• ADG Consultants

Technology: Finding success with Data Science


Introduction


Nowadays analysing Big Data has become an integral tool in optimising business performance and Big Data has transformed every industry by changing the way business make decisions from heuristic to a more objective and data driven approach. As a business owner, you have been hearing a lot about Data Science, Big Data and the dawn of the new digital era. Have you ever wondered how your business may also be able to leverage Data Science and Big Data to revolutionise your business strategy and unleash the true potential of your ideas?


Before we all jump onto the bandwagon of Big Data or Data Science and start hiring Data Scientists, we should first understand how these technologies are actually implemented in business across various contexts. Therefore in this article, we shall explore how some business have been leveraging data, statistics and machine learning to achieve tangible impact through increasing work efficiency and improving business value.



Customer insight and targeting


“A process to understand how different subsets of the population engage with your business and devising effective strategies to increase engagement”


Have you ever been on social media and receive advertising for a product or service that is relevant to you? Google, Facebook and many other service providers are collecting customer information such as personal profile, web browsing behaviour, product purchase history and others. Business who have been successful with online marketing often make use of such data by analysing the group of people who are highly engaged with these ads, analysing the click-through rates and conversion rates of various subsets of the population and constructing customer segments to model the behaviours associated with these subsets of the population.


Imagine an online bicycle retailer who is currently running multiple online ads to attract customers and increase engagement. Without understanding customer profiles, the retailer may heuristically decide to maximise reach by sending out the ad to everyone online which would not be cost effective. In an alternative scenario, the retailer may have done some analysis with their data and decided that there are three segments of customer. In segment A there are people who are cycling enthusiasts and often browse online for new gear. Segment B are people who are not cycling enthusiasts but based on their online activities and browsing behaviour, it is likely that they are going to purchase a bicycle within the next month. And finally there is segment C who has absolutely no interest in purchasing bicycles. Although there may be benefit in marketing to both segment A and B, it is segment B where the business marketing strategy may see the highest return on investment. Furthermore, it is also possible to tailer messaging and offers based on the in depth understanding of each of the aforementioned segments.



Customer retention


“Customer attrition is a process in which customer engagement has fallen below a certain level and results in a loss the business from the customer, customer retention is the process of retaining the business of these customers”


Customer attrition very often arise from customers switching to the product or service provided by a competitor or the natural lifecycle of that customer has run its own course. In the latter case, there is likely very little the business can do to proactively retain the business of the customer. However, in the first scenario there are many tactical and strategic solutions to improve customer retention.


Telecommunication service providers often face the above issue as customers can be easily tempted to switch services to a competitor who promises lower fees or other benefits with service plans. By using data, it is possible to identify the customers who are most likely to attrite given change in circumstances external to the business. Once this subset of customers have been identified, the business can devise plans and offers specifically catering to the requirements of these customers to reduce the likelihood of attrition. For example, a plan with lower fees could be offered if it is known that cost of service is the main reason for attrition given the similarity of the customer’s profile with other customers in the past who have switched services to a competitor. Through this process, the business can reduce customer attrition and improve customer retention.



Price optimisation


“A process to determine how customers respond to different prices for its products and services and finding the price that will maximise operating profit”


You may have experience browsing for flight tickets online and find that the prices constantly change depending on relative time from the scheduled date of flight, number of seats already reserved, and even the number of times you have searched up that flight. The constantly changing price for the ticket is precisely price optimisation through optimisation models that look at a number of factors including those relating to the flight and destination, and those relating to customer profile and browsing behaviour.


By leveraging data, airlines are able to provide different offers to customers to maximise their operating profit. For example, we have customer A who has very little travel experience with the airline and based on their browsing history it can be determined that they are very cost-conscious. On the other side we have customer B who has a long history of business with the airline and has historically purchased flight tickets that are on the higher end of the cost curve. Based on this information, a price optimisation model may offer customer A a lower cost ticket to persuade them to choose the airline, but may decide to offer customer B a much high cost ticket knowing that there is a high chance they would take up the offer anyway.


This is also a practice common in the insurance industry where tenure or loyalty is often associated with higher pricing for products and services, along with a range of other customer attributes and circumstantial factors.



Personalisation


“A process to refine customer experience such that it is relevant and contextual for the individual’s needs and wants”


This is one of the most common business use case of Data Science and Big Data, and is a capability that is at the very core of digital marketing. While some business may use a “one size fit all” strategy, your business can create experience that is perfectly tailored to your customer’s needs and interests.


For instance, online news media can populate the home page of their web or mobile application with news articles that are most relevant to the reader. This means the news media no longer have to rely on heuristic approaches to manually handpick stories that they think may be most important to all the viewers, but they can dynamically display stories or articles depending on the reader’s profile and browsing history and therefore personalising the experiencing of each reader. Furthermore, online ads that are often displayed next to the articles can also be optimised through personalisation to maximise operating profit and engagement as the platform can guarantee business higher levels of click-throughs and sales.



Demand forecast


“A process to forecast demands of various products and services such that the business can devise strategies to prepare for changes in demand and adjust inventory or prices accordingly”


It is important for big supply chains and large retailers to be able to decide precisely how much, when and which products should be stocked in inventory based on demand, historical trends and seasonality. Many supermarkets rely on analytics of data to understand the changes in demand and distribute their products accordingly to the various branches and stores. For example, a supermarket chain may leverage weather data to determine that there will be a heat wave in the coming month in certain areas. This can then lead to a sudden spike in demand for bottled water which the supermarket can then be prepared for and prioritise shipping of water to the affected areas according to their predictions.


Similarly, hotels can use data to analyse historical demands for rooms during both on and off-peak seasons and adjust the cost of hotel rooms accordingly to maximise their operating profit.



Conclusion


Business across the globe in many industries have already invested significantly into adopting and implementing Big Data and Data Science in order to succeed in the new digital era. In this article, we provided some examples of how the technologies have been implemented to drive tangible business benefits and impact and we hope that this inspires business owners to have a think of what their business can look like with the aid of data and what business use case there can be with the technologies.


While there are many benefits in adopting new technologies, we at Aurum (Data Governance) Consultants Limited also recognise the challenges that could arise in this transformation journey, including culture change, finding the right talents, developing the right expertise and how to measure impact. Perhaps the most challenging of all is to kick off the process of capturing data, storing and using data in a way that aligns with regulations and standards as enforced by the Personal Data (Privacy) Ordinance (PDPO) in Hong Kong.


At Aurum (Data Governance) Consultants Limited, we specialise in data risk impact assessment, data protection auditing, data breach handling, data policy design and data science advisory works. Our goal is to help businesses reduce data risk in their process of digital transformation and to help organisations drive business decisions through actionable insights by leveraging data, machine learning and automation.


If you would like to learn more about our consultation services, please feel free to reach us at info@aurumconsultancy.co or call us at +852 3725 4806.

26 views

© 2020 by Aurum (Data Governance) Consultants Ltd