Monday, 28 August 2017

The out of box salesman

AI in Email Marketing

Artificial Intelligence in Email Marketing

Email acted as the catalyst for "Internet Revolution".

The birth of internet may be credited to invention of Email before four decades.

Internet's 40th birthday :  First email crashes after just two letters received

The birth of email
The birth of email - The out of box salesman

It is the Email that created the Internet Revolution and gave birth to Digital Age. This revolution is equivalent to invention of steam engine in the 19th Century which made Industrial Age possible.

Email Marketing is the most important strategy of Digital Marketing which is going to rule the Digital Age.In the overall scenario, application of artificial intelligence in marketing and specially Digital Marketing will be the deciding factor for success or failure of any business organization.

Off course, SEO (Search Engine Optimization) is gaining popularity, but Email Marketing will always persist and it is better than SEO in many ways. It establishes personal relationship with your target audience or customers and it is not dependent on the mercy of any Search Engine or any company providing the services of Search Engine.

It is not wise for any business organization having online presence to solely dependent on the Search Engine services  for generating Inbound Traffic. The change in the rules and regulations or policy of the Search Engine Service providers may severely affect  the customer base in long run or future.

It is always better to have one to one relationship with each and every customer for any successful business organization, however it may be big or small. This is the strongest point in favor of Email Marketing which is going to survive in new shape and size with new features.


Artificial Intelligence and Machine Learning algorithms are bringing automation in "Digital Marketing"  with personalized rapport building with customers.

The rise of Artificial Intelligence and Machine Learning Algorithms are going to revolutionize the Email Marketing in the Digital Age. It is going to create a giant leap for Digital Marketing people.

"How AI and Machine Learning algorithms work?" and  "What are the characteristics of the software solutions designed using AI and Machine Learning algorithms?" These questions have been answered very well here:

Artificial Intelligence in E commerce


More resources can be found here to learn about these algorithms:
Artificial Intelligence
Machine Learning

Limitations of Current Email Marketing Methods


Limitations of email marketing
Limitations of email marketing- The out of box salesman



Current Email Marketing methods have some peculiar drawbacks:

If is not easy to get new subscribers swiftly or easily. Inbound Marketing which uses SEO techniques help in gathering new subscribers. "Subscription List" is grown cumulatively over a period of time using "Organic Methods".  There are options to buy "Subscription List" from free market which is called "Inorganic Method" of growing subscription base.

The content or of email or newsletter is not personalized. So it rarely appeals to each and every reader or customer and it is hardly opened by all of them. In short, it has less "Open Rate". "Open Rate" is a ratio of the number of emails sent and the number of readers who opened it.

It is difficult to have common time or the best time for higher "Open Rate" as every individual is different and unique. Every individual has got different lifestyle which is based on his/her demographic profile and geographical location. Everybody has distinct "persona".  We can not have a common time to suit every distinct "Customer Persona".  "Customer Persona" is a term to uniquely identify a customer based on his/her country,religion,income,age,educational qualification, work experiences, lifestyle habits , hobbies, expenditure pattern, etc.


The "Delivery Ratio" of  emails sent in bulk quantity  is low. Sometimes it goes to spam folder based on "Subject Line" of the Email. Spam  Filters segregate emails based on commonly used keywords in the subject line which are known to almost all the Email Service Providers.

It is difficult to maintain long term relationship with subscriber or reader as generally content is not emotionally touching and repeatative in nature.

It offers low Click Through Rates (CTR) for the links in the email. If is not easy to quantify or measure ROI of Email Marketing Campaigns from cost benefit point of view.

Common email marketing problems
Common email marketing problems - The out of box salesman

    
If you are eager to think deeply on the above problems which every digital marketing personal faces with Email campaigns, you will notice that the problem boils down to that:

So the crux of the problem is:

Email campaign including content or newsletter is not personalized to the customer's mental framework or "Customer Persona". It lacks the personal selling rapport which happens in face to face meeting and sales transaction. So it is very difficult to predict conversion rate of many "Email Marketing" campaigns successfully.


Simple Email Newsletter Design
Simple email newsletter design - The out of box salesman




How AI and Machine Learning Algorithms can help?

Using the AI and Machine Learning techniques , the above pitfalls can be circumvented. The algorithms can classify the online subscribers or online customers as per their "persona". It will be a CPU intensive computing task to gather  information about each and every customer or visitors to the website  and store  this "Customer Persona" in database, but it can be achieved in the following steps:


Step 1:

Installation of tracker code or program on your website to gather customer intelligence or web page visiting habits of the visitors or customers.

This step requires massive custom programming in the website or online store where each and every activity of the customer is recorded. The data of customer browsing habit will be sent to central database usually on cloud platform where "Customer Persona" will be stored as a record in the database with detailed schema having  vast attributes from personal details to lifestyle details and hobbies.
              

Step 2: 

Creating individual "Customer Persona" " with AI and Machine Learning Algorithms

Step 3: 

Creating Customer Segmentation as per "Customer Persona"

It is necessary to create segmentation in the "Customer Persona" database as generally, it is not feasible to have personalized content for each unique customer. However, there is no hard and fast rule and there is a strong possibility that in the future the content will be generated and targeted for each distinct customer.

