Transforming your company into a data-driven and data-aware company can be complex. Everything from knowing where to start, to executive buy-in, to grandfathered processes can slow data maturity and business growth. The journey begins with understanding the opportunities unique to your business based on your level of data maturity.
In this session, we will share findings and insights from customers, how they used this to secure executive sponsorship to ensure the data technology and business requirements were in tandem, as well as the use cases typically pursued. We will discuss the typical organizational constructs we see applicable based on the different stages of maturity and also discuss some best practices for driving best in class process for data driven transformation.
The explosion of data is catalyzing new business models and reshaping industries. No longer can you amble your way forward in the age of Big Data; the challenges are too great to address on an ad-hoc basis and the business potential too vast to simply dismiss.
3. 45% of F 200 firms want to
become and Integrated
Digital eco-system provider
47% of firms will introduce a new digital product portfolio in 18 months
69% say improving their
data strategy will be key
to their relationship with
the customer
48% believe sustainability as a
key reason to change their
digital business model by 2019
78% of F 500 organizations
have a medium to poor
Big Data and Analytics
capabilities
63% of $10bn+ firms are
witnessing their core
business model disrupted
Only 36% CEOs have a shared a
digital transformation vision
although 93% of the employees
believe it is the right thing to do
Digital and the Fourth Industrial Revolution through the numbers lens …
18. 18
Use Case: Single View of Customer
• Ability to identify # unique customers which directly impacts both the top-line ROI measurement and bottom–line optimization
• Increased customer loyalty, LFL sales, average basket size, redemption propensity on promotional activity, listing fees
• Dynamic real time targeted pricing which results in better margins from your most loyal customers
Business
Value
• Better customer experience leading to increased loyalty and customer advocacy
• Increased Marketing Effectiveness leading to higher ROI on every £ spent
• Cross selling and predictive promotional propensity means greater number of manufacturer partnerships
Why
Do It?
• Currently, Retailers, CPG firms and other manufacturers create shopper profiles based on historical data, SKU level data and
Basket data – They however struggle to marry that data with the behavioral data from multiple other channels (mobile, Social
Media, etc.) to map out the DNA of the customer and fail to predict futuristic buying patterns of customers across categories
and products
• Single View of the customer not only allows organizations the ability to create targeted campaigns based on shopping patterns
but also opens up new avenues of revenue streams through advanced marketing efforts such as cross device marketing,
beacon sensing, proximity mktg. etc.
Idea
Summary
The Single View of Customer combines historical sales data from structured systems with new, unstructured and semi-structured
data from social media, sentiment analysis, web activity, and blog posts. Single View of the customer helps create the DNA of the
consumer that can be used to target, re-target, personalize messaging to help address issues around loyalty, churn, cross-selling,
increasing the top line etc.
Innovate–Grow&Enable
19. 19
Contents
Big Data in Retail
– Digital Revolution
– Explosion of Data
Big Data Maturity Analysis
– Hortonworks Big Data Maturity Scorecard
– Retail and CPG Maturity Analysis
Big Data Use Cases
– Retail Use Case Maturity Map
– Single View of Customer
Big Data in Action
– Retail Case Study
– Call to Action
20. Quick Facts
Quick Facts
• For direct marketing, the lack of visibility into a customer’s credit and financial situation
restricted retailer's ability to pre-screen “right” customers to send the mailers
• Mismatch between Inventory Merchandising Ad Planner and Warehouse Inventory led
to incomplete sales
• Generation of various business reports took days to complete and even after that, not all
the information was available to the Business stakeholders
Situation Analysis
Innovation Strategy
• Retailer built an Enterprise Analytics platform based on Hortonworks Data Platform, breaking-
down silos and increasing historical depth of data available for analysis
• Drove targeted marketing strategy with insight driven customer segmentation analysis,
leveraging new data sources, including the available history
• Implemented near-real time simulation of new Credit Strategy with respect to approval or
decline of application process by collecting exhaustive set of variables needed for credit policy
coding for all customers
Business Impact
• Reduced Spend on Direct Mailers by optimizing mailing by Customer Segment: $3M in first 10
months of 2016 ($4.5 to $5.0M expected run rate savings)
• Reduced ads effectiveness analysis in Product Performance report: 300x improvement in
turnaround
• Reduced associate time in coding for red-flags and lookups for decline rules: 500x time
reduction in implementing credit policy
$3M
Marketing dollars saved to-
date from trimming the direct
mailers
Up to 500x
Time improvement in
implementing credit policy
Up to 45x
Time improvement in
generating Inventory
Merchandising Ad Planner
Digital revolution is transforming the industries. Data strategy is key part of this strategy and 2/3rd are working towards improving it.
