---
title: "You Know Big Data, but Do You Really Know Big Data Marketing?"
description: "This article discusses the practical application of big data in marketing, highlighting common pitfalls such as data inaccuracy and inductive reasoning fallacies. It proposes an alternative approach: tagging marketing scenarios based on instinct, emotion, and cognition, and tailoring marketing messages accordingly."
author: "呵先生"
publisher: "New Distribution"
email: "zhaobo258@gmail.com"
telephone: "+8615854817671"
published: "2018-10-09"
language: "en"
canonical: "https://xinjignxiao.com/en/articles/you-know-big-data-but-do-you-really-know-big-data-marketing-eddce2cb/"
markdown: "https://xinjignxiao.com/en/articles/you-know-big-data-but-do-you-really-know-big-data-marketing-eddce2cb.md"
original_source: "https://mp.weixin.qq.com/s/mUjsJ5xZbjkO8AK2f4qRxw"
translation: "https://xinjignxiao.com/zh/articles/%E4%BD%A0%E5%BE%88%E6%87%82%E5%A4%A7%E6%95%B0%E6%8D%AE-%E4%BD%86%E6%98%AF%E7%9C%9F%E7%9A%84%E6%87%82%E5%A4%A7%E6%95%B0%E6%8D%AE%E8%90%A5%E9%94%80%E5%90%97-eddce2cb.md"
attribution: "New Distribution — https://xinjignxiao.com/en/articles/you-know-big-data-but-do-you-really-know-big-data-marketing-eddce2cb/"
usage_policy: "https://xinjignxiao.com/ai-policy.txt"
---

# You Know Big Data, but Do You Really Know Big Data Marketing?

> This article discusses the practical application of big data in marketing, highlighting common pitfalls such as data inaccuracy and inductive reasoning fallacies. It proposes an alternative approach: tagging marketing scenarios based on instinct, emotion, and cognition, and tailoring marketing messages accordingly.

