Machine learning is one of the artificial intelligence (AI) methods. Did you know that the 'marketing' job is the best way to start working on this machine learning? In this article, Takafumi Nakano, a data scientist, teaches using machine learning in BtoB marketing and BtoC marketing with examples. What can you do with machine learning, from lead acquisition and development to upsell and churn forecasts, to ad delivery and customer experience (CX) improvement?

DataRobot, Inc. Data Scientist Nakano Takafumi

If you use machine learning for simple marketing tasks, you can spend time on more strategic tasks

(Photo / Getty Images)

<Table of contents>

مشاوره مدیریت و آموزش مدیریتشرح خدمات تیم مشاوران مدیریت ایران

• مشاوره بازاریابی و آموزش مدیریت بازاریابیمدیریت بازاریابی، بازاریابی کالا و خدمات
• مشاوره طرح توجیهی Business Planطرح توجیهی ، بیزینس پلن Business Plan
• مشاوره طرح بازاریابی تجاری Marketing Planبرنامه بازاریابی، سند بازاریابی، مارکتینگ پلن
• مشاوره تبلیغات – طراحی کمپین تبلیغاتیتبلیغات محیطی، تبلیغات آنلاین تحت وب
• دیکشنری مدیریت بازاریابیمعنی لغات و کلمات مدیریت بازاریابی فروش تبلیغات
• مشاوره برندینگ – آموزش مدیریت برندطراحی استراتژی هویت برند CI Guide
• مشاوره برندسازی – مشاوره برندینگانتخاب نام برند، طراحی لوگو/علامت تجاری

The reason why marketing and machine learning are outstandingly compatible
Use of machine learning in B to B marketing
(1) Lead generation x machine learning
(2) Reed Nursery x Machine Learning
(3) Lead qualification x machine learning
Upsell / churn prediction × machine learning
Use of machine learning in BtoC marketing

The reason why marketing and machine learning are outstandingly compatible
Related article 

Wasn't it like this? Marketing automation reconsidered, pitfalls after implementation
Paradox in the "onigiri" innovation that happened for the first time in 40 years From "Koto" to "thing"?
Why is "Pu Girls" born? "The essence of fan marketing" seen in pro wrestling popularity
Pringle's new digima strategy, why 'virtual live'
Kirin was also verified with "Tanrei" and "Honjo", and finally the "shock effect measurement of TV commercials" was realized.
Japan National Football Association president Kozo Tajima told "Japan representative, road of hardship" just before the W cup
"All items 1% discount" or "one Tada per 100", which customers will buy?
▼ Show All

　Marketing is an area where it is easy to introduce machine learning because there is little negotiation or regulation with other departments, and it is relatively easy to measure the effect, and since the impact is easy to understand, it is a suitable field to start. 

　Especially in digital marketing such as EC and CRM, there are many cases where considerable data has been accumulated from the beginning. Machine learning is used regardless of the type of industry, and it is also a theme with many cases. 

　Machine learning makes marketing work more efficient and improves performance. In addition to simply improving accuracy, it also prevents the personification of tasks that previously relied on the senses of certain people. 

　For example, even if a completely new person is in charge of personnel changes, formatted data can be used to operate the same high-precision targeting model as before. Now, let's look at specific practical use cases. 

Use of machine learning in B2B marketing
　B2B marketing is a mission to capture promising leads with high order accuracy and deliver them to the sales department. In addition to capturing leads, we will scrutinize those leads, increase the interest and interest of potential leads, and hand over to the sales department in a timely manner. 

　The lower the volume of leads delivered, the lower the number of orders, and the lower the accuracy, the higher the sales department's operation rate and the lower the order rate. The marketing department must keep in mind the quality and quantity of leads. 

　BtoB marketing is conducted with the following flow. 

BtoB marketing flow

　Then we will introduce how to use in each flow. 

(1) Lead generation (lead creation) x machine learning
　In this step, you will increase your awareness about your company. At first, prospective customers are attracted to events such as company websites and seminars by advertising and SEO. From the people who gather, get lead information on business cards and input forms. 

　You can use machine learning to analyze ad serving at the time of attracting customers to optimize delivery to potential segments. 

　Specifically, based on past deal information, a model is generated that predicts customerization. Use the generated model insights to identify likely user segments. Delivery to DSP (Demand-Side Platform) advertisements, etc., limited to those segments enables efficient customer attraction. 

　In addition, machine learning can be used to evaluate not only segments but also customer attraction channels. For example, using a budget for each customer channel as a variable, generate a machine learning model that predicts the number of lead acquisitions and closing numbers. Using this model, simulation can identify the channels that are effective for lead acquisition and provide optimal budget allocation. 

(2) Reed nursering (lead training) x machine learning
　It is a step to raise the interest, interest, and the degree of consideration to the company through communication such as e-mail and telephone to the acquired lead. Providing content that meets the information needs of the lead can increase interest and interest, but sending irrelevant content or sending frequent emails is not only ineffective and will not open the email in the future It may have negative effects. 

　If machine learning is used, the open rate and contract probability are predicted using the open rate of the past mail, the reading status of content, and CRM information (business type, department, post, etc.) as variables. Mail can be delivered only to highly responsive segments. 

　Not only that, you can also predict the next best action and deliver content that the person might be interested in, and predict when it is easy to open it to optimize delivery time. 

Practical use in read nursing

[Next page] We commented on machine learning utilization of "squeezing in prospects", "upsell / cancellation forecast", and "BtoC marketing".

1

2

next
Recommended articles

Why does your customer run away? Analytic techniques to improve “customer retention rate” by 3 types
Wasn't it like this? Marketing automation reconsidered, pitfalls after implementation
Why is Keyence so strong, what is "kaizen" that drives an operating margin of 55%?
Make a difference in "sales tech" in the manufacturing industry, "three technologies" that overwhelmingly automated US companies
"Chatbot" commentary for corporate personnel: ITR commentary on representative products and points of introduction
Brand improvement, marketing, PR genre seminar
[Tokyo] July 24th Adobe Symposium 2019
[Tokyo] July 24 Social whistleblowing · Internal risk measures to minimize information leakage risks
[Tokyo] July 26 Tokyo 2020 approaching! What should security do?
To List

Topics of brand improvement, marketing, PR genre
Why Kaneka's "compliance with laws and regulations" can be fatal
What is Customer Success? What kind of service do customers want? Introduce cases and KPIs
After all "people look 90%" The color of the suit changes work
How to use "red" to increase concentration, "blue" to draw ideas, and "color" at work
Was the hit secret in "color"? With the science of color that moves people
To List

IT promotion support information of brand improvement, marketing, PR genre
How to incorporate the customer experience (CX) into corporate strategy without "free running"
"How to make a strong corporate brand" as understood from Ichi, Hitotsubashi University
Looking back on 2018 "system failure cases" How to make a high recovery system "Learn from failure"
Why is it not "a organization that only pursues cognition and CVR" because Torikiki Motohiko says "rely on users"?
Why does employee health lead directly to business results? What is the deep-seated problem with industrial physicians
To List  طرح توجیهی