Is the Double Eleven really cheap or just a gimmick?

In order to explore the "double 1 1" marketing routine of Tmall merchants, we took beauty cosmetics as an example and tracked the total number of Tmall flagship stores of 22 brands from 201615 to114.

The 22 brands are as follows (in no particular order):

Domestic 4: herborist, Natural Hall, Appropriate Herbal Medicine, and Meijiajing.

Six in Europe: Lancome, Nivea, Guerlain, L 'Oreal, Vichy, Ave.

Five in the United States: Clinique, Olay, Maybelline, Mysi Buddha and Estee Lauder.

Four in Japan: SKII, Shiseido, Aubrey and Shu Uemura.

South Korea 3: Yin Feng, Shi Yue, Snow Scene, Lan Zhi

We have come to some small conclusions.

1.80% of the goods participated in the "Double 1 1" activity.

During the period of 1 1 for a total of 10 days, a total of 3,502 products were put on the shelves in these 22 Tmall flagship stores. However, not all these 3,000 items participated in the double 1 1 activity.

According to the loading and unloading time, we can divide the goods into seven categories and count the proportion of such goods in all goods:

In fact, there were only four kinds of ABCD products sold by Shuang 1 1 on the same day, accounting for 66.7% of the total of more than 3,000 kinds of products, that is, two thirds.

Where did the other third go?

We can divide the goods that are not in "sales" status on the day of double 1 1 into five categories:

Temporarily remove -F merchandise from the shelves.

Double Pre-sale 1 1- Some E-class goods

Completely off the shelf-some E-class goods

After being taken off the shelf, it was put back on the shelf in another place-some are E goods, and some are CDG goods.

After double 1 1, the products will be new-some CDG products.

Among them, the pre-sold goods and the goods that were put back on the shelves on the same day of Double 1 1 actually participated in the activities of Double 1 1, and we need to identify them.

From a technical point of view, we need to identify products that are pre-sold, re-put on the shelves and completely removed from the shelves, and find out the same products that were pre-sold and on sale that day. In other words, it is necessary to match E-class goods with ABCDG-class goods.

With the change of product id and web links, product names often change slightly, so we need to calculate the similarity of product names under the same brand for all kinds of products. The method we adopt is based on the disordered Levinstein distance, so the specific content will not be expanded.

Yes, 37% of the goods that did not participate in the day's activities were pre-sold, 1% were put back on the shelves with vests, and 62% of the goods were temporarily or completely removed from the shelves.

At this point, we can work out a set of numbers.

Goods on sale that day = goods already on sale+goods that were put on the shelves before Shuang 1 1 and snapped up on Shuang1+goods that were put on the shelves on the same day and snapped up on Shuang 1 1 and then put on the shelves again and continued to be sold on the same day.

Commodities participating in the activity = special commodities of the day+pre-sale commodities-special commodities of the day.

Total number of commodities = commodities participating in activities+commodities temporarily removed from shelves+commodities completely removed from shelves.

Through the above calculation, we can get that among the 22 Tmall beauty brands, products participating in the double 1 1 activity account for 80% of the total products.

How many products of each brand participated in the double 1 1 activity? Please see the picture below:

There are four items with more than 200 pieces: Shi Yue Yin Feng, herborist, Estee Lauder and L 'Oreal; Miss Buddha and SKII came last because the number of goods barely exceeded 50.

2. Real discounted goods account for 35%, and fake discounts are less than 1%.

In order to evaluate the discount more accurately, we only selected more than 2,300 kinds of beauty products that can clearly trace the price and participated in the double 1 1 activity.

Discounted goods can also be divided into three categories:

True discount: commodity prices fluctuate within 10 days, and the double price of 1 1 is the lowest price within 10 days, and there is no price increase.

Fake discount: there is a phenomenon that the price of goods increases first and then discounts.

No discount: the price of goods has not changed.

It should be noted that due to the limitation of data, the determination of "discount" in this study only considers the price, but fails to take into account other forms of discounts such as gifts, coupons, full reduction and postage, so the statistical results will deviate from the actual situation.

The results show that the number of goods with real discounts accounts for about 35.2%, and the number of goods without discounts accounts for about 64.5%. As for the remaining 0.3%, you know.

Let's take a look at the proportion of products that are really discounted by each beauty brand to the total number of products participating in the event:

It can be seen that the only brands with "all-store discount" are Appropriate Herbal Medicine and Honey Buddha. Some products of 13 brand participated in the discount, but 7 brands did not. Generally speaking, the proportion of domestic beauty brands participating in discount activities is very high, which is more than half.

3. Commodity discount rate, 15% off, up to 50% off.

How strong is the discount of real discounted goods?

As can be seen from the above picture, the number of goods with 15% discount and 50% discount is the largest. Among them, 15% discount accounts for 25%, and 50% discount accounts for 15%.

