revenues. It also identifies drivers and restraints for this market with insights on trends, opportunities, and challenges.

Global predictive analytics market is driving on the emergence of massive amount of data deluge and innovative technology implementations. Business enterprises focus has changed from traditional Business Intelligence (BI) solutions to predictive analytics, because they have understood the importance of data and its analysis for the future estimation.

Traditional BI solutions are striving to sustain in this highly competitive world. The transformation of BI to predictive analytics gives new opportunities to the big players as well as new startups in this market.

This article highlights the how big data is outpacing traditional BI – both in ability to deliver actionable insights and the technological infrastructure to handle massive amounts of data, much of which is ad hoc and unstructured.

#PredictiveAnalytics
#DataAndMarketing
#BigData
#CPG
#CPGMarketing

#ArkansasRazorbacks Football is Back for 2015 Season Opener #WooPig #Uncommon

SEC Nation Is Coming To Fayetteville

The University of Arkansas will host SEC Nation this weekend as the Razorback football program takes on UTEP at 2:30 p.m. on Saturday, Sept. 5. The show will broadcast live on Saturday morning from 8 to 11 a.m. from Victory Village North.
Find Out More . . .

#WooPig
#Uncommon
#NeverYield
#ArkansasRazorbacks
#GoHogs

What Do Marketers Really Want in #DataAndTechnology?

Marketers get data – or at least they get the importance of data. Data answers questions such as:

• Can you help me understand my customers?
• Which customers are my best customers and why?
• How can I find profitable new customers?
• How can I sell more to existing customers?
• How can I retain my existing share of each customer?
• How can I increase the velocity of my sales?
• How can I integrate my marketing through all available channels?
• How can I maximize the impact of my marketing budget?

However, data is just data unless you have the marketing technology to “make the data talk.” Marketers are increasingly in charge of marketing technology spend to drive better data outcomes. In fact, technology has become the core of marketing. According to research by IBM, marketing executives are adopting technology in the following areas:

• 88% Customer Relations

• 83% Digital Marketing

• 68% Customer Analytics

Contact us today to discover how Vortisieze analytics can take you to the corner of Marketing and Technology.

#DataAndMarketing
#BigData
#CPG
#CPGMarketing

Retailers to increase IoT spend fourfold by 2020 to US\$2.5bn | #BigData #TechnologyTrendsInRetail

Anticipating the winds of change, major retailers are expected to increase their investment in internet of things (IoT) technology fourfold to US\$2.5bn by 2020.

In the commercial space, IoT in retail is one of the clearest examples where connected network technology could have a significant impact on customer interaction in stores.

In particular, retailers have already begun investing in IoT hardware, including Bluetooth beacons and radio frequency ID (RFID) tags that allow them to send information to them in-store and keep a tab on stock and price levels, respectively.