4 ways data analytics can transform the retail sector

4 ways data analytics can transform the retail sector

Nakala Analytics Blog

The retail sector in Kenya is projected to be among the greatest contributors to the Gross domestic product in Kenya by 2030. However, over recent years the percentage of GDP contribution from the sector has been declining steadily. Two of the major retailers in the country have run out of business within 2 years. Though the entrance of international brands into the Kenyan retail market has greatly improved competition within the sector, it may, however, eliminate some of the local brands. This article aims to show how data analytics creates a competitive advantage within the retail sector

How is data analytics transforming the retail sector?

  1. Customer Acquisition and retention: Data analytics is helping retail stores answer the following questions;
  • Who are our customers?
  • What do our customers want or like?
  • When and Which mode of delivery do our customers prefer.

By clearly understanding customer behaviors a retail store is able to acquire and retain customers. Thorough customer data analytics enables the retailer to make personalized recommendations, give relevant offers and use customer feedback to improve customer experience.

  1. Sales boosting: Insights of next sell, upsell and cross-sell opportunities determine which promotions to offer to different customers. By personalizing the shopping of repeat buyers, a retailer will definitely increase the sales volume.
  1. Inventory management: This ranges from the positioning of products on shelves to re-order and restocking of goods in the stores. Data analytics gives insights on the impact of repositioning of goods on the shelves, real-time stock situation and sales fluctuations.
  1. Customer Experience: Data analytics helps create an appealing and attractive store layout, online websites and offers customers personalized experiences. Sentiment analysis of social media streams and customer feedback gives important insights on various retail products and services for decision making.

Here are some case studies of retail stores that have embraced data and analytics

1. Netflix vs Blockbuster

Netflix is a media services provider and production company founded in 1997. It started as a video rental-by- mail service provider at a time when Blockbuster had dominated the industry with over 2800 stores around the world. Blockbuster was a brick-and-mortar store while Netflix concentrated on an online platform. However, Blockbuster filed for bankruptcy in 2010 whereas Netflix’s subscribers have risen to over 151 million.

2. How did Netflix outplay Blockbuster?

One of the major reasons why Netflix succeeded over Blockbuster was customer focus and the ability to adapt to change. While Blockbuster was charging late unnecessary fees to its customers, Netflix offered subscriptions to its customers enabling them to watch a video as long as they wanted. This attracted the customers who were annoyed by the late fees charged by Blockbuster. Netflix further increased the ease of access to the videos.

3. How Netflix is using data analytics for growth.

The brand creates a detailed portfolio of its subscribers using data points collected from the customer’s interaction and response to its services. The portfolio is analyzed to discover customer behavior and patterns. These insights are used to customize marketing and make personalized recommendations to each subscriber. The customer feedback program through thumps up vs thumbs down on a particular TV show has significantly increased engagement with consumers.

How has Netflix benefited from consumer data analysis?

  • 75% of Netflix’s viewer activity is based on personalized recommendations.
  • The brand is earning over a billion in customer retention since the recommendation system accounts for over 80% of the content streamed.
  • An impressive 93% customer retention.

4. Amazon vs Toys r us

Amazon Inc. is a multinational technology company that focuses on e-commerce, cloud computing digital streaming and artificial intelligence.  Toys r us was one of the largest children’s toys retailer in the USA, founded in 1948 but filed for bankruptcy in 2018. Despite having market dominance over a long period of time the brick-and-mortar retailer faced stiff competition from online vendors like Amazon which forced her out of business.

How did Toy r us lose to Amazon?

The company didn’t develop its online platform during its 10-year partnership with Amazon thus since most consumers were adopting online shopping, it lost the majority of its clients. Amazon had also included other competitors in its toy retail sector.

Having lost the competition in technology Toys r us began competing on price alone. This approach failed since in a competitive environment there will always be a cheaper alternative. Thus failure to adapt to changing business environments and customer’s preferences greatly contributed to the downfall.

What is Amazon doing to remain competitive?