We compare the discount rate by brand, and the result is shown in the following figure (the lower the discount rate, the greater the discount. If there are discounted goods, calculate the average discount rate of discounted goods; If there are no discounted goods, the discount rate is 100%):

It can be seen that the discount rate of most brands is in the range of 60%-90%. Among them, there are 4 parties with 50% discount (Appropriate Materia Medica, oprah winfrey, herborist and Lan Zhi), 2 parties with 60% discount (Shi Yue Yin Feng and Nature Hall), 4 parties with 70% discount, 5 parties with 20% discount and 7 parties without discount.

4. Ten marketing routines of Tmall merchants

We compare the "double 1 1" marketing routines of Tmall merchants from the perspectives of goods shelves and discounts.

We divide the shelves of goods into two major strategies: "restricted purchase" and "pre-sale". Let's take a look at the situation of each brand first:

In the above figure, the horizontal axis represents the limited goods snapped up by flagship stores of various brands on the day of Double 1 1, and the vertical axis represents the number of goods pre-sold before Double 1 1, and the circle size represents the total number of goods that stores participated in the Double1/activity.

If two auxiliary lines are drawn by snapping up limited goods (35) and pre-selling goods (30), the marketing routines of 22 brands can be divided into four types.

Low-key: most brands belong to this type, which is characterized by less restricted goods and pre-sale goods.

Snap up: Lancome, Lan Zhi. On the day of Shuang 1 1, more than 40 products of both brands participated in the limited purchase.

Pre-sale: Clinique and Guerlain. Both brands have more than 30 kinds of pre-sale goods.

Pre-sale+restriction: Estee Lauder, herborist. Estee Lauder is the brand with the deepest marketing routine. 1 * * * 74 models participated in the purchase restriction, 40 models participated in the pre-sale, and the total number of products that the brand participated in the activities was only in the early 200 s.

By the same token, brands can be divided into five categories with the proportion of discounted goods (50%) and the discount rate (80%) as two auxiliary lines:

No discount: including Estee Lauder, SKII, Clinique, Shu Uemura, Avenir, Shiseido and Snow Show. (Note: Please refer to the previous explanation about why there is no discount, that is, brands that do not discount may have gifts, full discounts, etc. )

A small number of big discounts: including Lancome, Guerlain, Vichy, Yue Shi Yin Feng, Lan Zhi, Oprah and other six brands. This kind of brand discounts less goods, but the discount is bigger.

A large number of small discounts: including Olay, Meijiajing, Nivea, Maybelline, Mysore Tathagata and other five brands. More than half of these goods have participated in discount activities, but the discount is not strong.

A large number of concessions: including three domestic brands, namely Appropriate Herbal Medicine, herborist and Nature Hall. These brands not only have more than 90% products participating in discount activities, but also have great discounts.

Balance: L 'Oreal. L 'Oré al is an old brand in the beauty industry. Over the years, I have accumulated rich experience through numerous marketing activities, and I performed just right in this activity of Tmall Double 1 1. Discount rate and discount rate are in the middle level.

Finally, let's take a look at how the brand plays the combination of "up and down+discount". The marketing routines of 22 brands can be summarized into ten categories:

Category I: No discount. Including SKII, Shu Uemura, Shiseido, Yang Ya and Snow Show.

The second category: many small discounts. Including Nivea, Olay, Meijiajing, Maybelline and Mysi Buddha.

The third category: a small number of big discounts. Including Yueshi Fengyun, Oprah Winfrey and Vichy.

The fourth category: big discounts. Including two brands: Appropriate Herbal Medicine and Nature Hall.

The fifth category: purchase restriction+a small amount of big discounts. Including Lan Zhi and Lancome.

The sixth category: pre-sale+purchase restriction+no discount. Estee Lauder.

The seventh category: pre-sale+purchase restriction+big discount. Herborist.

The eighth category: pre-sale+a small amount of big discounts. Guerlain

Category 9: Pre-sale+no discount. Clinique.

The tenth category: equilibrium. L 'Oreal.

What are the results of the top ten marketing routines?

In terms of sales volume, no matter which brand or routine, the sales volume of Double 1 1 on the same day has increased by leaps and bounds.

However, the sales volume of products is related to the double 1 1 environment and the fan base of each brand. High sales volume does not mean good marketing effect.

We need to define the marketing effect:

Marketing effect = double 1 1 daily sales of a brand/double1/daily average sales before.

Sure enough, the eyes of the masses are discerning, and a large number of concessions are king! Which strategy routine are you more likely to be imitated by businesses? Readers have different opinions, please.

The data and technical support of this paper is BDP personal edition.

The writer of this article is the secretary of the Communist Youth League, originally published by the independent third-party media "City Data Group".

Thanks to the machine learning Daniel of the data group? Deo? Special contribution to this article.

This paper uses the data generated by Tmall flagship stores of 22 beauty brands on 2016.16.16.1.14, and the conclusion cannot represent other brands and other kinds of goods.

This paper uses a variety of algorithms to calculate the situation of goods on shelves, discounts and sales. Due to the imperfection of the algorithm, the result may be slightly different from the actual situation.