In a data-rich environment like insurance, it is often easier to quantify opportunities associated with speed than it is in other industries. A global financial services company knew it took them 18 months to get new nationwide variable annuity products to the market. The company’s vast database and proven models allowed them to estimate that every day they could get to the market sooner would generate about $750,000 in extra revenue. It was obvious that cutting development time would be a good thing to do.

So the company created a cross functional team with representatives from marketing, actuary, illustrations (the people who do the models of “how much you get when you invest X dollars over X years”), risk evaluation, legal and finance. The team’s charter was to reduce the variable annuity development time by at least eight to nine months. The team was now ready to begin the DMAIC journey.

Measure and Analyze: Where Does the Time Go?

If the goal is to eliminate time, the first stage is understanding where time goes now. That is why the team started by creating a value stream map (VSM) of the process. The map started out as a basic flowchart – capturing the flow of work – onto which is layered:

• Information flow – where and what information is needed to get the work complete.
• Data on a number of process characteristics, including how long work items wait “in queue” (that is, in someone’s in-box) before being worked on, how many work items are in the process at any given time, how much time is spent on the value-added time, what quality and time metrics are monitored throughout the process, and so on.

Creating the VSM for this project was time-consuming but critical. The process involved about seven major phases and 200 detailed activities, all of which were captured on a master map along with all of the associated data. A portion of the final VSM is shown in the figure below.

Working on the VSM allowed the team to pinpoint the process steps where the biggest delays occurred. They then combined that knowledge with cause-and-effect brainstorming and identified a few factors that were slowing down the process: 

1. The team knew going in that some states had much more complicated regulations for variable annuities than other states. So it was no surprise that working out the details and the exemplars (illustrations) for those states took much longer than for the easy states. What did come as a surprise was that a number of functional areas were measured on how many “state details” they submitted per month. Naturally, they all chose to work on the easy states first so they could get more done. This meant that all the work that was hardest to do (and would take the most time) was not even started until late in the process. 

2. Those who did the exemplars and the risk models started anew for each state and each variable annuity product. In “Lean speak,” the issue is phrased as they had zero re-use. Re-use is one of the best ways to cut time in any process because people do not have to constantly reinvent the wheel. 

Improve and Control: How to Speed Up the Process

Key changes that led to the biggest gains in time included: 

1. Moving from metrics for individual functions to one that reflected the speed of the process as a whole. Thus, interim functions in the process were no longer measured based on the “number of completed requests/ month.” Rather, the primary metric used to gauge the whole process was the “number completed in the last phase” (which was sales/illustrations). 

2. Moving to a triage system, which is simply a set of criteria used to determine what work is released into the queue next. The criteria will vary by process and business needs. Here, the criteria stated that at least one “complex” state had to be worked for every three “simple” states being processed. Getting started on the more time-consuming work earlier in the process contributed about a two- or three-month reduction in development time. 

3. Creating a re-use library for illustrations, which had example scenarios that the staff could easily (and quickly) adapt to new models. Having something to start with cut between one and two months from development time. 

4. Similarly, creating a common set of codes and assumptions for all 50 states for the risk assessment team. Every model generated by the team was archived and named in such a way as to make it easy to retrieve and compare down the road. This provided a one- to two-month reduction in development time. 

The table below summarizes the impact of these changes. It shows that the company ended up cutting eight months from the development cycle – representing about $120 million in added revenue. 

Conclusion: Commit to the Upfront Work

One of the lessons that can be learned from this company is that there is enormous opportunity in speed. The company did not have to invent new annuity products; it just had to get faster in delivering its current products to the market. It did not have to use a lot of complicated data tools either; it just needed to understand how time was spent in its processes.

Perhaps the biggest barrier that companies face is the commitment to doing the upfront work. There is a lot of resistance when managers hear that a team will likely spend a month or two or three just gathering data and creating a process map. Managers and project teams are impatient for progress, and want to go out and do something that will have an impact. 

In a company with a data-rich environment, it is likely that somewhere there exists the data that can help quantify the opportunity in speed. Once it is clear that there is a financial justification for slashing time, take a lesson from this company and spend the time doing a good job on a value stream map. In situations like this, making the changes turns out to be the easy part of the project. It is figuring out what changes to make that takes the time and discipline.