Amacis and HSBC Bangladesh

Amacis and HSBC Bangladesh


HSBC Bank plc is a wholly-owned subsidiary of HSBC Holdings plc. With more than 5,000 offices in 80 countries and territories and assets of US$569 billion at 31 December 1999, the HSBC Group is one of the world’s largest banking and financial services organizations. Through a global network linked by advanced technology, the Group provides a comprehensive range of financial services: personal, commercial, investment and private banking; trade services; cash management; treasury and capital markets services; insurance; consumer and business finance; pension and investment fund management; trustee services; and securities and custody services. In 1999 HSBC Bank plc launched the first nationally available interactive TV banking service and was also developing an Internet banking offering. HSBC was quick to realize that e-mail and web forms are the natural channels for Internet based customer contact. However, many issues had to be resolved if HSBC was to manage effectively the customer interactions that the new services would generate. Not only would there be extremely large volumes of electronic interactions with these customers, but the bank would need to ensure it maintained high quality customer service. The majority of the customer interactions would require responses incorporating data that was stored in existing customer databases. Pulling information from these systems manually would make the large volume of interactions virtually impossible. In addition, with a large financial and intellectual investment in these systems, HSBC didn’t want to replace them. HSBC needed to implement a system in which it could invest full confidence. High accuracy classification, precise distribution and optimization of agent productivity were required to meet this need, in addition to safeguarding the confidentiality and integrity of each customer interaction. Finally, HSBC’s diverse customer base meant that a number of languages needed to be supported. All customers needed to be handled in a consistent fashion regardless of the customer service request language. In short, HSBC realized that it needed a system that integrated the electronic customer service requests that would result from its front-end implementation with its back-end systems, while satisfying a host of other vital requirements. HSBC initiated a search for a message management solution that could satisfy these needs.

 System Requirements:

To begin the search for a message management solution, HSBC first endeavored to translate its business problems into a set of system requirements. Weighing their existing infrastructure against the needs of customers using these new channels, the following specification was defined:

 Integration and Scalability

1.Platform: The solution had to be portable across multiple platforms in order to support existing and proposed server platforms within both the bank and its contact centers worldwide.

2.Scalability: The solution had to be capable of handling the considerable volume of interactions anticipated from both TV banking and Internet banking initiatives.

3.Integration: The solution had to integrate with HSBC’s existing customer systems, databases and other applications rather than demanding the replacement of these systems.


 1. Classification and Routing:

The solution had to demonstrate a high degree of precision (the degree to which the system returns correctly categorized responses) and recall (the degree to which the system returns a complete set of responses) in order to provide accurate classification and skills based routing functionality.

2. Security:

The product had to provide an infrastructure that would guarantee secure customer interaction.

 3. Multilingual Support:

The solution had to support multiple languages allowing for the probability of global roll-out.

4. Management Features:

Various features that allowed HSBC to monitor and manipulate existing electronic requests including real-time monitoring and system alert functionality.

5. Administration Features:

 Requirements included service level parameter setting, system back-up and archive scheduling, template management and routing policy set-up.

Vendor Company Issues

1. Vendor background:

The solution vendor required reputable references to ensure the highest quality not only of the technology but also of the vendor’s project management team.

2. Understanding of issues:

The vendor would be judged on its understanding of the various issues regarding message management within the bank.

3. Flexibility:

The vendor was required to be flexible enough to work with HSBC to resolve any issues that may arise during the course of the project.

4. Implementation:

The solution required a rapid implementation time.

