Pipeline Management. David Kirjassoff. DEK Management Consulting

Size: px
Start display at page:

Download "Pipeline Management. David Kirjassoff. DEK Management Consulting"

Transcription

1 Pipeline Management David Kirjassoff DEK Management Consulting

2 Focus How can a development engineering organization Increase the flow of new products By taking the perspective of operating a development pipeline?

3 Pipeline Thinking Common for operations/manufacturing (and chemical engineers!) Processes and workflow are visible, and explicit Work flowing through the pipe is routine and repeatable Capacity, throughput, bottlenecks are designed in and managed The pipeline is managed for continuous improvement But it hasn t always been that way

4 Pipeline Evolution in Operations Lean Manufacturing Efficiency Load and Chase Production Control Job Shop Time

5 Pipeline Thinking Less common for design/development Process and workflow is invisible, and in a constant state of flux Work flowing in the pipe is highly variable, often indeterminate Capacity, throughput are often secondary considerations Projects are managed for improvement, often the pipeline is not

6 A Sample Development Pipeline

7 Common Development Pipeline Problems Good new ideas spoil; waiting for someone to start working on them Too many projects for too few resources Constantly shifting priorities New work slips in unannounced Half-baked work is rushed to completion Lots of work is sort of on hold, but nothing ever goes away

8 Pipeline Management Definition: The process of managing a portfolio of separate development projects across a common group of resources (people, processes, tools) Three main focus areas: Pipeline Design What do we want our development process to look like, and what capacity do we want to establish? Pipeline Operation How do we best operate the pipeline we have, and how do we know how well its working? Pipeline Improvement How do we take what we are learning during pipeline operation, and translate it into improvements in pipeline design?

9 Pipeline Design Considerations Length Where does the development pipeline start and end? Width How many projects do we want to work on at once? How much capacity do we need to provide? Shape What should the pipeline look like? One pipeline for all, or special purpose pipes?

10 Pipeline Length Start Point Beware the fuzzy front end Idea Concept Product Consider separating discovery from development Discovery Team Development Team Idea Concept Product

11 Pipeline Length End Point Avoid the sustaining engineering black hole Development Sustaining Separate design resources from sustaining resources Clearly define the handoff from development to sustaining

12 Pipeline Width How many projects do we want to work on at once? Consider two extremes: Start Many Start Few Resources Project A Project B Project C Resources Project A Project B Project C Time Time

13 Advantages: Start Fewer, Finish Quicker Multiple resources involved, easier to deal with resource disruption Smaller window to introduce changes or deal with reprioritization Easier to forecast customer needs over shorter time horizon Cycles of learning for both team development and project content Benefits of completed work begin accruing earlier

14 Pipeline Width How much capacity do we need to provide? Average Cycle-Time Capacity Utilization What has the greater impact your profitability Time-to-market or Productivity 100%

15 Pipeline Shape

16 The Pipeline(s) have to Supply the Product Portfolio New Existing Markets Derivatives/Spins Revs/Updates Existing Technology Breakthrough Products Performance Enhancement New

17 Pipeline Operational Considerations How do projects get defined and initiated? How are priorities established and maintained across pending and active projects? How are resources allocated to projects on an initial and ongoing basis? How is forward progress tracked and maintained for active projects? Let s look at these from both a getting started and a best practices perspective

18 Getting Started Inventory current projects Drain the swamp Prioritize what s left Get some control on the input

19 Inventory Current Projects Generate a list of all active and on-hold projects Identify (or assign) an owner for each project Ask owners to estimate: Effort remaining (man-hours, calendar time, $$) Business benefit ($$, H/M/L, etc.) Team members involved (and time %) Current status (active, on-hold, stalled) Obstacles to completion Tabulate and circulate summary for review/comment

20 Inventory Format Project Name Description Status Effort Remaining Business Benefit People Involved Completion Obstacles System Upgrade New call center script system On-hold 6 man-weeks, $12,000 High, allows growth with current staffing System analyst, IT Pending resources from IT

21 Drain the Swamp Schedule a meeting to review the inventory Evaluate the work Expedite the easy wins Identify what to stop doing

