Idempotency in Data Pipelines

Insights

Are data lakes stuck in the black and white world of the big data infomercial?

Insights

Hackathon 2018

Insights
backgorund

Periodic Model Monitoring Framework for Marketing Models

Insights
Vanquis Bank Case Study -webcover

Optimizing Inventory of Slow Moving Products Using SAS Optimization

Insights

Leveraging Analytics to Optimize Call Center Operations

Insights
CoreCompete-Big-Data-World-background

Deciphering & Leveraging Retail Returns

Author Bright Masih | Senior Account Manager Retail Sector | Core Compete
Date: 27th February 2018
 

Retailers are cracking the omni-channel transformation challenge with aplomb with shoppers now hopping between device, location and channel at will, much to the delight of heavily invested brands. In line with innovation, In the UK in 2017 mobile shopping rose to 49.7% as a share of all online retail spend, compared to 42.3% in 2016. Whilst this increase is promising for retail on the whole, retailers are painfully aware that the success is not an entire win, and in some cases a double-edged sword, especially in the context of e-commerce returns costs.

consumers are naturally taking advantage of generous shipping and returns policies designed to meet expectations and encourage repeat purchase. These carefully crafted policies are designed to tempt shoppers to visit more frequently with personalisation supporting conversions. Policies also inadvertently encourage intentional returns but satisfy consumers who expect convenience at every step. Intentional returns also have the undesired effect of slicing through profits designed to be delivered by expensive and meticulously curated Christmas ad campaigns as well as early and ever more aggressive sales discounting.

Intentional returns have costly repercussions for retailers, when you consider that on average a returned item touches seven different sets of hands before the item is finally restocked, often at a significant discount before being resalable. The cost of handling forward and associated reverse logistics is seriously hampering healthy trading results. Clear Returns estimates returns cost UK retailers an eye watering £60b last year, with £20b from items bought online with approximately one in three purchases returned.

Phenomena such as “Take Back Tuesday” isn’t helping. According to the Logistics Consultancy LCP an estimated £17bn worth of goods were bought online in the Christmas period between Black Friday and Boxing Day. As 15%-20% of the goods bought online are eventually returned, £2.5bn worth of goods would be refunded which doesn’t include the processing costs. Intentional and non-intentional returns are now part of the fabric of retailing with consumers frequently purchasing several SKU’s in a bid to find the right product.

According to Geekwire Amazon lost a staggering $7.2b in 2016 on reverse logistics alone, up £2m on 2015. Even though returns are a burden, retailers know that nothing can kill a relationship faster than unfairly charging for returns or making the returns journey cumbersome Despite this we have found that many retailers appear apathetic. According to UPS, 81% of customers want an easy returns process and over 66% of online consumers read the retailers returns policy before committing to any significant purchase so the returns journey and how they will be treated is clearly important to consumers.

Just as with the initial purchase, consumers expect to know where their purchase is and what stage it is at. Retailers on the whole have failed to invest in the customer the same way when it comes to returns with consumers often left wondering whether the return has been received and processed and critically when money will be refunded. It could be argued that consumers are being treated poorly despite that fact that future stickiness and repeat purchases are dependent upon their returns satisfaction.

According to UPS over half of online consumers are dissatisfied with the online returns processes they have experienced. This isn’t an unknown issue to retailers; however, many are falling short of the expectations placed upon them by ever more demanding consumers who simply expect convenience and communication. Consumers are never going to be happy with paying for returns and online retailers need to invest and innovate to reduce the net 30% returns average.

By ensuring the process is painless and simple retailers can delight consumers, a recipe for advocacy and repeat purchase. With the lifetime value of the customer at stake, in an ever more ruthless marketplace, todays retail leaders need to face the returns issue head and be decisive. Retailers must insure data scientists unpick returns data to establish trends, examine customer feedback across every channel, identify problem outliers and model returns processes in order to understand potential.

We have found that our retail clients are addressing the online returns issue by taking a two-pronged approach. Firstly, they are seeking to understand the root causes of the customer returns by analysing the massive amounts of transactional data and product reviews through in depth, cost-effective analytics through the solutions that we have implemented. For this we consult and recommend an analytics stack which delivers the right combination of processing power, elastic scalability and affordability.