 The Vendor Selection

HSBC evaluated a number of e-mail management products and vendors against its requirements. In April 2000 Amacis was chosen as the vendor to provide the customer interaction management application to answer the bank’s needs. The Amacis Visibility product – and the business partnership between Amacis and IBM – impressed the bank. Amacis Visibility met all of HSBC’s non-negotiable system requirements and also exceeded some of its expectations, in particular with regard to scalability and integration with its existing systems. “It wasn’t a difficult decision to make,” said John Rendle, Internet Services Manager at HSBC, “in the end it came down to technology. The IT view was that the Amacis Solution was truly scalable and a good fit for our architecture.” Amacis CEO, Tom Montgomery was delighted with the outcome, “Amacis Visibility was designed with this kind of organization in mind. HSBC made the right decision, realizing that our product was the only solution that would meet its current needs, and could also scale up to its future global requirements. Amacis Visibility’s connectivity features mean that HSBC can maximize current investment in their existing IT infrastructure.” He continued, “In addition to the product advantages, HSBC was attracted to our personal customer approach. Amacis sees its customer partnerships as paramount. We listened to their needs, providing HSBC with a solution to fit their core requirements and also tailoring the product to best fit their wants.” Lee Longhurst, Manager of Internet Systems at HSBC concurred, “It was the only product that could handle the message volumes we proposed. We knew that Amacis would be flexible enough to provide a solution for our needs on a timely basis.”

 The Project

Amacis began implementing Amacis Visibility in May 2000, over a period of six weeks. HSBC intends to roll out the new service in 3 key stages. Each phase will allow time for agents and administrators to come up to speed with the new skills required for e-mail response management and will enable web administrators to optimize the service.

Phase 1: A number of specific web forms to cover technical support and feedback for Internet Banking Customers.

Phase 2: E-mail support for customers of specific business units in the Commercial sector.

Phase 3: Will offer management of secure, authenticated, free format messages delivered through Internet Banking.

HSBC has no doubt that it will be a global leader in Internet banking – embracing pervasive computing developments such as digital TV. The bank will be rolling out Internet banking in other key geographies and will implement Amacis interaction management as a key component of its Internet proposition.

 About Amacis Ltd.

Amacis Ltd, the e-channel and m-channel management company is a leading provider of customer relationship management software that transforms the way global enterprises manage customer interaction. Amacis offers a unified e-channel and mobile-channel management environment called Amacis Visibility. An Amacis solution allows traditional brick-and-mortar companies and e-business enterprises to maximize customer retention and profitability, through provision of the most scalable and secure customer interaction management environment available. Amacis manages all forms of Internet-based customer contact consistently across a variety of channels and integrates with other corporate customer systems and call centre operations. With its main R&D facility in Belfast, Northern Ireland, Amacis also has offices near London and its US headquarters are in Boston, MA.

 News Release

Below provided is the news release through which HSBC PLC declared that they have chosen Amacis as their business partner:


Amacis and IBM implement strategic solution for customer interaction management in one of the world’s largest banking and financial services organizations BOSTON, MA, August 14, 2000– Amacis, Inc., the only company to provide global-class solutions for customer interaction management, today announced that it has been chosen by HSBC to implement its Amacis Visibility solution. HSBC Group’s international network comprises more than 5,000 offices in 80 countries and territories, including APAC, EMEA and the Americas. This implementation by Amacis and IBM will allow HSBC to build better relationships with its online customers by directly managing all forms of electronic interactions. Amacis was selected by HSBC specifically for its e-suite’s scalability, portability across multiple platforms and ability to integrate with the bank’s existing customer systems. In addition, HSBC saw the strengths of Amacis and IBM’s relationship as a significant factor that impressed them relative to Amacis’ offering. HSBC was quick to realize the need for a strategic customer relationship management solution as it brought its services online, and selected Amacis in an open bid. HSBC has already implemented within the UK and will use Amacis Visibility to support all of its customer interactions that are anticipated from digital television, Internet and wireless banking initiatives. The ability of Amacis Visibility to handle large volumes of customer interaction coupled with the e-suite’s language capabilities meant that it was robust enough for HSBC’s highly distributed environment. “It was the only product that could handle the message volumes we proposed,” said Lee Longhurst, Manager of Internet Systems at HSBC. “We knew that Amacis would be flexible enough to provide a solution for our needs on a timely basis.” HSBC needed to implement a system in which it could invest full confidence. High accuracy classification, precise distribution and optimization of agent productivity were required to meet this need, in addition to safeguarding the confidentiality and integrity of each customer interaction. “Amacis Visibility was designed with this kind of organization in mind,” said Tom Montgomery, CEO, Amacis Inc. “HSBC made the right decision realizing that our product was the only solution that would meet its current needs, and could also scale up to its future global requirements. In addition to Visibility’s product advantages, HSBC was attracted to our personal custom approach. Amacis sees its customer partnerships as paramount. We listened to their needs, providing HSBC with a solution to fit their core requirements as well as further tailoring the product to best fit with their business.” Amacis began implementing Amacis Visibility in May of 2000, over a period of six weeks, meeting HSBC’s need for rapid implementation. The product integrates with HSBC’s existing customer systems, databases and other applications, protecting its financial and intellectual investment in these systems. Amacis stood out from the competition by providing a strategic enterprise solution, rather than a small tactical one.