In retail, there is big potential of Big Data .. 60% improvement potential in operating margins, 15-20% improvement in marketing potential. This is driven by big transformational changes in retail – mobile devices, IOT sensors and that measure customer movements in store. Cross-channel is becoming important where customers shop across the stores and online
Use the tombstone
Retail firms are in the exploration stage where firms don’t currently have full-fledged Big Data vision and strategy or primarily using structured data. But they are actively working on it – across the board average maturity scores will improve and most of them will be in Optimizing phase. This will be driven by having enterprise-wide vision and strategy, usage of unstructured data and leveraging inhouse and outsourced skill set.
Most of the firms are currently in the Exploration stage
Firms lack enterprise vision around Big Data and funding is unbudgeted
Although most of the firms still use structure data, some have started to collect unstructured data as well
Firms have also started to adopt analytical tools for project specific objectives
Big Data skills are mainly located among technologists and most of the work is outsourced
There is lack formal process for planning Big Data programs
In 2-3 years, firms are planning to attain Optimizing stage
Firms plans to attain enterprise-wide vision and alignment with sponsorship and funding
Firms expect to make big strides in storing their data through Data Lake
Firms will use tools that fit the purpose with centralized administration of tools and integration among the tools
Organizations are investing to gain advanced analytical skills and will leverage mix of in-house and outsourced skill set
Planning and budgeting for Big Data will be part of cyclical budgeting process
In vision and strategy, majority of the firms currently lack enterprise-wide vision. Funding is unbudgeted and seems to come from IT projects, there is little executive sponsorship and there is very little business case development around Big Data.
In 2-3 years, most of the firms will have enterprise-wide vision and strategy for Big Data. Funding will be part of cyclical budgeting process. There will be increased alignment among executive sponsors to support Big Data with Big Data business cases being developed
In Data and Analytics domain, most of the firms are still using structured data while discarding most of the data they collect. The firms are also focused mainly on measuring key business metrics for their business rather than doing advanced analytics.
Over the next 2-3 years, firms plan to leverage unstructured data, store it in the data lakes while keeping the data even if it isn’t being in use at that time. Firms are planning to perform advanced and predictive analytics on top of this data lake.
Currently firms store their data on-premise in the traditional EDWs. They are staring to adopt analytical tools for specific objectives but aren’t able to conduct cross-functional analysis as there is little integration of tools across the organization.
W.r.t. technology, the firms are mainly moving toward hybrid hosting strategy of on-prem and cloud based storage and analysis. As discussed, firms are planning to have data lake, which will be based on multiple Hadoop clusters with tools on top of it that will be integrated and with ability to provide cross-functional insights.
In terms of organization and skills, whatever Big Data skills firms have, are located in the IT organization currently. Firms outsourced quite a bit of work for Big Data projects. Firms also don’t yet have CoE to enable best practices in the whole organization and to achieve cross-group collaboration.
But firms expect things to be different in 2-3 years. Most of the firms are investing in gaining advanced analytical skills and expect mix of in-house and outsources Big data skills set within the organization. Majority of the firms are also planning to have centralized COE group for cross-functional collaboration and institutionalizing best practices.
Within process management, firms currently lack planning around Big Data programs as projects seem to be driven within IT, by IT budgets. Given this, there is hardly any evaluation of results from Big Data projects.
In future, majority of the firms are planning to have budgeting at either business-unit level or at the enterprise-wide. This will result in businesses looking for Big Data programs that drive new value streams and business models. With this, will come more effective measurement around Big Data projects and their outcomes.
Firms do have wide spectrum of capabilities when it comes to data security. Some have basic security and governance process while others do have enterprise-wide standards in these areas. These capabilities will only improve in 2-3 years.
We analyzed the maturity scorecard data for one of the top European retailer. The retailers already has an enterprise-wide vision and strategy, which drives rest of the organization. The form ingests and analyzes unstructured data in Hadoop to perform advanced analytical and predictive skills. It is investing in Big data skills with has already included Big Data programs in its budgeting and planning cycle.
The firm is already in Optimizing stage and will be in Transforming stage in 2-3 years