Retailers are able to attribute what portion of the problem is caused by what issue and then set-out to improve problem areas such as online sizing, product displays, inaccurate descriptions etc. We have enabled retailers to understand the returns process as a step in the customer journey, and by correlating the pre-purchase browsing intent to the returns outcomes the implemented solution is able to provide actionable insights that are leading to significant operational improvements and reductions in reverse logistics processing cost.

Retailers are also experimenting with ways to drive traffic into stores by enabling quick and easy returns in order to benefit from impulse purchases. Through the analysis of the sales data of returns customers, one of our apparel customers decided to implement a concept where instead of waiting in a queue to return an item, a shopping assistant checks you in, let’s you browse the store, then comes to you when they are ready to process your return using a mobile device.

Through a carefully crafted test they determined that through this subtle change, they were able to gain 0.72 units for every return unit they processed. Devising effective strategies to address the problem and realise the opportunity requires a precise understanding of your customers, deciphering what they are telling you through their data, then making the returns customer journey a convenient one for the customer and a profitable one for you.

Recognised as an international leader in retail sector big data analytics consultation and service delivery explore our retail sector expertise to see how we have helped retailers stay ahead of their challenges. Visit us on stand B2916 at Big Data World on the 21st and 22nd March 2018 at the Excel Arena London to hear Travis Perkins, Shop Direct and Advance Auto Parts talk about their experiences as Core Compete retail sector clients.

Core Compete is Google Cloud Partner, Hortonworks Gold Partner, SAS Gold Partner, AWS Advanced Consulting Partner, AWS Managed Service Partner and an AWS Big Data Competency Partner.

Insights

Why IT Fumbles Analytics

Why IT Fumbles Analytics
By Donald A. Marchand and Joe Peppard

In their quest to extract insights from the massive amounts of data now available from internal and external sources, many companies are spending heavily on IT tools and hiring data scientists. Yet most are struggling to achieve a worthwhile return. That’s because they treat their big data and analytics projects the same way they treat all IT projects, not realizing that the two are completely different animals.

Read more

Insights

Gartner viewpoint on Systems of Innovation and how they are different from Systems of Record

Accelerating Innovation by Adopting a Pace-Layered Application Strategy
Analyst(s): Yvonne Genovese

Gartner’s Pace-Layered Application Strategy is a methodology for categorizing, selecting, managing and governing applications to support business change, differentiation and innovation…

Read more

Insights

How to use customers’ life stage changes for deepening customer loyalty?

Lokendra Devangan
David is an investment banker based in New York. At least twice every month he visits Redwood hypermarket to shop for groceries and other personal needs. Three months ago, he married his girlfriend from China.

David was lounging around on a lazy Saturday when he received an SMS from Redwood. The message offered 40% discount on woman fashion brands at a partner retailer and 30% discount at famous Chinese restaurants in central New York.

After seeing the offer David was impressed with Redwood.  He was also surprised as most of their earlier offers were bundled for grocery items. Why did Redwood think that these offers were relevant for him?  He then smiled and thought that he will definitely avail both the offers next month on his wife’s birthday. He will gift his wife a lovely handbag and take her to one of the restaurants for dinner.

How did Redwood figure out this highly relevant offer for David?

David is a loyal customer of Redwood. Redwood Customer Loyalty team uses advanced analytics method to identify the customers’ life stage. They track the customers’ basket size, product mix and frequency of visit to segment customers into groups such as large family, family with kids, family without kids, Single, occasional visitor etc. Loyalty team observed that in recent months David’s monthly spend at Redwood store has increased and the basket composition has changed. They noticed that apart from larger baskets, transactions for his household now routinely include women’s beauty and hygiene products.

Redwood employs automated analytics where the system captures such changes in behavior, and identifies opportunities to enhance marketing around this life stage changes.