 Amacis and HSBC Bangladesh

 How It Began

In this year, the CEO of HSBC Bangladesh, advised the marketing and public relations department to partially implement the Amacis Visibility solutions for the customers. In Bangladesh, internet banking has not been introduced by HSBC. However, marketing through sending e-mail messages is a viable option. HSBC mainly operates in the upper end of the market, that is, customers who are able to deposit higher amounts are targeted. It is quite logical to believe that at least half of the customers of HSBC have their own e-mail addresses. The Marketing and PR department of HSBC Bangladesh expects that it can get high response from its existing customer base, through sending them e-mail information regarding new products and services. These messages will tell the customers how to get the most out of their existing product subscriptions, as well as how availing new services will enhance their wealth.

The interns of HSBC’s marketing department have been assigned the task of building the e-mail database of the customers of HSBC. Way back in 1996, not many people had e-mail addresses in Bangladesh. Thus the e-mail addresses had not been collected from the customers then. Equipped by the home and office phone numbers of those customers, the researchers will collect the e-mail addresses from the customers, in a polite and professional manner. After collection e-mail addresses, e-mail messages will be sent. Amacis has all the ability of managing this campaign. Once all the e-mail addresses are inputted in the software, it starts working. It sends the e-mails. It can also track how many person received the messages, how many of them read it, how many of them deleted it without opening, etc. information. This information is critical in judging the performance of the campaign.

Though Amacis Visibility 2 provides a spectrum of facilities and options, HSBC Bangladesh has opted for the email campaign option. The project has been assigned to the GPA (Group Public Affairs) or Marketing and Public Relations department, as already mentioned. The IT (Information Technology) department is providing the technical support for this project. HSBC Bangladesh is using a shared version of the software. HSBC PLC procured a version of the software which’s license enables its Bangladeshi concern to use the software without incurring additional costs. Before launching the campaign, the Marketing department had to consider many factors.

The Email Marketing Process

The email marketing process of HSBC Bangladesh has been divided into three phases.

These are:

1. The Preliminary Phase

2. The Interim Phase

3. The Final Phase

The Preliminary Phase

In the preliminary phase, all the details of the campaign are finalized. No email is sent during this phase. HSBC Bangladesh went through the following steps in this phase:

1. Key strategy formulation

2. Choosing the tools

3. Writing an eye-catching copy

4. Following up effectively

5. Starting an Ezine or Newsletter

6. Complementing the email messages with links to website resources

7. Including unsubscription/remove from mail list option in the mail message

8. Choosing the Appropriate Software for Sending Automated Email messages

 Key Strategy Formulation

Here are some key strategies adopted by HSBC for their email marketing campaign:

1) Developing powerful benefits packed sales letters and ads for the products

HSBC plans to write the email messages in a way so that it clearly indicates the key benefits of the discussed products.

2) Continuously comparing email marketing with other forms marketing

HSBC plans to monitor the effectiveness of the campaign, continually, by comparing it with other traditional means of marketing. However, this will only be effective if the customers can buy their products online. If a customer clicks on a link provided in an email, and eventually ends up availing a new product or services, that can be treated as a success of the campaign.

