Alteryx provides retailers with the unique ability to easily discover, prep, blend and analyze all of their data using a repeatable workflow, which can be deployed and shared at scale for deeper insights, insights that come within hours, not weeks. Retail analysts can utilize the Alteryx platform in a multitude of ways, including connecting to and cleansing data from their data warehouses, sending and receiving data sets from cloud applications, as well as gleaning information from spreadsheets or a whole host of other sources. Retail analysts can easily join their data together, then perform analytics—descriptive, dignostic, predictive, prescriptive, statistical and spatial—using Alteryx's intuitive user interface, which is, if not a code free environment, then a very code friendly one.

The emergence of new channels has changed the way customers interact and shop. To engage smart-device toting, always-on consumers, retailers need to better understand their preferences, anticipate demand changes faster, and personalize communications and promotions – all to stay ahead of the competition. In this fight for customer attention, data is both the biggest ally and the biggest challenge for retailers. Alteryx enables retailers to derive deeper insights about customers, transactions and operations, across all channels, without any coding or IT intervention.

Customer Churn Modelling

Customer retention is key to sustaining cash flow and profitability. In today’s big data and AI environments, understanding ex-post facto churn no longer strategically works. Organizations need to predict churn proactively to prevent it. For an industry as competitive as the retail industry, retailers need to understand the factors leading to churn long before the customers intends to walk away. Intelligencia has built a chunr model utilizing Alteryx designer that is almost as simple as plug-and-play. 

 Alteryx Workflow

 

Customer Segmentation

Customer segmentation is a deceptively simple-sounding concept. Broadly speaking, the goal is to divide customers into groups that share certain characteristics. There are an almost-infinite number of characteristics upon which you could divide customers, however, and the optimal characteristics and analytic approach vary depending upon the business objective. This means that there is no single, correct way to perform customer segmentation. That being said, customer segmentation is often performed using unsupervised, clustering techniques (e.g., k-means, latent class analysis, hierarchical clustering, etc.), but customer segmentation results tend to be most actionable for a business when the segments can be linked to something concrete (e.g., customer lifetime value, product proclivities, channel preference, etc.).

Customer segmentation is the process of dividing customers into groups based upon certain boundaries; clustering is one way to generate these boundaries. There is an important caveat though—clustering assumes that there are distinct clusters in the data. Oftentimes, customers are distributed more or less continuously in multivariate space, and they aren’t in neatly defined groups. A customer segmentation model provides a view of the retailer from a customer perspective: such models have many and varied applications. Customers are segmented according to what they present to the retailer. Views include:

·       Interests and needs

·       Gender and age

·       Marital status

·       Spending history

·       Demographics

·       Psychographics

Generally, the data is used to determine the appropriate segments for these views. The result of this analysis presents a detailed view of how the retailer is populated at different times and can allow for appropriate strategic decisions to be made. These decisions could be a function of marketing, operations or strategy. The output is also used for the building of acquisition models.  Other potential for analysis would be a master segmentation model that uses the preference results described. Customers are clustered based on their preferences to gain a global view of the retailer that is concise and understandable. Furthermore, such models can help measure the impact of strategic decisions.

Alteryx's K-Centroids Cluster Analysis Tool allows users to perform cluster analysis on a data set with the option of using three different algorithms: K-Means, K-Medians, and Neural Gas.

Demand Forecasting

Retailers use forecast demand distribution and stock replenishment models to support the business' need for quick and accurate short term demand forecasts. At fast fashion retailers, every 3-4 days (one forecast period), a new distribution demand forecast is developed for each branch store and new shipments are sent to restock the inventory required. Because fashion retail demand is so volatile, forecasts are made shortly before the expected date of shipment arrival. 

Time Series is comprised of a variety of tools within Alteryx, which are part of the standard Alteryx Designer License. Alteryx customers use predictive analytics to identify patterns found in historical and transactional data to identify risks as well as opportunities. Alteryx Predictive analytic tools are built on Open source R. Alteryx users are not required to know R to execute predictive models because all of the models in Alteryx are packaged into easy-to-use macro tools that only require configuration. All predictive tools are macros, and therefore not a “black box”. Macros provide the user with the flexibility to open all models and dissect the logic, as well as see and modify the R-script(s) being executed.

The following is a high-level description of the two pre-packaged time series methodologies our predictive tools support:

ARIMA  is the Autoregressive Integrated Moving Average and is the most commonly used forecasting approach.

ETS model uses exponential smoothing method and is a commonly used forecasting approach based on a weighted average of past observations with the weights declining in size for more distant past values. In short, all past values are factored in the forecast although with decreasing importance as the values are further back in time.

The following video provides a brief tutorial of the Alteryx Times Series tools:

Natural Language Processing

Natural language refers to language that is spoken and written by people, and natural language processing (NLP) attempts to extract information from the spoken and written word using algorithms. NLP encompasses active and passive modes: natural language generation (NLG), or the ability to formulate phrases that humans might emit, and natural language understanding (NLU), or the ability to build a comprehension of a phrase, what the words in the phrase refer to, and its intent. One of the major use cases for AI is sentiment analysis, which uses NLP to gain insight into how a business is seen on social media. For retailers, NLP can help them understand what customers are saying about their company and their competitors. Managers can use these insights to increase customer intelligence and customer service. Intelligencia has built an Alteryx-based NLP solution for retailers utilizing Python that can be used as a powerful social media listening tool. 

 

Alteryx Workflow