Using Reviews To Improve Product Listings In Ecommerce

To improve ecommerce product performance: don’t ignore customer reviews. Most brands do next-to-nothing with this valuable feedback. Yet, they are a goldmine because: 1. Customer reviews are generally more honest than surveys. 2. Which means the information in these reviews can effectively inform improvements for headlines, testimonials, content, or even sales pitches. At Enavi we utilize this information through our Human-Obsessed approach, based on the following set of questions: Identifying Pain Points 1 - What issues were customers trying to solve with the product? 2 - Is there a common thread that led users to shop for the products? Recognizing Recurring Features: 3 - Which aspects of the product are repeatedly mentioned, positively or negatively? 4 - How does that compare to what we “thought” was important for users? Noticing Benefits: 5 - Are there any benefits in the customer reviews that we didn’t consider previously? Identifying Outcomes: 6 - Which specific outcomes have customers highlighted? Acknowledging Concerns: 7 - Were there any hesitations before the purchase? Use Cases: 8 - What frequent uses or applications of the product are mentioned? 9 - Do the use cases align with what is mentioned in the product description and key messages? 10 - Could these reviews be harnessed for testimonials? By following this 10-step process, we've effectively enhanced product-specific conversion rates and overall performance. Why does it work so well? Because review mining with a Human-Obsessed focus isn’t just about making adjustments. It’s about building better products and growing your business. Where data ends, human insight begins.

I've been in ecommerce for 18 years, and have made hundreds of millions of dollars of incremental profit for clients. One of the key factors I see stopping brands from scaling is putting too much emphasis on a tactic. For example: On Amazon, it is widely understood that your page needs to have robust reviews to perform well. This is true. However, brands will often compress this into reviews = money Which is not true. This is where you lose money.  Brands who get fixated here will “do what it takes” to get a review, fast. If reviews=money, then you aren’t going to wait for them to show up organically and Vine takes a little while. Now you're tempted to incentivize your sister in laws friends to write you reviews. This costs you revenue in two ways: 1. Your account is going to get suspended. The largest data company on earth can see what you’re doing 2. Your conversion will go down and returns will go up Good reviews can print money. They help customers validate their search and answer questions before they buy. This increases conversion and reduces returns. How? The best, highest converting, lowest churn goods have star ratings between 3.7- 4.2 1. No one believes a 5 star review- they always look like your mom wrote them 2. There is nothing useful in “This is great”- and your customers probably don’t all agree on what great is 3. 3 and 4 star reviews usually have justification- “I liked this but not that”- This is the money maker. This helps people decide if they care more about this than that. Example: You sell a moisturizing shampoo It is 3.7 stars All reviews mention that it is not moisturizing. But it's also a best seller. How? The clues are in your 3 star reviews “It isn’t moisturizing at all, but it smells great and works fine on my normal hair. Don’t buy it if your hair is dry, but I love it” This works in three ways -Increases conversion by validating purchase decisions from people with normal hair -Reduces returns from those with dry hair because they are aware it won’t work -Tells you how to update your copy and search to maximize sell through Honorable mention- people who don’t like your product will rarely tell you. They will return it and talk about how much it sucks to their friends, not you. Feedback is a gift that you want desperately. This process is what makes reviews valuable on a page. Good reviews can make you money, but bad ones will cost you and this is not based on star rating. Do not hyper focus on a tactic at the expense of your strategy. Shortcuts are often incredibly expensive. #strategy #retail #topretailexperts #ecommerce #amazon

New Ecommerce SEO article: How to use gen-AI + reviews to write product descriptions at scale. In 10 minutes, you can create 250+ descriptions: This is an article that I just wrote for the Search Engine Land site. Reviews are a goldmine of data for ecommerce sites. However, this is a data source that's often underutilized on my sites. Many commerce sites have reviews that contain a ton of information about real customers. Fans of your company, people who have used the product, users will real-life experience. This is the type of content that's perfect for using as a source for product descriptions as users are likely highlighting the product's most important features. By using Screaming Frog's OpenAI integration, you can have the tool extract your reviews, run them through ChatGPT and have it create product descriptions. Here's the process. 1. First you'll need to have Screaming Frog extract the reviews from your site. Navigate to Configuration > Custom > Custom Extraction > Add. Add a new extraction called "Reviews" 2. Then select the "Globe" icon to pull up the visual editor. This will help you create the regex/CSS to extract your reviews. 3. Run a sample product page through the editor. Click your reviews section and look at the "Suggestions" feature. Toggle through these until you find a code snippet that will pull just your reviews. Save this configuration. 4. Next, you'll need to push this extraction to OpenAI. Navigate to onfiguration > API Access > AI > OpenAI. Ensure your OpenAI key is added in "Account Information". 5. Next go to the Prompt Configuration tab and click Add. Name your prompt “Product Descriptions.” In the Page Text dropdown, select Custom Extraction > Reviews. 6. In the "Enter Prompt" field add your prompt. You'll want to tell OpenAI to create reviews from the submitted product reviews. You can click the "Play" button to test the output. 7. Finally, start your crawl by using Screaming Frog on "Spider" mode. When the crawl is complete, go to AI > Product Description in the right-side bar to find your new product descriptions. What I've found is that the product descriptions outputted are generally pretty strong. They take into account a product's most prominent features from reviews and integration them into the actual description. It's worth noting that this content only serves as a STARTING POINT for product descriptions. You 100% need to have a copywriter review the output and makes adjustments. However for commerce sites with thinner resources, this is a good way to quickly generate product descriptions using real-world customer data.