3) Building and marketing towards the contact database and then following up

HSBC has to make sure that only interested people are included in the email database. If someone is not willing to receive promotional offers through emails, he or she should be removed from the mailing list. The other things that HSBC are planning to do are:

  • Offering and advertising free reports and other information
  • Following up systematically and consistently with the contacts.
  • Writing articles that appeal to the target market
  • Providing great content and information which are valuable for the contacts.
  • In the sales letters and follow-up, providing hyperlinks so that contacts can be driven back to the web site and order forms

4) Know who the target market is and working hard to find them

Currently, only the existing customers of HSBC are being included in the mailing list. However, there are many other people who are interested in the different products and services offered by HSBC. HSBC should work hard to trace out these people. Similarly, the uninterested people should be immediately removed from the mailing list.

Choosing the tools

The types of tools that are required for launching a typical email marketing campaign are:

  • Follow-up Auto responders
  • Secure Ordering Pages
  • Email Marketing Software
  • Web Page Designer Software

Auto responders are software program, which reply to email messages, through an objective judgment of the email content. That is, it has a list of certain words in it—in order to describe certain customer attitudes. If the program sees “don’t send”, “useless”, “I don’t want”, etc. phrases in the emails sent by a customer, the program will assume that the customer is not satisfied with the emails sent to him. It will then remove that respondent from the mailing list. However, there are ample chances of errors in such an automated system, due to the fact that human beings use the same words to describe different matters. Moreover, email marketing is still a test case in Bangladesh. HSBC does not want to use Auto responders, at this moment. HSBC wants to maintain a personalized touch in this phenomenon.

Secure ordering pages are very crucial for the success of an email marketing campaign. If a customer cannot order the products and services offered in the email messages in a secured manner, he or she will never feel confident about any such approaches. HSBC Bangladesh gets most of its web resources from HSBC group head quarters. Thus the security issue is taken care of by world class experts, who have the experience of handling online banking transactions for a long time.

The email marketing software is very important. HSBC went through a lengthy research procedure before selecting Amacis Visibility 2 as the appropriate software for this campaign. The decision for HSBC Bangladesh was easy—as the group is using this software, they did not have to look elsewhere. The reasons for selecting Amacis have already been discussed in another part of this paper.

The web page of HSBC Bangladesh ( is the heart of the email marketing campaign. Every email message contains some links to the different sections of this site. The software used for designing this site was selected after taking advices from the concerned authorities. Only globally accepted and licensed, full version software had been used in the designing process. The web content was mainly created in html (Hyper Text Markup Language), using Notepad™ as the editor. The graphics and images were created by Macromedia Flash™ and Adobe Photoshop™, as well as Adobe Illustrator™.

Writing the Copy

Copy writing is a creative process. This requires imagination, creativity, as well as in depth knowledge of the subject matter. All the members of the marketing department sit down, and decide upon the copy before converting it into an email message. During writing the copy, they seek advice from the product team of HSBC. Their input is of immense value, because they are the creative brains behind the product.

 Effective Follow Up

Following up is often the toughest task in email marketing. Many email addresses will bounce back, some emails won’t be sent at all, and to make things worse, some customers will send replies to the sent mails! To make the campaign truly effective, HSBC needs to find out why some mails are bouncing back. There might be a number of reasons behind bounced back email messages:

 Firstly, if the address is invalid

Secondly, if the user has exceeded his mailbox size

Thirdly, if the address has become inactive

If the address is invalid, HSBC can ask the owner of the address to provide an updated email address. The same measure can be taken for inactive email addresses. However, for the second case, HSBC cannot do much. The main problem regarding bounced back mails is that it is almost impossible to know why the mail bounced.

Starting an Ezine or Email Newsletter

The email marketing campaign can be further enhanced by starting an email newsletter. This way HSBC can let the customers know more the bank itself. This could work as a welcome change for the customers who would get bored by viewing product information, on an on. Currently, HSBC has a newsletter that is printed out and sent to every customer via courier. The software version of this newsletter can be sent to the users via email.

