Customer Experience (CX) as an architect to bridge buyers and sellers
“Focusing on the customer makes a company more resilient.” — Jeff Bezos
A Shift to Customer Obsession
An emerging digital technology has changed the behaviours of customers dynamically. Exposure to media have continuously driven their needs from one to another.
According to the Temkin Group, 86% of those who received excellent Customer Experience were likely to repurchase from that company, compared to only 13% of those who had a very poor Customer Experience (more than 6 times more likely).
They also found out that those who received excellent Customer Experience were 11 times more likely to recommend the company than those who had a very poor CX (77% vs. 7%, respectively).
Amazon is the largest eCommerce player in the world and used worldwide. Jeff Bezos — founder of Amazon had a vision to dominate e-commerce industry and not only to sell books but everything that customers need. Focusing on what customers wants or needs has driven many of Amazon’s most profitable business moves.
“The No. 1 thing that has made us successful by far is the obsessive compulsive focus on the customer as opposed to obsessing over the competitor,” Bezos said in a talk at the Economic Club of Washington on September 13.
Tokopedia Customer Experience (CX) Team
To follow Amazon success, Tokopedia have 3 DNA, “Focus on Consumer”, “Growth Mindset” and “Make it happen & Make it better”. Embedded to it’s employee, to improve not only Tokopedia as a company but also it’s users.
To support one of the DNA “Focus on Consumer”, Tokopedia built a Customer Experience team specifically designed to understand customer expectations. We gather insights from our customers on what they are experiencing in overall (between pre- and post- purchase). One of the sources that we use is Net Promoter Score (NPS) survey.
Net Promoter Score (NPS)
Net Promoter Score (NPS) is a survey used to measure the willingness of your customers to recommend your company, product, or services to others based on their experiences.
The typical question of NPS is “How willing are you to recommend product/brands to your family or friends?” with the scale rating answer 0–10. Then we categorized the scale answers to:
- Promoters (10 & 9)
- Passives (7 & 8)
- Detractors (0–6)
Further questions should be asked to analyze more on why they gave the score. The picture below shows how you calculate the NPS Score.
The benefit of NPS itself is to provide the big picture and also customer relationship with your brand. A low NPS would indicate that there are several things needed to be improved. On the other hand, your business is on the right track if you have a high score.
To understand more about the NPS, let’s see the study case below on how we close the gap between buyers and sellers!
Study Case: Analysis of buyers and sellers expectation gaps within Tokopedia marketplace.
Several months ago, there was a significant increase in passives and detractors in the NPS survey. After analyzing the feedbacks on why customers gave a low score, these are the top feedbacks:
- Buyers demand faster response
- Buyers are frequently rejected because out of stock issue
- Buyers receive mismatch products
To prove and investigate further, we validate to our internal and external (either qualitative /or quantitative research) data. Below are the explanation on how we gain insights and give our recommendation.
Gap of expected response from buyers & sellers
The analysis is done based on buyer’s journey perspective to order products in the marketplace. The product page itself is the last touch-point for buyers to chat sellers. When buyers reach out sellers to ask them on product stock or details, there is no indicator to tell buyers that the sellers can give either fast response or not.
We interviewed users how long sellers needed to response and which lead to there is no certain amount of numbers, but they demand to be replied as soon as possible.
Hence, we shift our attention to our internal data to create the response grouping based on fast, slow, or average using first and third quartile based on their first response time. Discussion feature is where you can discuss a product between buyers and sellers publicly while chat is more to a private channel.
There is also additional proof testing to make sure that response grouping is sufficient by analyzing the average response time of successful transaction if it’s enough for the user to wait that long. From the graphic below, it’s proven that the user is capable of waiting that long in the normal and fast group to checkout.
We also assess this issue using a scatter plot on number of transactions againts sellers response time. There is no correlation between First Response Time (FRT) and number of transactions but there are a tendency for a high number of transactions with a faster response.
To sum-up, our recommendation is to create response group (fast, average, slow) to bridge buyers & sellers response expectation and educate sellers to respond faster which may lead to an increase of successful transactions.
Rejected transaction due to out of stock products
To understand more about this issue, analysis on our internal transaction data to see what sellers use to indicate the stock status (limited stock, always available or unavailable) on rejected transactions. The result is most sellers use always available compare to the limited number of stocks. In addition, sellers that also use limited stock often reject orders when their product stock had less than certain number.
To do further analysis, we interview sellers via phone call that reject orders using limited stock. We found out that sellers often use available stock due to several products refresh quickly. In addition, sellers sometimes forgot to update their stock when they run-out while using limited stock.
To tackle these issues, we propose to the related team to change the default user interface (UI) from always available to limited stock in order to shift the behaviour of users to fill their number of stock amount. Furthermore, we will add an additional feature — stock reminder, which is used to remind sellers with a push notification if their stock has reached a certain amount of number.
Received Mismatch Products
User occassionally received product which are not suitable to it’s description or picture shown on the product page. To gain more insights, we looked-up to our internal data which are tickets complain about these issues and also at what transaction status when it occurred. The complaint issues are about product quantity, details and broken issues while these kind of issues take place on finish transaction status.
We evaluate our internal procedure in handling such seller issues, where we give point deductions that will impact the seller’s reputation. The analysis result shows that the point deduction was not impactful, due to the small point amount deducted.
So, in order to cover this kind of issue, we gave recommendation to our operations team to add a bigger point deduction if the sellers make accumulated mistakes. In the same time, education to seller is also done to increase their awareness of product mismatch issues and enhance their product details (pictures and description) to create a better satisfaction to our customers.
If you guys have read until this part, we have similar approach for these issues.
- From the NPS survey feedbacks, we analyze on why they gave the score.
- After that, we gain more understanding by checking up on our internal data and/or external data (research).
- Finally, after proofing the specific pain-points, we create recommendation based on it to close the gap.
There is still much moment of truth to be discovered between Tokopedia users and the product or feature itself. It is our job to explore more pain-points, close the gap between it and exceed user expectation.
We will share more stories about uncovering more stories on how CX drive the company closer to their customers in the future so stay tuned!