Click 'Read Original' for details.
Source: He Yuanwai (ID: Yuanwai-HE)
Author: Mr. He
In recent years, many people in both work and life like to discuss big data with me. They feel that looking at big data is like having collective presbyopia: it's clear from afar, but the closer you get, the blurrier it becomes, and you can't find the right approach.
We all know what big data means—it's **massive, high-speed, multi-dimensional, low-value-density real data**. It's a mouthful; simply put, we can know detailed real data about an individual, but it's hard to think of specific help for marketing execution with this data (that's the difficulty).
In reality, big data often provides managers with a sense of **control**, which is similar to superstition in ancient times. Seeing various data about consumers makes you feel you understand them well and have everything under control.
But when it comes to actually using this data, it often reverts to simply pushing what consumers have browsed or purchased in the past (which is also the logic behind Taobao and Baidu's recommendations).
Last year, I led a project that connected a retail company with the data of a well-known data bank. After accumulating a certain amount of data, we had consumers' age, gender, occupation, income distribution, family size, car and house ownership, mobile app interests, reading interests, and much more.
Then everyone got stuck in the mud: **how can we use this data well?**
Going back to basics, I understand that big data has two main uses: **one is to look at trends, market, and consumer behavior development direction (long-term). The other is to drive sales, by analyzing consumer data that can link with your business to increase marketing efficiency (short-term).**
Many large data institutions, such as Alibaba Cloud, CBNData, or consulting firms like Deloitte and KPMG, regularly provide consumer reports for different markets. These reports mainly analyze long-term trends.
This article focuses on the latter: **how to use your daily consumer data to help improve company performance**.
Currently, big data applications in many large companies mainly use two methods. Besides the simple and crude push of repetitive information, a more advanced approach is **tagging + precision marketing**.
Simply put, it's two steps:
First, tag consumers by age, gender, interests, habits, and life stage;
Second, determine which consumers the marketing message targets and reach them directly;
But this big data application method also has **two drawbacks**:
**1. Data Source**
Although big data captures target consumers' behavior records indiscriminately, when we use this data in a **local context**, there is still a high possibility of **interference from inaccurate data sources**.
For example, I once did a consulting project using big data to capture foot traffic and store visitor data around a coffee shop, to analyze why sales at a location with decent foot traffic were not improving.
After two months of data collection, we found that the store's consumer age distribution was mostly middle-aged or elderly, and they stayed the longest.
We then fell into a misconception, thinking the store should add more elements to attract middle-aged consumers, such as more prominent discount information, or even enlarge the menu font. However, these measures had little effect.
Until one day we decided to visit the store ourselves to see what was happening that prevented these measures from increasing conversion rates even slightly.
It was a hot summer day; we walked to the store entrance and noticed a strange phenomenon: many middle-aged and elderly people were sitting at the door looking at their phones.
It turned out there was a bus stop in front of the store. Because the store had air conditioning and Wi-Fi, many elderly people waiting for the bus would sit at the door to enjoy the AC. That was the truth behind the so-called **main customer group being middle-aged and elderly**.
Big data can mislead in other ways too. I also did an analysis project for a beauty e-commerce store. Data showed that during a promotion, over 50% of transactions came from male consumer accounts.
So, did the promotion seem more effective for men's products? No. When we analyzed the shopping baskets, we found that male accounts bought women's products. In fact, women used their boyfriends'/husbands' accounts to pay.
So if you blindly trust the data presented on a dashboard, your judgment may be misled, because **the consumers seen in the data may not be who they really are in reality**.
**2. Inductive Reasoning Fallacy**
Currently, many large companies' use of big data is at the stage of inductive reasoning. That is, data shows that most consumers have characteristics A, B, C, and then they infer that the consumer tag is D, and then reach out with information.
For example, a hotel finds that its guests mostly have these characteristics:
> Need parking spaces
> Mostly stay one to two nights
> Little or no minibar consumption
Based on these three conditions, it's easy to conclude that the hotel's main customers are families on short trips. So the hotel can increase user stickiness by adding family meal packages and family tickets to nearby attractions. They would then promote through family travel forums and public accounts.
This seems like normal reasoning, but the tag could be completely wrong.
Those who meet the above three conditions are not only family tourists.
They could also be business people attending meetings at nearby companies, who bring their own drinks for entertaining or even go out for nightlife.
They could also be couples driving over to check in at a nearby trendy spot, which is a short-term purpose.
To avoid this inductive reasoning error, the best way is to find "**key circumstantial evidence**" across different data dimensions. Just like Sherlock Holmes sees a pipe burnt on the right side and infers the user is left-handed.
For example, in the above case, the three conditions cannot filter out an accurate tag. But if you add an extra data dimension, "these guests' rooms require extra beds," then you can basically confirm they are a family traveling together.
So when using big data to tag target customers, **finding the key tag can effectively increase subsequent conversion rates**, because if the tag is wrong, precision marketing is impossible.