 Complementing the email messages with links to website resources

This point has already discussed in the strategy formulation part. HSBC intends to include links in the email messages, so that the users can seek further information on the products. As an example: the email about HSBC’s “Day & Night Banking” included basic information about the whole service package, but it only had links for each of the different services (ATM Card, EasyPay Machine, etc.). The user can get more information on each components of the service package by clicking on the links. This is done with the intention that after the user has seen the particular page, he or she might like to browse the other pages of the website as well. Thus the marketing objective will be further enhanced.

 Including unsubscription/remove from mail list option in the mail message

This is a very critical option. The customers do reserve every right of opting not to receive any promotional mails. Thus an “Unsubscribe” option should be included with every email message sent. This will ensure customer satisfaction, which is in the top priority list of HSBC.

Choosing the Appropriate Software for Sending Automated Email messages

This topic has been thoroughly discussed in the previous chapter.

The Interim Phase

In the interim phase, the customer database is created. There were several criterions for this process. There are different types of account holders at HSBC. Some are:

a) Power Vantage Account Holder (PVA)

b) Corporate Employee Privileged Scheme (CEPS) Account Holder

c) Non-Ceps/Normal Account Holder

The PVA customers have the highest deposits in the bank. To become a PVA account holder, one has to deposit at least 5 lac Taka in HSBC. These customers were approached first. Then gradually CEPS, CEPS with account balance above 50 thousand, CEPS below 50 thousand taka, Non-CEPS above 50 thousand and finally Non-CEPS below 50 thousand balance customers were approached. The customers were approached through telephones, both land and mobile. The guidelines for approaching customers were:

a) Try not to say or do anything that might dissatisfy the customer

b) Do not ask any irrelevant question

c) When calling at someone’s home, apply extra caution

d) If a person could not be reached after calling at the same number twice, that particular number should be leaved alone

e) If a person did not provide a local number, he or she would not be contacted

f) Customer’s of Chittagong will not be contacted

g) If the customer asked for any information during the collection process, the researchers were advised to forward them to the concerned departments.

These guidelines were strictly followed during the interim phase. This ensured that customer satisfaction levels would not go down. After 1,000 customers were contacted, the information was inputted into computers. Two databases were actually made, from this data. The second database is a derivative of the first one, and it is required for the third phase. The following information was included in the first database:

1. Serial Number

2. Branch Code

3. Name of the Customer

4. Email Address

5. Home Telephone Number

6. Office Telephone Number

The final database looked like the one provided below:










MR S. M. JASHIMUDDIN Updated Number Required818199812579



MR AKBAR HOSSAIN SONNETSENIOR MANAGEMENTintech@aitlbd.net018-216661408682



MR MOHAMMAD AMINUR RAHMAN Telephone Number Required  

 The serial number is an arbitrary number assigned for each customer. There were repetitions, because every new list started from 1. The lists were based on the type of account and average balance, as already mentioned.

The branch code denotes in which branch the customer opened his account.

1 means the account had been opened in HSBC DAK, the Head Office of HSBC Bangladesh at Sonargaon Extension Road

2 means the account had been opened in HSBC DMO, the Dhaka Motijheel Office

3 means the account had been opened in HSBC DGU, the Gulshan Branch

4 means the account had been opened in HSBC CHG, the Chittagong Branch

5 means the account had been opened in HSBC DHA, the Dhanmondi Branch

At first the researchers were provided with secondary data, which included at least these two data, the branch name and the name of the account holder. The rest information— employee name, email, home telephone number and office telephone number were provided for some accounts, while some were not. The objective was to make sure that the column “email” did not have any blank entry. Only five types of entries were given into this column. These were:

1. Email address

2. N/A (Not Available)

3. UNR (Updated Number Required)

4. TNR (Telephone Number Required)

When a customer could not be found in any of the provided phone numbers, or when all the provided number are overseas numbers or number of other districts of Bangladesh, the entry would be “UNR”, which means that an updated phone number for that customer is required.

When no telephone number was provided for a particular customer, the entry would be TNR, which means a telephone number is required.

If the customer declined to disclose his or her email address, or a customer did not have an email address, the entry would be N/A or Not Available.

