You knew before buying this lawn mower that it was a safe bet. Not only did your neighbor advise you to buy the same one she had, but a colleague also recommended the brand. More importantly, though, the dozens of comments and reviews you read online were mostly positive and described a similar usage scenario. This is the way we buy more and more products now — we look up online reviews to make up our minds.
Actually, these comments, ratings, and reviews are crucial to the decision processes of 70% of online customers, according to the OECD. Moreover, the fact that reviews are available on the website increases the chances of visitors who will consider buying from it. Whether the purchase is made online or not, customers read reviews.
Businesses have understood this expectation, as more and more online stores offer customers the possibility to post a review and/or to rate their purchase. Actually, more than 70% of companies collect the feedback of customers from more than one channel, as revealed in this report from BCG. Not only are customer reviews expected by customers themselves, but companies understand the potential of using this source of information.
Read Your Customers' Minds? Nearly!
This lawn mower might have been a good purchase, since it worked so well for the last two mowing duties, but you didn’t expect that the grass catcher was sold separately. This little frustration might make you answer the post-sales feedback email sent by the online store and mention it in the customer review form. This will be useful information for the seller, maybe should they specify it clearly on the product page?
By openly sharing their opinion on a service or product, customers give business owners the opportunity to hear what is good or bad about their offering. It could be about the product itself, the experience of the purchase, or even the customer service — reviews cover any topic.
The challenge that quickly appears when integrating a review system on an e-commerce website is volume. Reviews are multiplied by the number of products and can be counted in dozens or hundreds over time. Understanding customer sentiment is hardly doable manually: With the help of NLP-powered business solutions, businesses can overcome this challenge and start to better match with their consumers’ expectations.
To that extent, Dataiku has recently launched a Business Solution aimed at analyzing customer reviews. It allows organizations to quickly deploy an analysis of past reviews to understand the overall opinion of customers. It also includes the ability to make that review more specific and zoom in on certain types of reviews. Finally, brands can also explore for specific keywords thanks to a dedicated NLP-powered search engine: You want to know which products were the main ones considered as a “bad fit” in order to better size your next collection? The search engine will automatically match relevant reviews, not only the ones that mention “bad fit,” but also those that are relevant from a topic perspective, such as “too small size.”
Note that the current solution leverages a Google TensorFlow encoder model: Bear in mind that this is used as a template and that other models may be used for better results depending on your data/usage.
How Can Dataiku's Business Solutions Help You Reach Full Potential?
Business Solutions are Dataiku add-ons accelerating the way to achieve advanced or foundational industry-specific use cases within your organization. They are an operational shortcut to achieve real-world business value. Taking advantage of Dataiku’s core features, they are built to be fully customizable and entirely editable.
They come with:
- A user-friendly interface that enables fine tuning to match specific business requirements
- Ready-to-use dashboards that can be customized
- Documentation and training materials
Dataiku industry specialists develop solutions for every vertical, among which:
- Retail & CPG: Distribution spatial footprint, market basket analysis, RFM-enhanced customer lifetime value, product recommendation
- Financial services and insurance: Interactive document intelligence for ESG, modern insurance pricing, AML alerts triage
- Health and pharmaceuticals: Omnichannel marketing, accelerating drug repurposing
- Manufacturing and energy: Predictive maintenance, outliers detection
As a result, business professionals experience a boost in AI productivity and can rationalize their resources.
How Does It Work in Practice?
The Customer Satisfaction Review solution provides a reusable project to accelerate the development of analytics tailored to your data and business structure.
Learn more about Dataiku's Customer Satisfaction Reviews solution in the video above.
With this solution, marketing managers, digital media analysts, or product managers are able to:
- Identify categories/products which are the most reviewed, along with their average ratings.
- Compare a product rating to the aggregate rating of a pool of products.
- Identify the top two consecutive words in reviews according to specific filters (country, product category, product id).
- Explore reviews that mention the selected keyword.
From a user perspective, the solution is made of the following easy-to-use components:
1. Review Analysis:
Analyze the number of reviews per product category/id and their associated ratings.
2. Review Explorer | Analytics:
Analyze the top two consecutive words within a set of reviews associated with a product category/id.
3. Review Explorer | Semantic Search:
Explore reviews based on your chosen keyword.
Start leveraging the information customers share with you right now, with these simple requirements:
- Datasets of customer reviews
- Dataiku version: 10.0 or later