Question

Market Research: Find What Consumers Value Most

There are plenty of stats that can tell you consumers are interested in other’s opinions and experiences. These statistics reveal that at least 90% of us are influenced by what we read. Even more so if it’s a negative review – the sentiment resonates. In recent years, multiple sites have collected reviews for local eateries, vacation destinations, and, of course, consumer products.

A positive review is a form of social proof.

If your company is considering entering a new market or needs to research product ideas, why not start with online reviews from real users? This was the very idea that drove a major Amazon case study on pricing by a young analyst team.

They wanted to understand the best speakers available to purchase at the $150 price point. They theorized that if you’re going to develop and market a new product, it is useful to understand what features are most valued. What a great business use case and reasonable example of analytics!

The team extracted data for five speakers based on popular brands that Amazon customers had reviewed. The data contained consumer ranking, price, and all customer reviews.

This data was a mix of structured data (ratings, price) and unstructured data (review text).

Using the customer rating, these junior data scientists wanted to learn which product characteristics influenced scores. The following figure shows the products with the final text topic extraction analysis.

It makes sense that a speaker’s most outstanding quality should be its sound quality. There are many choices in this market space. Within the target price range, consumers must choose the most valuable features.

By reviewing low-rating topics and considering the reviewers’ sentiment toward that topic, you learn which features are essential. Its battery life, speaker material, and a charge port.

The business analysis paid off! But more importantly, the business is able to understand the customer experience.

When starting your speaker re-design or even a marketing campaign, you understand what features are essential to consumers. This market research from this data could have been expanded to include multiple sites or even all products. It is vital to understand what features drive purchasing decisions and leads to the most product dissatisfaction.

Questions:

1. The analysis shown in this case study is typical descriptive analysis, as it identifies what has happened. Based on this scenario, please give an example of predictive analysis and an example of prescriptive analysis, as well as discuss the needed input data to each in the context of the case study

2. To identity which features influence customers most, you may need to use association rule mining to find which features go with high ratings or low ratings more frequently. If we find a rule {charger}=>{low rating}, [35%, 65%], what are the meanings of 35% and 65%?

3. Following Question 2, if we have found another rule {noise}->{low rating}, [35%, 75%]. How will you decide which feature is more influential to customers, and why?

Answer 1

Solution:

Answer to Q 1):From this Case study we have to use online Product reviews to be used to estimate the product quality and by that its marketing strategy is to be planned. If a product has good online reviews it has a very good market so that profitability will be increased also.So this case study focussed on Amazon's Reviews database which has ratings,price and all review comments as unstructured data, by analyse this, product market can be predicted and user can decide the best and appropriate product of Price,quality and service facility etc.

This case study uses 3 types of Analysis:

1) Descriptive Analysis

2) Predictive Analysis

3) Prescriptive Analysis

1) Descriptive Analysis: This provides Product positive descriptions as well as negative descriptions along with Delivery feed back, e-commerce web site reliability of Return policy and Secure payments etc. This provides Price of the product,Rating(1 to 5 stars),specific comments on various features of product in pro and cons manner. For example a mobile gadget of a company : Overall rating , Battery life, Screen Resolution,Operating system, RAM,Processor, Charger, Warranty,Insurance etc parameters will be analysed and ranked. So descriptive analysis is used to characterise the pros and cons of customer experience with product and amazon(ecommerce tool)

2) Predictive Analysis: This is used to predict the future market of a particular prodcut based on the present review statistics of the product on multiple websites.A new customer when he want to buy any product(Mobile or Tv or Laptop etc) he will go through the reviews of that product on online websites.If he found satisfactory report from that review he will proceed to buy else he will choose alternate product. In this manner predictive analysis is used to buy a new product of a specific company by a specific type of customer will be predicted using Association analysis as well as Regression modeling.

3) Prescriptive Analysis shows the constraints of customer to buy/choose a product model or the specific conditions to occur a prodcut sale will be shown by a prescriptive Analysis like which customer of particular age,area,education,period will be shown. Like laptops purchased by university students during June to August .

Answer to Q 2) Association Analysis is also called as market basket analysis which is used to find the buying habits of the customer and which items will be purchased by a customer during a single visit to a shopping mall. This will be determined by analyse the Transactional database of customers. Association find list of items a customer buy along with another Item. Like when a person buys milk how likely he also buys bread and/or eggs.

The given Rule is :

{charger}---->{low Rating} [35%,65%]

Association rule is shown above it has two measures ie Support and Confidence

Here Support is 35 % ,this meansThere are 35% Chargers among all the transactions are low rating

Support count will be calculated based on percentage of an Item over total transactions. Here 35% is the minimum support count. Among total transactions 35% chargers are low rating

Another metric is Confidence which 65% here. confidence specifies how may batteries having low rating among the trasactions of batteries only. here minumum confidence is 65%.

So association rule is said to be strong when it has above 35% support and above 65% confidence count else rule is weak so need not be considered for marketing purpose.

Answer to Q 3)

{Noise}---->{low Rating} [35%,75%]

Similarly Support count of low rating due to noise is minimum 35% from all trasactions

and Confidence count is 75% having low rating for the prodcuts having noise issues only.

Rule is strong if support is >35% and Confidence is >75% else Rule is weak

So these two rules are used to applicable for finding good quality gadgets with better battery life and No noise problem