During the interim phase, daily results were calculated and strictly monitored. The average expected efficiency was fifty percent. Efficiency was calculated through using the following formula:

Efficiency=           Total Number of Emails Collected   × 100

                                           Total Attempts

Where total attempts = N/A + UNR + TNR + Email

On 21st July,

Total Number of Emails Collected were: 198

Total Attempts were: N/A (41) + UNR (16) + TNR (5) + Email (198) = 260

Efficiency = 198/260 × 100 = 76.15 %

Shown below is part of a result sheet:






Total Attempts












































 On Thursdays the number of email addresses was comparatively lower, because, on Thursday the researchers would work half day, from 9 AM to 2: 30 PM, compared to 5:30 PM on other days of the week. After all the lists had been completed, the final phase starts.

 The Final Phase

This phase begins with the creation of the second database. This database is called the “Name & Email database”.  This database was created from the database created in the previous phase. The first database had six columns. The second database had only two columns; the name and email column. There were some criteria for this database. These were:

1. There should be only one name for each email entry.  If two or more customers had a joint account, only one person’s name had to be used.

2. Customer names cannot include these signs: . (period) , (comma) & (ampersand) – (dash) \ (slash) / (back slash) ` (quote sign) etc.

3. Only one email address had to be used for a customer, even if he or she has multiple email accounts.  As a rule of thumb, the web mail addresses were preferred over the ISP (e.g. is preferred over accounts, because people are prone to changing ISPs every now and then.

The first two criterions are the requirements of the software, Amacis. If a name entry includes any of the forbidden signs, that particular entry will not be counted by Amacis.

After creating the Name & Email database, all the entries are saved in a CSV file (Microsoft Excel Comma Separated/Delimited File). This CSV document is then inputted into Amacis Visibility2™. Amacis analyzes all the entries, and deletes the duplicate entries. An entry is a duplicate entry; if another entry exists with the same name (same spelling of name) and same email address. However, sometimes people with same name can have the same email address. The CEPS account holder often prefer to use their organization’s email address, and it is not very unusual to find two people with the same name in a large organization. This problem would not occur, if a third variable could be added into Amacis, e.g. home phone number, or home address.

After creating the CSV file, the email message is created. The messages are usually a combination of text and graphics. The text is in html format, and the image is usually in JPEG format. Usage of the compressed image format JPEG keeps the mail size minimal. Using html enables HSBC to include hyperlinks within the mails. The significance of hyperlinks has already been discussed.

Once the email message is finalized, the campaign supervisor has to run a dummy campaign. Some local and overseas staffs of HSBC are sent the email message through Amacis. If everything runs smoothly, the CSV file is fed into Amacis. Then Amacis sends emails to the customers. Amacis provides instant statistics regarding the campaign. The campaign results are saved in the HSBC’s group intranet. Authorized personnel can download and view the results by using an appropriate user ID and password.

Analysis of Campaign Results

 Email Campaigns at HSBC

Till now, HSBC’s marketing department has conducted two email marketing campaigns. Another campaign is on the way. The first campaign was regarding the monthly interest bearing time deposit scheme. This campaign was codenamed MITD. The second campaign was regarding “Day and Night Banking”, the latest service package offered by HSBC. This campaign was code named DAY.

Campaign Results

The results of both the campaign are given below:

Campaign NameTotal No. of TargetsTotal E-Mails SentNot SentBounced Back EmailsMails OpenedNot Opened/No ActionClicked on The Links
Day & Night960496013200022195382771

Eighteen thousand (18,000) customers were approached, during the research phase of the campaign. Out of these 18,000 customers, 9604 customers provided their email addresses. The rest 8396 entries had to be marked “N/A”. It has already been discussed in sections 1.5 and 6.4, for what reasons an entry has to be termed as N/A.

The following table and chart shows the email collection results, in brief:



Email Address Collection Results

Customers Contacted


Email Entries


Not Available Entries



Statistical Analysis of Results

The researchers had to work with a sample of all the customers of HSBC Bangladesh. The population consists of all thirty five thousand customers of HSBC. For this reason, direct inferences cannot be made, from the obtained results. Although the sample was quite big, it cannot be said that HSBC’s 53% customers have email addresses. Thus we have to apply statistical analysis in order to get some reliable inferences.