There should be balance between fine grained segmentation and coarse segmentation. If there is more fine grained segmentation of "Customer Persona" database, more unique and different content is required.
       

Step 4: 

Designing Email Campaign as per segmentation and executing it.

This step requires drafting Email content as per the characteristics of  segmentation in which all the customers have been classified. Drafting content will be a creative manual process and in near future it is hard to believe that this task can be automated using AI and Machine Learning.


Step 5: 
Monitoring the campaigns and using analysis data to modify "Customer Persona" and segmentation.

This is the most crucial step in order to algorithms work. It is a kind of feedback loop used by the algorithms
to get insight into the customers and understand them in a better way.  This step will be required to continuously update "Customer Persona" database and keep it intact and current.

What benefits will be derived for Email Marketing? 

The above exercise may give the answers of following questions:
  • What is the best time to send to increase "Open Rates"?
  • Who are the customers who open his/her email account with specific to  the email content delivered?
  • Does personalized product recommendations with customized subject line work?
  • Does personalized product recommendations increase click through rates (CTR) and ultimately sales?
  • Can we get accurate information about cart abandonment rate in online sales?


Can we use AI and Machine Learning techniques without using above algorithms?


Yes up to certain extent.
No because it is time consuming in terms of human effort along with working with common software.

It is possible to use above techniques without AI and Machine Learning algorithms and software. It is possible to implement the above techniques using traditional Email software and  popular Spread Sheet program like Microsoft Excel,etc. But it requires lot of human effort to maintain Customers Data as "Customer Persona". It requires lot of human work to create and maintain segmentation in "Customer Persona" data in the form off groups or communities.

Almost, all popular client Email software offer facility to organize emails into "Folders".  They also offer to search and query emails based on "Filters". It is possible to classify customer emails based on various demographic attributes like:
  • All customers between age 18 to 25  or 40 to 50 or above 60
  • All customers who open email between 8AM to 10AM , during lunch time 1PM to 2PM or               at night after 10PM
  • All customers  who have income in certain range or above certain range
  • All customers who buy in the first week of month, at the last day of month or quarterly , weekly
  • The group of ladies and the group of men... and the classification or segmentation is endless.

We can also term this segmentation or classification as group or communities. We can go on classifying like taxonomy of customer attributes.  We can use Microsoft Excel  sheet also to store the above segmentation or groups.

Now, the tedious tasks like updating the data has to be done every time "Email Campaign" is executed and  feedback report is obtained. Also, human effort in form of judgement , intuition , analysis and trying different strategies is required to determine predicative behavior or next step of the customer which is generally very well handled by AI and Machine Learning algorithms.



AI in email marketing
AI in email marketing


Email marketing solutions based on AI and Machine Learning algorithms 


Some of the solutions (there are many more...) from the world which use the above techniques and methodologies in less or more degrees:

http://www.optimail.io
 

This software solution provides the complete toolkit to supercharge your email campaigns with AI.
Optimail’s AI explores thousands of strategies automatically to find the strategy that works best for individual customers and segments. This permits us as data scientists the opportunity to take a peek at the data and see which strategies tend to work in general — across campaigns — in order to gain new insight into email campaign optimization.
www.automizy.com


This solution offers features that automatically tests the subject line and content variations of  auto responding and follow-up emails of various "Email Campaigns" executed. Besides it offers features to check instantly conversion rate, best performing subject lines and contents, delivery ration, list healthiness, engagement rate, open and click rate, world map and more.

www.phrasee.com 

The strong point of this solution is that it uses AI algorithms to write better email content using marketing language than human. The strength of this solution is it's AI algorithms which provide better results from better use of language combined with better statistical analysis.


Summary


It is interesting to know that Email gave birth to Internet and  we are standing at the threshold of a new era where automation in day to day life whether personal or business will be dominating factor. It is again the very basic Email which is leading the way.


The out of box salesman

9 comments:

  1. Useful information!
    Get discounts on gaming chairs & accessories by using DXRacer Coupon

    ReplyDelete
  2. Hi Ksenai,

    Thanks for appreciating my post.

    Regards

    ReplyDelete
  3. Great article! A sneak peak to the future that is very much here. My blog has a long way to go however. Need a bigger email list and I hope to take some tips from here.

    ReplyDelete
  4. Hi Vineeth,

    Thanks for appreciating the ideas. Yeah, blogging is a tough and it takes hell lot of time and patience to go ahead. We all need bigger email list and intelligence to plan and execute "Email Marketing" campaigns.

    Please feel free to share your ideas here as "Guest Post" or in comments section. It will be of great help to everybody.

    Regards

    ReplyDelete
  5. Super post.Rado te citam samo tako nastavi ☆
    https://kraljicadrameblog.wordpress.com/2017/10/05/jesen-u-mome-sokaku/

    ReplyDelete
  6. Thank you for sharing such valuable information and tips. This can give insights and inspirations for us; very helpful and informative! Would love to see more updates from you in the future.

    content marketing

    ReplyDelete
  7. Hi,

    Thanks Nina for appreciating my thoughts.

    Regards

    The out of box salesman

    ReplyDelete