The change in the product-mix in David’s basket can be permanent or temporary (a friend or a relative visiting). Redwood loyalty team can test the hypothesis that this is infact a change in relationship status and that the partner likes Asian food. Redwood to test their hypothesis has created a special personalized offer for David and extended discounts at partner fashion retailer and restaurant chain. They also plan to send such multiple offers over the period of next two months. If David responds to offers regularly, Redwood will be able to conclude that David is married. This information can be used for more personalized offers.

Redwood has realized the power of data analytics and used it efficiently to enhance the relationship with the loyal customers.

Insights

Need More Capacity for your Data Warehouse:  We can save you at least $990,000

Data volumes in your organization are exploding and your data warehouse is bursting at the seams. You have heard of all the hype about Hadoop but it is not well-suited for your workloads of real-time reporting. You don’t seem to have any other options other than to buy that dreaded expansion license (1 TB = $ 1 Million) for your data warehouse. Does this sound like your predicament? Read on, it’s worth at least $990,000.

For one of our customers utilizing the Core Compete A3 service, we identified a model where they could move a majority of the newer, transient work-loads onto a cloud based data warehouse, while utilizing their current data warehouse for traditional workloads. Leveraging Amazon’s Redshift data warehouse as a service, we proved to them that data from syndicated data providers can be made ready for analysis by analysts in 1-2 weeks cutting down the 3-6 month process that is currently employed by the standard data warehousing process. This significantly increased their ability to respond to demand signals when they launched a new drug.

The cost of the AWS Redshift solution, was less than $10,000/ Terabyte. In addition, this system was integrated into other services such as AWS S3, that allowed the customer to store a significantly larger volume of data in an environment that is a fraction of a traditional data warehouse cost (1/10000th), while being able to bring it online for analysis in a matter of minutes.

One of the early concerns that the customer had was around the security model of the cloud based solution and potential risks around this. However, through a series of reviews, we were able to convince that our AWS based architecture was designed to address the demanding security expectations of the client.

We integrated AWS Redshift into the client’s SAS and Tableau environments. The enterprise class SAS Environment which was behind the client’s firewall, was able to send SQL pass-through commands to Redshift, allowing us to leverage the MPP capabilities of AWS and bring back summarized data into SAS for further analysis. The Tableau server environment was deployed directly on top of the Redshift database to enable real-time access to the data for the dashboards that are now widely used to understand launch performance.

Now, the client is also engaged with us for a variety of pilots with integrated delivery networks (IDN), where they bring the IDN data into the Core Compete A3 Service to identify insights specific to the IDN to reduce re-admissions and increase Medicare related shared savings for the IDN.

Insights

Forecast Health: How we built a Secure HIPAA Compliant Big Data Environment leveraging AWS

How does your doctor make sure that you don’t get re-admitted? Generally, they are awash with data about your health, but have very limited data on your behavior outside the health system. On top of it, they really don’t know how to synthesize all this information in the limited time they have with their patients to come up with an actionable and tailored set of actions.

Forecast Health, a Durham, NC based start-up precisely aims to address this problem. Synthesizing data from Electronic Health Records (EHR), Insurance Claims and more than 100 consumer behavior oriented data sets, they provide actionable predictive risk scores that are integrated directly into the EHR system that the physician already uses.

While what they do is amazing in its own right. What’s even more amazing is that they were able to get from the ground up and running with a lean staff in less than 60 days. Leveraging Amazon Web Services, Core Compete built a HIPAA Compliant Cloud Environment that enabled the Forecast Health staff to bring large datasets from Center for Medicare Services (CMS), syndicated data providers that provided patient level identifiable data and from within the health system in a cost-effective manner.  In addition, leveraging the elasticity of the AWS cloud and the variety of on-demand services such as EMR (Hadoop) and RedShift, Core Compete provided an environment that was designed to scale for the most highly demanded workloads. The entire environment was validated by external auditors.

“The Core Compete team enabled us to get up and running very quickly allowing us to prove our capabilities very quickly to both our investors and our early customers. Without Core Compete and the AWS cloud, our business would have never taken off”, said Dr. Michael Cousins, President, Forecast Health.

Insights
Our site uses cookies to improve your experience. You can control cookies by adjusting your browser or device settings. By using our site and closing this box, you consent to our use of cookies.