 Sample Size Determination

Stratified Sampling Technique will be used for determining the sample for this research. The total population of 18,000 customers will be divided into nine stratums. Each stratum contains 2000 entries. From these 9 stratums, stratum 2 (entries 2000-4000) is selected through lottery. Now a sample of 50 will be determined from this stratum using SRS (Simple Random Sampling).

 The 50 entries are:

2236, 2413, 3757, 2891, 3263, 2933, 2967, 2597, 3997, 3685, 2138, 3236, 2700, 3306, 2059, 2641, 2559, 2252, 3336, 3378, 2242, 2883, 3906, 2566, 2988, 3327, 2420, 2643, 2642, 3099.

The values obtained from these entries will be used in all statistical inferences.

After analyzing the values, the following results were obtained:



Not Available




Hypothesis Testing

Hypothesis testing is an effective statistical tool. It is a very popular marketing research tool. There are eight steps involved in this process.

 Formulating the Hypothesis

The null hypothesis for this test is:

H: Most customers of HSBC do not have an email address

The alternative hypothesis for this test is:

H: Most customers of HSBC have an email address

Thus the appropriate formulation for the hypotheses is as below:

H: π ≤ 0.50

H: π ≥ 0.50

 Selecting an Appropriate Test

As the sample size n>30, the appropriate test here will be z test.

z= p – π/ σp

Where σp2 = π(1- π).n-1

 Choosing Level of Significance

Level of significance is denoted by α. The α for this test is 5%, which means 5% error levels will be tolerated, and the confidence interval is 95%.

Determination of Sample Size and Calculating the Value of Test Statistic

Sample Proportion p = 13/30 = 0.433

σp2 = π(1- π).n -1

      = .5(1-.5).(30) -1

σp =   0.091

thus z = p – π/ σp

          = (.433-.5)/ 0.091

          = -0.734 = -0.74 (approx)

 Determining The Probability  

Using standard normal tables, the probability of obtaining a z value of -0.74 can be calculated.

The lined area between –infinity and -0.74 is 0.2296 (0.5-0.2704). Therefore, the shaded area to the right of z= -0.74 is 0.2704 + 0.5 = 0.7704

Comparing the Critical Value and Making the Decision.

Here, the probability of getting a p value of 0.433 is 0.2296, which is lower than the level of significance (0.5).

Thus we can reject the null hypothesis.

Here probability of TSCAL (0.2296) < significance level (0.5)

Final Decision: Most Customers of HSBC Have an Email Address

Chi Square Test

The chi-square statistic (χ2) is used to test the statistical significance of the observed association in a cross-tabulation. The following two-factor cross table will be used for this test:

Mail Subject Line and Mail Opening Rate

Campaign NameMail Opened/ReadMail Not OpenedTotal
Column Total480110401 

 Note: For both campaigns. 2000 mails were bounced back. That category has been overlooked for this test.


fe = nrnc/n

Where fe = Expected cell frequency

            nr = total number in row

            nc = total number in the column

            n   = total sample size

For the data in the table, the expected frequencies for the cells, going from left to right and from top to bottom are:

(7601 ´ 4801) ¸ 15202 = 2400.5(7601 ´ 10401) ¸ 15202 = 5200.5
(7601 ´ 4801) ¸ 15202 = 2400.5(7601 ´ 10401) ¸ 15202 = 5200.5

 Then the value of χ2 is calculated as follows:

χ2 = ∑all cells(f0-fe)2 ¸ fe

f0 here is the observed value for each cell

For the data in the table, the value of χ2 is calculated as:

χ2 = {(2219 – 2400.5)2 ¸ 2400.5} + {(5382-5200.5) 2 ¸ 5200.5} + {2582 – 2400.5) 2 ¸ 2400.5} +

         {(5019-5200.5) 2 ¸ 5200.5}

      = 13.72 + 6.33 + 13.72 + 6.33

      = 40.1

Now, for this chi square test, the null hypothesis is that there is no association between people’s tendency to open email messages and the subject line of the mail.

