Process Mapping Cuts Costs, Increases Throughput by 30%
![]() |
HSBC Bank USA is the U.S. subsidiary of HSBC Group, one of the world's largest banking and financial services organizations. In the past several years, its retail banking footprint has grown dramatically to include many parts of the United States including New York State, where it has more than 400 branches.
Providing marketing support to those branches can be challenging to say the least. There are regular new branch openings, constant improvements in signage, and a steady stream of new promotions that require not just in-branch materials, but training support. If that isn't enough, each branch manager needs to be able to execute community-specific direct mail campaigns or run advertising and sponsorships. And of course, there are minor differences in each local market that make each project just a little unique, making off-the-shelf solutions somewhat impractical.
Keeping up with this growing demand for support was proving difficult for the marketing support team. The branch managers were complaining that not only was it taking too long for marketing to fulfill requests, but there was great variability of delivery, making it nearly impossible for them to form realistic expectations of when they would get their program materials. This concern was showing up in the satisfaction scores marketing was getting, which had recently dipped to an average of 2.9 on a 0 to 5 scale.
HSBC had a Six Sigma team leader (on loan from IT) review the situation with the vice president of marketing. They agreed that they needed to first get a baseline on just how bad things were compared with the customer (branch manager) expectations for faster, on-time delivery of programs better suited to their local markets. They formed a team comprised of marketing and branch associates to investigate.
Getting a Baseline
In reviewing job jackets for several hundred projects completed in the past year, the team found that projects could generally be categorized as Long Duration, which were less routine and more customized (things like ad campaigns and branch opening plans), or Short Duration, which represented the vast majority of projects and were fairly routine (projects including seminar invitations, flyers, posters, and in-branch materials). Overall they found that:
In reviewing job jackets for several hundred projects completed in the past year, the team found that projects could generally be categorized as Long Duration, which were less routine and more customized (things like ad campaigns and branch opening plans), or Short Duration, which represented the vast majority of projects and were fairly routine (projects including seminar invitations, flyers, posters, and in-branch materials). Overall they found that:
- project duration ranged from 1 to 370 days;
- on-time delivery varied from 101 days early to 307 days late; and
- 75% of projects were at least 10 days late.
Figures 1a and 1b provide a summary of the problem:
![]() |
Mapping the effects back to causes seemed to suggest that there were several steps in the project execution that were adding more time (and probably human error) than value.
Process Mapping
By drawing a detailed map of the project management processes within the marketing group, the team identified a communications group whose contributions were largely redundant with tasks being performed by others in the loop. Studying the job jackets further, the team determined that restructuring the department to eliminate the communications group and allowing segment managers to interact directly with agencies and production vendors would clarify roles and responsibilities, reduce project cycle times, and reduce the variability surrounding project completion dates. (See figures 2a and 2b.)
By drawing a detailed map of the project management processes within the marketing group, the team identified a communications group whose contributions were largely redundant with tasks being performed by others in the loop. Studying the job jackets further, the team determined that restructuring the department to eliminate the communications group and allowing segment managers to interact directly with agencies and production vendors would clarify roles and responsibilities, reduce project cycle times, and reduce the variability surrounding project completion dates. (See figures 2a and 2b.)
![]() |
Solution and Measurement
Reorganizing the department wasn't going to make the problem go away completely. To ensure continuous improvement over time, the team needed to create a new measurement system to capture the data necessary to monitor success on key variables. It proposed a series of eight key metrics to provide diagnostic and evaluative insights as seen in the table below.
Reorganizing the department wasn't going to make the problem go away completely. To ensure continuous improvement over time, the team needed to create a new measurement system to capture the data necessary to monitor success on key variables. It proposed a series of eight key metrics to provide diagnostic and evaluative insights as seen in the table below.
![]() |
But team members still needed a way to capture the data reliably to feed the measurement system. That's when they hit on the idea of automating the project request process and allowing the branch personnel to submit requests via a Web interface system that later became known as MERF (Marketing Ez Request Form). For about $20,000, the IT department built MERF based upon the analysis the marketers provided with respect to the types of projects and parameters of flexibility for each.
Branch managers loved MERF because it made ordering marketing-support elements fast and easy. The marketing team loved MERF because it:
- answered many of the requestors' questions automatically;
- captured all the data necessary to monitor the key metrics; and
- linked directly to the agencies and vendors, completely automating the delivery cycle for many of the most routine requests.
Results — Big Gains in Efficiency and Satisfaction
In the end, project cycle time was reduced nominally, but variability was reduced dramatically, making it easier for branches to manage their needs effectively (figure 3). As a result, on-time performance went from 26% to 80% in the first year. And thanks to the new process flow (and MERF), marketing administrative costs were reduced by $700,000 in the first year with an estimated increase in project capacity of 30%.
In the end, project cycle time was reduced nominally, but variability was reduced dramatically, making it easier for branches to manage their needs effectively (figure 3). As a result, on-time performance went from 26% to 80% in the first year. And thanks to the new process flow (and MERF), marketing administrative costs were reduced by $700,000 in the first year with an estimated increase in project capacity of 30%.
![]() |
Oh, and marketing's satisfaction scores awarded by branch managers took a jump in a positive direction, too.