Through its site, Amazon collets information on how its clients interact with the various elements of the site which processes to improve site performance. It further uses predictive analytics for targeted marketing and personalized recommendations to improve customer satisfaction and loyalty.

Can data analytics save the slowly dying retail sector in Kenya?

Nakumatt and Uchumi supermarkets case study

Despite having dominated the Kenyan retail market for many years, the retail giants were faced with numerous challenges and forced to quit. One of the main reasons was the massive debts they incurred after failing to pay their creditors and suppliers.

Could data analytics have saved the situation?

Data analytics could have been applied to;

  1. Monitor whether products are meeting sales target
  2. Monitor real-time stock situation
  3. Make predictions of future stock orders on certain products.
  4. Monitor sales revenue and make comparisons to costs on stock at specified periods of time
  5. Inform management of any irregularities in finance management.

With the above applications, the retail giants could have survived.

Conclusion

Over recent years, data analytics has proven to be a powerful game-changer in the retail sector. Retailers who have embraced the role that analytics play in customer segmentation, product recommendation, internal operations, and market analysis have gained a competitive advantage over competitors. In Kenya, though majority of the retailers are yet to fully integrate analytics into their operations, the field proves to be significant for most competitive markets. Attracting customer loyalty trust and satisfaction has been a core component for almost all successful brands. This can be achieved through adequate customer data analytics.

References

  1. Study on Kenya's retail sector prompt payment.

http://www.trade.go.ke/sites/default/files/Study%20on%20Kenya%20Retail%20Trade%20Sector%20Prompt%20Payment%2C%20June%202017_0.pdf

  1. The Downfall of Toys R Us — Don’t Blame Amazon!

https://medium.com/@brand_minds/the-downfall-of-toys-r-us-dont-blame-amazon-c88856516383

  1. Here's how Amazon may have led to Toys "R" Us' demise

https://www.businessinsider.com/heres-how-amazon-may-have-led-toys-r-us-demise-2017-9?IR=T

 

 

 

 

What is a data audit for data science?

What is a data audit for data science?

Nakala Analytics Blog

What is a Data Audit for data science

A data audit for data innovation is a process which seeks to provide recommendations that facilitate realistic collection, processing, usage, securing, transmission and storage of data in a manner that supports affluent innovation and attainment of expected ROI. 90% of all data analytics projects fail because businesses have not put in time to understand whether the existing data assets can help the business achieve its strategic milestones. Data assets simply refers to all components that are involved in capture, storage and consumption of outputs of a business process which could be an application output file, document, database, or teams that companies use to generate revenues. Data assets are some of the most valuable assets in the technology era, and organizations spend billions of dollars to manage the assets.

A data audit takes an inventory of your data assets and provides guidelines on its usability. Critical to performing a data audit Do you know what data you have available? Are you making the most of it or Do you know how to make use of it?

Why conduct a data audit before a data innovation project?

A data audit seeks to answer three key questions.

  1. From your existing data (primary & secondary), are you able to address all your strategic objectives?
  2. Are there any anomalies in your data that could prevent you from achieving your business goals?
  3. What is the ‘Rich & Reach’ in your data?

 Lack of a data audit for data innovation?

This is so far the most important stage before a business invests in a data project. A data audit for data innovation is a process which seeks to provide recommendations that facilitate realistic collection, processing, usage, securing, transmission and storage of data in a manner that supports affluent innovation and attainment of expected ROI. To perform a data audit for data science means being able to:

  1. Review all existing data and data sources (both primary and secondary).
  2. Match strategic objectives with data assets available.
  3. Identify gaps in the data ecosystem that may prevent the organization from achieving its strategic objectives and recommending way forward.
  4. Identify additional innovation avenues that the organization has not thought about also called the “Data Rich & Reach”.
  5. Assess data collection tools and data pipelines.
  6. Assess organization wide data storage and protection needs.
  7. Assess data collection points, identify gaps and recommend corrective measures hitherto.