However, this "tag + precision marketing" approach still has a major flaw.
**From consumer behavior perspective, even if the tag is correct, pushing marketing messages at the wrong time and place is useless.** Just like a new mother is precisely tagged, it doesn't mean she needs to buy milk powder and diapers anytime, anywhere. Not to mention that a middle-to-high-income fashionista doesn't necessarily have to buy a certain trendy brand.
Fundamentally, consumers' purchase motivations are diverse.
Another drawback of tagging consumers and marketing is that it **reduces the possibility of non-target groups buying your products**. For example, I am a college student who loves studying and preparing for the postgraduate entrance exam; that doesn't mean I'm not interested in trendy brands like Supreme. Or a high-income senior manager doesn't necessarily dislike simple and plain Toyota cars.
But if precision marketing doesn't reach these groups, then it's foreseeable that potential sales opportunities are missed.
So why not **try a different way** to understand big data and precision marketing?
Sometimes, **tagging marketing scenarios and developing several sets of copy based on the scenario is more effective.**
Below, we'll discuss what tags marketing scenarios can have from the three dimensions of **instinct, emotion, and cognition**.
The reason for choosing these three dimensions is that in motivational psychology, a person's intrinsic motivation for behavior is mainly influenced by these three aspects, where instinct and emotion are hereditary, and cognition is learned.
When these tags are independently or simultaneously marked on a marketing scenario, they tell us what information the product copy should highlight.
**I. Instinct**
Those who have read Maslow's hierarchy of needs know that the bottom two levels are human instincts. The bottom level is related to survival and reproduction, such as breathing, water, food, sex, etc. The second level is related to safety, such as health, assets, morality, etc.
Safety is so important because **humans naturally seek certainty from the surrounding environment and spend our entire lives increasing this certainty**.
In hunter-gatherer societies, having your own cave meant being free from large predators, and rituals were used to try to ensure favorable natural conditions.
The continuous pursuit of certainty has driven human progress from ten thousand years ago to today.
Even now, we still try to increase the **certainty** of the world:
> Countries increase international certainty through organizations like the UN
> Individuals increase income and promotion certainty by joining Fortune 500 companies
> Mothers-in-law increase the certainty of their daughters' quality of life by requiring a wedding house
These certainties all give stakeholders a sense of security.
Researchers have found that **when we are in a secure environment, we tend to choose more personalized products. When we are in an insecure environment, we tend to choose mainstream products.**
For example, if an app's marketing channel is before a movie on a video site, for many merchants, all movie ads are the same channel, attracting people who like watching movies. But it's more effective to tag TV series and movies.
If the viewer is watching a thriller or suspense movie, the app's marketing message should highlight that **many people are already using it**, such as "a second-hand car platform chosen by 30 million people."
Conversely, if it's a romance or sci-fi theme, the app's copy should highlight its **uniqueness**, such as "no middlemen taking a cut, sold within three days."
These two tags apply not only to streaming content ads. For example, ads in office building elevators should convey mainstream information, while ads in residential building elevators should be personalized, because people's sense of security **at home is usually higher than at the office**.
**II. Emotion**
Different marketing scenarios give consumers complex emotions. To simplify, we'll divide tags into two categories: familiarity and unfamiliarity.
In psychology, there's a concept called the **priming effect**, where our eyes see anything and the brain starts associating concepts related to that perceived target.
For example, when I see a grassy field, subconsciously I activate cognition and memory to associate concepts related to "grass."
At this time, because everyone is watching the World Cup, concepts related to football and the World Cup are more easily "primed." Then things like national football teams, even football players, give us a sense of familiarity.
When consumers face **familiar** concepts, they tend to think about cost issues, that is, **what will hinder me** from taking a certain action.
When facing **unfamiliar** concepts, they tend to think about benefits, that is, what **benefits** this thing can bring.
Therefore, in this dimension, before tagging a marketing scenario, we need to think about **how direct the connection is between the product and the marketing scenario content for the target group**.
For example, if we sell clothes and place an ad on a public account with emotional content, then this marketing channel has high familiarity for target consumers. At this time, consumers will first think, "Oh, there are clothes for sale, let's see the price first."
Similarly, on an emotional content public account, if the product to be marketed is tea sets, then it's low familiarity. Consumers will immediately think, "Why sell tea sets here? What's so good about these tea sets?"
**III. Cognition**
In a previous article "When We Talk About Traffic, What Are We Really Talking About?", I mentioned the concept of **cognitive closure mode**.
Simply put, in different scenarios, we have different acceptance levels for ambiguity in answers to questions.
For example, when a shopaholic is buying clothes on Taobao, they have in mind "I want to buy a dress to wear when shopping with my boyfriend next weekend." At this time, when seeing a style, many questions arise:
> "Is this style the standard for this season?"
> "Does this fabric look see-through/hot/easy to wrinkle?"
> "Does this store have discounts recently? Will I lose money?"
> "This style is similar to the one I saw before; what's the difference in wearing effect?"
> "..."