In order to accept or reject this hypothesis, we need to determine the degrees of freedom

Degrees of freedom (df) = (r-1) ´ (c-1)

Here, r= number of rows and c=number of columns

Thus df= 1´1 = 1

The critical value = 3.841 (for df =1 and level of significance 0.5)

The calculated value = 40.1

Tcal (40.1) > Ttheory (3.841)

Thus, the null hypothesis of no association cannot be rejected indicating that the association is not statistically significant at the .05 level.

Final Decision: There Is No Association Between The Subject Line of The Email Messages,and Viewing Tendency

 Statistical Findings

From the first test, we found out that most HSBC customers have an email address. This means that it is actually feasible to run more email marketing campaigns in the future. If most customers did not have email addresses, running an email campaign would be too expensive and less results bearing.

From the second test, we found out that the subject of the email messages do not play a significant role in motivating people to open the email. It can be assumed that people open the emails because they are sent from HSBC. People do not open the emails because they are interested about MITD or Day and Night Banking—they open them because the messages are from HSBC. This is an alarming fact for HSBC, because after some time people will realize that the mails sent from HSBC contains only product information, and therefore is not important. They will then simply delete the messages, as soon as they realize that they have been sent from HSBC.  In order to change this situation, HSBC has to make the email messages more informative. Ways of improving the quality of the messages has already been discussed in another part of this report.

 Other Findings

Campaign NameTotal E-Mails SentMails OpenedNot Opened/No ActionClicked on The Links
Day & Night960122195382771

 From this table, we can get the following data:

34.75% People Responded to The Day & Night Campaign, by clicking on the links for additional information

41.9% People  Responded to The Day & Night Campaign, by clicking on the links for additional information

 This data is true for those people who opened the email messages. Thus, this data cannot be used for making conclusions. Out of 2219 people (23.11% of the population that has email addresses), 34.75% clicked on the links. Thus the effect created by the email messages is quite poor. As most people did not even bother to open the emails, the justification of the entire campaign comes under question. There are also a high number of bounced back emails. This matter can be resolved by double checking with the customers, and updating the database once more. The only ray of hope for HSBC is that most of their customers have email addresses. They need to increase the value of the email messages sent. The email message should bear some value for the customers instead of providing raw facts.


After completing the research, the researcher has came up with a number of recommendations for making email marketing campaign more successful. These are:

1. Whenever a customer opens an account, he should provide his email address in the account opening form.

2. Customers should be asked whether they want to receive email messages or not

3. The email address collectors should be trained extensively, so that their actions do not result in any dissatisfied customer

4. In order to decrease costs, only land phone lines should be used in collection

5. Email databases should be checked manually, and “same name, same email address” cases should be carefully analyzed

6. Similar name entries should be differentiated by using variants and shortened forms of one’s name, e.g. Mesbah Ur Rahman can be written as M U Rahman, Mesbah U Rahman, etc.

7. More people should know how to operate and use Amacis, so that absence of a certain individual does not harm the campaigns

8. Customers replies should be read and acted upon accordingly

9. Customer requests for removal from the list should be fulfilled within the shortest possible time

10. Customer requests for additional information should also be fulfilled within the shortest possible time

11. The value of the email messages should be enhanced by providing discount coupons, raffle draw tickets, critical information, new investment opportunities, etc.

12. People who click on the links should be treated in a special way, both in the web site and through future email messages. This can be implemented by using “cookies”.


Email marketing is still an experimental case for HSBC. Although there is a gulf of difference between expected results and the actual scenario, the future seems bright for email marketing. More than 50% of HSBC’s customers have an email address, which is quite amazing. Although  we are reportedly going through a large scale IT adoption process, the usage of internet is quite low in the corporate context. If HSBC can create maximum value within their concise email messages, they might remarkably improve their sales and eventually they may end up with a higher market share. No other bank is pursuing email marketing at the moment, which is a great opportunity for HSBC. Whether HSBC can capitalize on this opportunity or not, will depend largely on the clever execution of the campaigns.

Amacis and hsbc Bangladesh