In this age of rapid e-commerce growth, online product reviews hold a great deal of clout over customer shopping behaviors.

Online reviews essentially function as actual referrals for the product or the business itself and are treated like gatekeepers to the eventual purchase by the customers (see the Spiegel Research Center’s “How Online Reviews Influence Sales”).

The first thing a consumer does before buying a product online is to scrutinize its reviews and gauge the overall sentiment. If the average sentiment is negative, they look for alternative options. On the other hand, if the sentiment is veering more towards positive, they are more likely to go ahead with the purchase. Thus, the reviews play a powerful role in whether the product itself gets sold or not.

The difficulty of achieving uniformly positive reviews

Reviews are generally given both for the quality of the product as well as various related factors, such as shipping speed, packaging, ease of placing the order, availability of online information, and so on. With millions of people shopping online every day, I believe, from my very brief personal side hustle years ago, it is impossible to get 100% positive reviews in the long run. If the product and the service is good, you will inevitably have outstanding positive reviews, but it would never be 100% perfect. As the saying goes, you simply cannot please everyone.

A predominantly bad review could severely impact the sales of the product. Oftentimes, the support department of the companies respond to negative reviews expeditiously by working with the customer to resolve the issue. However, it is not guaranteed that the customer will go back to update the negative review, even when a replacement is delivered promptly or a refund is issued.

Some sellers also choose to take their listing off if the bad reviews climb too high, but then they lose the visibility of history of sales, which is also a big driving factor in influencing the sales.

Handling negative reviews

Various studies have been done on handling negative reviews, and a ton of related material is available online in this regard. Personally, after you have scaled to a certain size, it is far more important and outright efficient to prevent negative reviews from cropping up in the first place, as opposed to trying to rectify them after they have already been posted.

Of course, one can argue that the obvious way to achieve this is to simply provide good quality products and related services that would automatically lead to positive reviews. However, a seasoned online seller would agree that, sooner or later, you will run into a displeased customer.

What if we somehow get to know in advance that a customer is less than delighted with his/her purchase and there is a high probability of a negative review, far before the review has actually been placed on the website? Would that information be beneficial to the seller? Is there anything the seller could choose to do with that information to redress the situation and avoid a negative review?

Avoiding a negative review might seem unfair to future buyers but remember that the customer still retains the dreaded power to post his/her overall experience, regardless of how the situation panned out. Furthermore, sellers can always enhance the feedback pages by bringing in the transparency by displaying the remedial measures taken.

How to “catch” potential negative reviews using AI and Big Data

In this article, I want to theoretically illustrate how we can “catch” such a potential negative review before it actually occurs so that an honest seller has an opportunity to rectify the mistake before if affects the business.

Whenever a purchase goes wrong, there are a few things that a typical customer may do in response:

  1. Browse the same product’s webpage again to check the reviews to see if anybody else has had a similar issue and if there was a helpful response from the seller for the same.
  2. Contact the customer care service to lodge a complaint.
  3. Browse similar products of other brands on the same website or on a different one.
  4. Change their purchasing pattern (mostly a downward trend).
  5. Go ahead and place an order for the same product on the same website or buy from a local retail store
  6. Discuss the bad purchase in social media.

Now, what if the company employs artificial intelligence software implemented on a Big Data distributed processing engine on cloud that will constantly scan all the above data points and look for an instance where the customer is expressing dissatisfaction directly or indirectly regarding a particular purchase?

In reality, it may only be a window of a few minutes to a few hours from this data occurrence before the bad review gets posted. However, today’s super-fast fiber internet speeds and booming Big Data technologies should provide an automated system sufficient time to use this timeframe judiciously to initiate a corrective action, thereby improving customer satisfaction and maintaining reasonably better product reviews.

How the Big Data system works

Below is a very high-level diagram and the steps about how this Big Data system could work:

  1. Data from various sources containing the customer’s digital footprints are fed to a Big Data distributed processing system like Apache Spark or Hadoop in Cloud in a periodic manner.
  2. The Big Data processing system would then process this data and extract the insights needed for the decision makers in quick time. The results are then forwarded to the “Action” system.
  3. The “Action” system would then analyze this feed and take the decision to contact the customer with possible remedial measures, where appropriate.

Actions the seller can take based on the data

The Big Data system would probably generate all the needed results, but then what can the seller do once the software has determined that there has been a bad purchase?

  1. Immediately contact the customer requesting feedback about the purchase and/or provide contact details for resolution, if any.
  2. Point the customer to the replacement department if customer desires to replace the product.
  3. Provide coupons or discounts for future purchases.

Do you think the sophistication of today’s technologies allow for such an implementation? What do you think could be the challenges in implementing such a Big Data system?

Learn more

For more information on artificial intelligence (AI), Big Data, and IBM Cloud, see the following resources:


More from Cloud

Kubernetes version 1.28 now available in IBM Cloud Kubernetes Service

2 min read - We are excited to announce the availability of Kubernetes version 1.28 for your clusters that are running in IBM Cloud Kubernetes Service. This is our 23rd release of Kubernetes. With our Kubernetes service, you can easily upgrade your clusters without the need for deep Kubernetes knowledge. When you deploy new clusters, the default Kubernetes version remains 1.27 (soon to be 1.28); you can also choose to immediately deploy version 1.28. Learn more about deploying clusters here. Kubernetes version 1.28 In…

Temenos brings innovative payments capabilities to IBM Cloud to help banks transform

3 min read - The payments ecosystem is at an inflection point for transformation, and we believe now is the time for change. As banks look to modernize their payments journeys, Temenos Payments Hub has become the first dedicated payments solution to deliver innovative payments capabilities on the IBM Cloud for Financial Services®—an industry-specific platform designed to accelerate financial institutions' digital transformations with security at the forefront. This is the latest initiative in our long history together helping clients transform. With the Temenos Payments…

Foundational models at the edge

7 min read - Foundational models (FMs) are marking the beginning of a new era in machine learning (ML) and artificial intelligence (AI), which is leading to faster development of AI that can be adapted to a wide range of downstream tasks and fine-tuned for an array of applications.  With the increasing importance of processing data where work is being performed, serving AI models at the enterprise edge enables near-real-time predictions, while abiding by data sovereignty and privacy requirements. By combining the IBM watsonx data…

The next wave of payments modernization: Minimizing complexity to elevate customer experience

3 min read - The payments ecosystem is at an inflection point for transformation, especially as we see the rise of disruptive digital entrants who are introducing new payment methods, such as cryptocurrency and central bank digital currencies (CDBC). With more choices for customers, capturing share of wallet is becoming more competitive for traditional banks. This is just one of many examples that show how the payments space has evolved. At the same time, we are increasingly seeing regulators more closely monitor the industry’s…