PDLP Chapter 47 Intranet Strategy
Chapter 47 Intranet Strategy
Alibaba's "Buy One Item Anytime" Feature Internal Guide
Inside Alibaba's headquarters, the latest weekly performance report on the "Buy One Item Conveniently" feature has been released.
Lin Mu didn't even crack the top 5 on the new orders list. However, he was at the top of the list in terms of first-time purchase rate.
His proposed user mindset and audience operation model has proven to be feasible, with first-purchase rates for products like yoga mats and niche aromatherapy reaching 15%.
In other words, out of every 100 yoga mat buyers, 15 will also purchase a niche aromatherapy product.
While this can't compare to the 30% to 40% first-purchase rate of toothbrushes paired with toothpaste, anyone could create such a strong association. What we need to uncover are the underlying needs hidden beneath the surface.
Lu Hai approached Lin Mu and said, "Da Yan, your suggestion is excellent. Why don't you write a strategy guide and post it on the intranet? That way, when the 'Buy One' feature is officially launched, everyone will have a reference. This is also President Lu's suggestion."
Lin Mu naturally had no objections and said, "Okay, I'll write it now."
After editing for a period of time, Lin Mu finished writing the strategy post.
[#Taobao Insider Guide: Advanced Strategies for Using the "Buy One On the Go" Feature, from Linking Items to Linking People]
[Shared by: Dayan]
back ground
[Regarding the platform's manual product recommendation feature during its gray-scale testing, this article shares how to think outside the box, achieve higher-value product pairings, and uncover users' latent needs, based on recent pilot data.]
Hello everyone, we recently launched the "Buy One Item Now" feature.
Due to technical limitations, this feature cannot currently achieve automatic matching; it requires manual recommendation and association by Taobao staff.
I received one of the first batch of test slots, conducted the tests, and, through internal discussions and summaries, have generated some interesting data. I'd like to make that announcement now.
Firstly, I believe this functionality has three levels.
The first layer is associated items; the second layer is associated scenes; the third layer is associated people.
[Let me give you an example. In the beta testing, we discovered a combination with the highest first-purchase rate, reaching 40%—a washboard and soap.]
In other words, any user who actively searches for a washboard has a 40% chance of also buying a bar of soap after purchasing one.
The underlying logic is easy for everyone to understand: complementary functions create related products. However, this level has a core limitation: it relies on common sense, has a low ceiling, and cannot drive greater purchasing demand.
Let's take a second example: a USB fan connected to a screen wipe.
The logic of this second layer is about relating scenarios. Why is it better than the first layer? Because it represents a breakthrough compared to the first layer, extending from "what to use" to "where to use".
Imagine this: by linking these two products, isn't it easy to sketch out a user profile? A white-collar worker who values their work environment?
So, are his real needs just a USB fan and screen wipes?
Can we recommend a good computer chair for him?
[Could you recommend a good height-adjustable computer desk for him?]
Does he get sore shoulders or back pain from spending so much time in front of the computer? Could you recommend a neck and shoulder massager for him? Or even some snacks?
Okay, now that we've given examples of the second level, let's look at the third level.
I believe the third layer of connection logic is the complementarity between groups of people and their mindsets. It connects to individuals, specifically user profiles with specific lifestyles, values, and underlying desires.
[This might sound a bit abstract, so let me break it down with a real-world example—a snack sealing machine paired with imported coffee beans.]
The first two logics are based on our experience and sound judgment, while the third is the key to unlocking a new world. Why did snack sealing machines and imported coffee beans succeed?
Here's some data: their first-time purchase rate reached 10%.
[Some might say, "A 10% first-purchase rate isn't higher than the traditional combination, is it?" No, that's a shallow view. Dude, your perspective is narrow. The key point here isn't the data difference, but rather to think carefully about why the first-purchase rate for snack sealing machines and imported coffee beans isn't 0%.]
[Because they both target the same user group: young urban professionals who pursue a refined and convenient lifestyle and value quality details. Those who buy sealing machines aren't just worried about their snacks getting damp; they also value a clean, organized, and efficient lifestyle. This overlaps significantly with users willing to try higher-priced, uniquely flavored imported coffee beans, highlighting their emphasis on quality of life. We're not connecting two products, but rather two lifestyles of the same person.]
This may sound abstract, but it can be broken down into three executable methods.
The first step is to move from problem-solving to defining identity. Don't think about what else people who buy A need; instead, think about what kind of person people willing to pay for A typically want to be.
To give another example, someone buying a yoga mat isn't just looking for a mat; they're likely embracing a lifestyle focused on health and inner peace. Therefore, associating it with niche aromatherapy oils is far more targeted and persuasive than associating it with a sports towel, satisfying their need for identity recognition.
The second step is to move from usage scenarios to the behavioral journey. Don't just focus on the moment a user uses the product; instead, reconstruct the entire process from when they have an idea to when they complete the experience.
[For example, I have a top-selling combination here: pairing a retro game console with an HDMI adapter. This solution has an 80% first-purchase rate.]
Why is this the case? Users who buy retro game consoles are driven by nostalgia. However, their journey after the nostalgia ends immediately encounters a practical pain point—how can they play without an AV input on their TV? Therefore, using an HDMI adapter isn't just icing on the cake, but a lifesaver, directly removing the core obstacle to their experience. This is the next step in their journey.
Here's a little secret to teach you how to find the behavioral journey from usage scenarios: Ask around, directly ask the merchants, collect data on user inquiries with merchants, and find the behavioral flow from the data.
Finally, the third step is to move from category data to behavioral data. Make good use of backend data, but don't just look at category rankings; dig out hidden clues such as repeated searches across categories and joint purchases across categories.
[Finally, let's take an example: why is there a strong correlation between postgraduate entrance exam English test papers and noise-canceling earplugs? Because the data speaks for itself; a large number of users searched for these two products at the same time. Behind this is a clear profile: users who buy postgraduate entrance exam test papers are in the extreme scenario of high-pressure exam preparation; their core need is extreme focus, and noise-canceling earplugs represent a direct investment in a focused environment. This is how behavioral data is found from category data.]
Conclusion
The future of connectivity, the ultimate form of the "buy one thing at a time" feature, will not be about mass-linking hundreds of thousands or millions of product pairs—that's just behavioral laziness. The real future lies in our ability to understand users' desires before they even say "I need," and to prepare "you might also want" before they even say "I need."
Today, through manual association, we are practicing this ability to understand people. This ability will be the cornerstone of all future personalized recommendations and intelligent marketing. From associating with items to associating with scenarios, and ultimately to associating with every unique, living person, this is the path we should embark on first. Hopefully, this guide will offer some inspiration.