To jump start your data innovation journey and realize a profitable venture, seek to create a data strategy which works hand in hand with a data audit for data innovation. The data innovation strategy will create an ideal road map bespoke to your organizations current and future market positioning. The deliverable will present a report which maps strategic pillars to profitable initiatives, creating a clear road map on products, services and data innovation for future growth and profitability.

To be sure of your data investment, the data audit will identify content and structural gaps in the data ecosystem that may prevent the organization from achieving its strategic objectives. The deliverables will consist of a detailed report with recommendations and adjustments in readiness for data innovation.

 How to conduct a data audit for data science the Nakala Way?

In Nakala Analytics, we have adopted a unique cost effective way to help you jump start your data innovation journey. Our essential building blocks for creating a workable data strategy demands that we give each & every business unit a chance to think through the past and envisage the future.

  1. Create a detailed data collection process which seeks to understand the current positioning of the business as well as the needs of the organization.
  2. Evaluate all data resources while looking out for areas of innovation that will maximize achievement of your objectives.
  3. Prepare a data report and proceed to create your data strategy.
  4. Following an environmental scan and a thorough data audit, SWOT Analysis will be used as a tool to evaluate internal and external influences which are dependent on data and its effects towards supporting the overall mission and vision of the company now and in the future.
To implement the data strategy, a broad range of efforts which focus on the transformation of strategic intentions into action shall be undertaken. 
Sentiment analysis on customer feedback

Sentiment analysis on customer feedback

Nakala Analytics Blog

Without doubt, big data have caught the attention of many businesses across the globe. The availability and richness of data available in public data sources have immensely grown to such levels, that it cannot be ignored. Efforts have been made to analyze and make sense of structured sources. However, unstructured data which comes in many shapes and forms is yet to be utilized to full scale. One type unstructured data sources in text available on social media sites. Just like the viral nature of word of mouth, the digital shapes present eWOM.

An eWOM is a form of communication defined as “a statement made by a potential, actual or former customer about a product or company which is made available to people via the internet”.

eWOMS are less personal, they are emotional and reflect the subconscious opinion. Only 20% of the tweets mentioning a company’s brand contains sentiment, according to Jansen et al. (2009) study. As a result, a great deal of data is required to get a valid sentiment scoring.

Among many other methods that can be used to extract insights from social media sites is sentiment analysis. I believe much has been said about sentiment analysis and as you take time to read through this piece, you are probably familiar with the methods of how to extract insight using sentimental analysis methods. Just a quick mention, to classify an eWOM, you might consider the following methods (not exhaustive)

1.      Topic in the text– Job advert, product advert, news, clothing

2.      Type of user – Person or Organization

3.      Sentiment – Positive, Neutral & Negative

4.      Text content e.g.! Indicates happiness? means doubt

5.      Intent in text

Having been in this industry for a while, I would like to bring in business contexts to business users about the power of social media data in forecasting sales. To start off, I feel obliged to define sentimental analysis? Sentiment analysis is a method of extracting value from unstructured textual data in order to mine people’s opinions by classifying them as either positive or negative. Opinions could emanate both from peoples public profiles or organization profiles.

No alt text provided for this image

 

In spite of the diversity of content found in online sites, predicting product sales is highly domain specific. For sales prediction to be effective, your product must receive a lot of attention which means a lot of review. Consumers are increasingly posting their opinions on social media, commenting on their experiences with products and services they purchased. To my amazement, Positive tweets by organizations do not correlated to sales.

Fortunately, positive tweets of people have more effect than by organizations. Social media activity does not affect products and services that are less socially-oriented. There is a strong evidence that social media activity has a strong effect on sales. By this, positive tweets by persons can be used to forecast sales

While doing the analysis of how online reviews can affect product or service sales, I often used to ask myself question. I believe I share my sentiments with many of you outside there.

Is there a relationship between tweets of a certain type and sales in following weeks?

Well, there seems to be a high correlation between positive tweets by persons and sales in the third and fourth week following the tweets.