22 Evaluation Matrix High Low Hanging Fruit Prioritize for Completion Low Impact Finish Quickly or Stop Candidates to Stop Low Effort High

23 Prioritize what s left Rank order remaining significant projects Assign resources in priority order Communicate priority across the team Establish ground rules for priority: Queues are handled in priority order In case of resource constraints, assign to higher priority projects first

24 Get control of the pipeline inputs Declare a temporary moratorium on initiating new work until inventory process is completed Dedicate separate resources to sustaining tasks and projects

25 Best Practices Input control Phase gate review process Pipeline control room

26 New Ideas Input Control Methodology Establish a sequence of two queues for potential new projects: Valve 1 Is this a good idea? List of ideas awaiting review Prioritized list of good ideas awaiting resources Valve 2 Do we have the resources? Prioritized list of active projects Active Projects

27 Valve 1 Good Idea Test Establish a process for submitting and collecting ideas for projects Minimize bureaucracy, keep it simple Review ideas frequently, avoid spoilage Rank ideas for fit with team strategy, business impact Select ideas to pass through the valve Estimate resource requirements

28 Idea Evaluation Matrix Fits our Charter High Low Just say no Just say no Prioritize and pass to second valve Refer elsewhere Low Good ROI High

29 Valve 2 - Resources Test Note: Periodically review the good ideas queue, priorities often shift Starting with the top idea, apply the following process

30 Valve 2 Decision Process Flow Begin here Are all needed resources available? Yes Could they No Start new accelerate an project active project? No Yes Idea remains in queue Yes Apply resources to active project

31 Phase Gate Review Process Establish a standard set of phases that fit your types of projects Maintain a virtual status board showing where each project is in the pipeline Conduct reviews of every project in the pipeline as it passes a phase milestone (next slide) Allow for emergency review any time a project in jeopardy

32 Phase Review Questions Is this project still worth doing? Market/customer needs Opportunity cost Are the planned resource requirements on track? Are there any anticipated obstacles to completion?

33 Pipeline Control Room Pipeline Metrics Development Project Status Improvement Project Status

34 Metrics Wall- Example % Of Sales from New Products New Product Revenue Growth 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1st Qtr 2nd Qtr 3rd Qtr 4th Qtr >12 months 6-12 months < 6 months P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 Product D Product C Product B Product A Aggregate Schedule Performance 80 Normalized Development Cycle Time (Days/1000 gates) P1 P3 P5 P7 P9 P11 % of Planned Milestones Met Goal Q1 Q2 Q3 Q4 Actual Goal

35 Development Project Status Wall Project C Idea Queue Input Queue Project Priority 1. A 2. C 3. E 4. B 5. D Project Schedule B Performance Project Schedule A Resource Performance Utilization Schedule Resource Performance By Utilization Name Resource By Utilization Hours/month, Name By Hours/month, Name planned vs. actual Development Hours/month, planned vs. Spending actual Development planned vs. Forecast Revenue Spending actual Plan Development Forecast Revenue Spending Plan Forecast Revenue Plan

36 Pipeline Improvement Considerations How do we identify potential improvements in our pipeline design? How do we plan and implement desired improvements? How do we allocate resources to improvement work?

37 Pipeline Improvement Wall Improvement Project C Idea Queue Input Queue Active Improvement Project Priority 1. A 2. C 3. B Improvement Schedule Performance Project B Improvement Schedule Resource Performance Project A Utilization Charter/Goals Resource By Utilization Name Schedule By Performance Hours/month, Name Resource Hours/month, Utilization planned vs. actual Development By planned Name vs. Spending actual Development Forecast Hours/month, Revenue Spending Plan Forecast planned Revenue vs. actual Plan

38 In Summary Teams that manage a portfolio of projects often don t view themselves as managing a production pipeline We have a natural tendency to bite off more than we can chew which has unintended consequences to productivity Controlling the front end of the process generally results in a faster flow of completed work at the back end Maintaining a clear and common set of priorities for project queues and active projects can help ensure that at least the high priority work gets through