At this time, there is **high cognitive closure need**, because to perfectly complete the task of buying clothes, these questions **must be accurately answered to make a decision**.
When consumers browse information and think they are completing a task, there is a high cognitive closure need.
When are consumers completing a task?
The answer is transactional scenarios, such as e-commerce platforms like Taobao and JD, or offline retail stores like supermarkets and personal care stores, even flight and hotel booking websites.
Conversely, in content scenarios, such as video sites, short video apps, public accounts, and Moments, consumers enter **low cognitive closure need, meaning they are prone to impulsive decisions due to one or two product advantages.**
(PS. This is also the fundamental reason why WeChat business conversion rates are so high)
So when we tag a marketing scenario as transactional, the marketing message should be a detailed comparison of data and address consumers' main doubts. This scenario is more suitable for marketing supplementary products, like daily care items.
For content scenarios, marketing messages should be concise and use emotional connections to stimulate. This type of scenario is more suitable for new and novel products, like newly imported foreign products or an unknown romantic hotel.
**Summary**
Today I shared a new way of thinking about using big data.
Currently, many large companies use:
  1. **Repeat push based on browsing/purchase history**
  2. **"Tag + precision marketing"**
These methods are affected by data source and inductive reasoning fallacies, and cannot effectively improve conversion rates.
Therefore, here's another option: first use big data algorithms to tag marketing scenarios, then push personalized information in different marketing scenarios.
The advantage is that the marketing scenario, product information, and the information to be communicated are all objective and certain.
Merchants only need to customize several sets of marketing copy based on the tag combination of the marketing scenario, and then use big data to distribute to different channels.
Simply put, it's **"use the right bait in different ponds, so the fish in that pond are most likely to bite under the current water temperature and light conditions."**
**On October 23-24, during the Autumn Sugar and Wine Fair, New Distribution will host the '2018 FMCG City Distribution Logistics Conference'.** At that time, we will invite industry experts, FMCG warehousing and distribution specialists, and distributors who have transformed to unified warehousing and distribution platforms, to discuss and answer questions about the future development trends of FMCG city distribution logistics and practical cases of distributor transformation to unified warehousing and distribution, under the theme "New Distribution, New City Distribution". We hope to bring you different inspiration and thinking! The specific meeting topics are as follows:
**List of Participating Companies**
In no particular order
Hunan Zonglan Diandan Network Technology Co., Ltd.
Jingbang (Wuhan) International Freight Forwarding Co., Ltd.
Mengniu Dairy
Qinghai Hanxiang E-commerce Co., Ltd.
Unilever Service (Hefei) Co., Ltd. Shanghai Branch
Huicong
Hunan Xuan'ang Food Co., Ltd.
Guangzhou Tongdaoren Information Technology Co., Ltd.
Qingdao 888 Trading Co., Ltd.
Uni-President Enterprises (China) Investment Co., Ltd.
Hunan Province Zhongxiang Gongpei Logistics Co., Ltd.
Shenglong Ingredients
COSCO Shipping Logistics Warehousing and Distribution Co., Ltd.
Guangxi Yongpai Liquor Co., Ltd.
Shangqiu Kangrong Trading Co., Ltd.
Jinan Dingzhong Economic and Trade Co., Ltd.
Liaoning Bimai Agricultural Technology Co., Ltd.
Kunming Xiongjia Trading Co., Ltd.
Shaanxi Houheng Trading Co., Ltd.
Guangzhou Dingwo Enterprise Information Consulting Co., Ltd.
Shaodong Jiajiale Commercial Firm
Boda Trading
Industrial Bank Changsha Branch
Wuhan Muchen Convenience Store Chain Co., Ltd.
Fujian Fuxing Yuncang Logistics Co., Ltd.
Guizhou Yilimi E-commerce Co., Ltd.
Jiangxi Xiao Laoer E-commerce Co., Ltd.
Jinshankoufu
Shanxi Taihang Yuanjing Supply Chain Management Co., Ltd.
Shanxi Dezhun Supply Chain Management Co., Ltd.
Shaoyang Tongdeli Trading (Xiangbang Logistics)
Huanfu
Tongda Express City Distribution
Beijing Xinjingxiang Food Co., Ltd.
Wuhan Huizhong Tianhong Liquor Co., Ltd.
Changsha Paide Biotechnology Co., Ltd.
Chao'an Tuqiang
Guizhou Yihe Bopin Supply Chain Management Co., Ltd.
Jiangxi Kang'en Industrial Development Co., Ltd.
Xiangtan County Yisuhe Town Yuhua Paper Store
Luoyang Yuanlang Trading Co., Ltd.
Tongchuan Yaozhou District Huayuan Supermarket Co., Ltd.
Hunan Yongfu Jiujiu Trading Co., Ltd.
Zhejiang Chengchengtong Logistics Co., Ltd.
Chongqing Kaiguo Materials Trading Co., Ltd.
Beijing Xianmaixianmai Data Technology Co., Ltd.
Hanchuan Qixing Trading Co., Ltd.
Tongxin Jiuzhiru Trading Co., Ltd.
Guizhou Meiguo Guoguo Network Technology Co., Ltd.
......
**Distributor Transformation Representatives (Tentative)**
In no particular order
Jiangsu Huashang City Distribution Network Co., Ltd. Chairman, Rong Jun
Hubei Yijiaren Logistics Co., Ltd. Chairman, Wang Bo
Sichuan Chengdu Xingrenxing Trading Co., Ltd. General Manager, Jiang Shuming
Shandong Yunbang Warehousing and Logistics Co., Ltd. Chairman, Liu Jichen
Chongqing Lingyu Consumer Goods Supply Chain Management Co., Ltd. Chairman, Tu Mingyu
Guangzhou Zhongshan Wanrong Marketing Co., Ltd. Chairman, Yang Su
Sichuan Bajie Supply Chain Management Co., Ltd. Chairman, Yuan Xia
Hubei Pengdun Meiyitian Supply Chain Management Co., Ltd. Co-founder, Li Qiangyun
Henan Xuchang Jiulegou E-commerce Co., Ltd. Chairman, Zhang Jianyong
Hebei Changyi Logistics Co., Ltd. Founder, Ma Haichao
Hebei (Chengde) Wulian Yuncang Co., Ltd. General Manager, Meng Yucun
Xinjiang Urumqi Su'an Jinchi Logistics Co., Ltd. Chairman, Zhang Xun
Jilin Sanxing Lianguo Chairman, Zhang Hailing
Hebei Dunjie Supply Chain Management Co., Ltd. Founder, Qiang Huitao
Hunan Damei Supply Chain Management Co., Ltd. General Manager, Liao Lei
......
-END-


---

## Copyright and AI use

This article is sourced from New Distribution. Search, quotation, summarization, and model training are permitted, but every use must credit New Distribution and retain the canonical source URL.

Contact: zhaobo258@gmail.com · +86 158 5481 7671