To what extent the number of tweets of a certain type can be used to predict sales?

Interesting, tweets can be used to forecast sales 5 weeks after people comment positively about your product or service.

Does the high number of tweets lead to an increase in sales in the following weeks.

There is some evidence that suggests that a high number of positive personal tweets is followed by an increase in sales. During high peaks, there is a chance of reaching more positive people

Pipeline for sentiment analysis for sales people

  • Identify an inbound lead (Person, Company)
  • Fetch leads online comments/reviews/posts/networks
  • Classify opinions based on topic, intent, type, sentiment etc.
  • Pitch to customer based on their current experience or challenges
  • Allow user to share opinion with other public users online to create a ripple effect

In addition to sales, there exists a growing body of knowledge on [predicting large social and economic events, in particular

  • Unemployment rates
  • Influenza epidemics – ability to predict peaks of influenza infection rates
  • Election prediction – The fraction of attention certain parties receive on twitter corresponds to the outcome of the elections.
  • Furthermore, sentiment analysis of tweets has proven useful in a financial context. One example is the study conducted by Bollen, Mao, and Zeng (2011) that uses it as a tool in order to predict the stock market. They performed a sentiment analysis on tweets about a company. The result was then compared to the share price. It was proven that a correlation between the share price and the sentiment could be found.
Data Analytics Strategy

Data Analytics Strategy

Nakala Analytics Blog
Sports Analytics

Sports Analytics

Nakala Analytics Blog

As I was watching AFCON 2019 kick off ceremony, I thought to myself, what factors could result in a national team being eliminated at the group stage of the competition? Like many other people, I believe you are anxious to see your countries move the stages! Should you really trade your emotions?

I decided to pull AFCON 2017 results with an objective to predict the likelihood of a team progressing past the group stages. The final results were impressive with an ROC score of 0.833 running a decision tree classification which emerged as the best model among many other.

No alt text provided for this image

For those of us who are not familiar with machine learning, A decision tree is a decision support tool that uses a tree-like graph to model possible event outcomes. The ROC (Receiver Operating Characteristic) curve tells us about how good the model can distinguish between two things in this case if a team will advance to the knockout stage or not! ROC is a probability curve and AUC represents degree or measure of separability. An ROC score of 0.833 is pretty good for football match outcome prediction. These results were computed on a python backend.

Taking a look at other summary metrics:

No alt text provided for this image

Some variables I processed include:

  1. Number of historic appearances
  2. Historical matches drawn
  3. GA
  4. GD
  5. GF
  6. LOST
  7. PLD
  8. PTS
  9. WON
  10. Did they qualify at the Qualification stage?
  11. Matched lost at Qualification stage
  12. # Qualification Matched Played
  13. Qualification Stage Points
  14. # Matches won at Qualification stage
  15. # AFCON titles
  16. Qualification stage Group winners
  17. Qualification stage Group Runners up
  18. Qualification stage Group Others

The resultant decision tree looks like this:

No alt text provided for this image

Variables of Importance

No alt text provided for this image

Model Evaluation Parameters

Confusion matrix

* “Optimal” cut was found by optimizing for F1 Score. One way to assess a classification model's performance is to use a "confusion matrix", which compares actual values (from the test set) to predicted values.  

No alt text provided for this image

 

 

 

 

 

What is the quality of the model

No alt text provided for this image

This density chart illustrates how the model succeeds in recognizing teams which will proceed from the ones which wont. It shows the repartition of the actual classes in the validation set according to the predicted probability of being of said class learnt by the model. The two density functions show the probability density of rows in the validation set that actually belong to the observed class vs rows that don't.

A perfect model fully separates the density functions:

·      the colored areas should not overlap

·      the density function of Advance should be entirely on the left

·      the density function of Eliminated should be entirely on the right

The dotted vertical lines mark the medians.

No alt text provided for this image

Testing the model on Kenyas performance at the AFCON 2019, it appears that Kenya will be eliminated at the group stages.

 

Subcategories

Videos

Initiatives