Building the IT Value Stream

I am struck by a few facts that surround the IT Value Stream. First, a recent survey by Planview? indicated that about 8% of the PMO portfolio is delivered on time. Second, that 30%-40% of the release train for every project is spent in testing/QA and release.

While both of these facts may sound negative my perspective is that we have a baseline. If your organizational baseline of portfolio metrics is similar or close, then great. The reason this is great is it means you have the wherewithal to assemble facts, data, and reports or you possess a value stream management suite like Planview? has built allowing your organization to develop a forecast for portfolio management.

Why is a forecast for the portfolio important, you ask? The annual struggle to predict a budget against an ever-changing product offering challenges even the best minds to submit a budget that is reasonably accurate. This is frustrating for the CIO, CFO and CEO especially when business customers and departments demand immediate changes (ad hoc) adding cost.

I don’t have to tell anyone reading this article that human resources are technology’s greatest constraint. So how do we succeed at building a forecast that is reasonable and flexible at the same time? We start with the knowns.

1.???At any given moment we know how many employees and contingent workers are on the payroll that perform portfolio services.

2.???We can express these workers as a two items: actual cost and hours at work:

o??A worker cost us the sum total of all salary and benefits.

o??A worker “at work” can be expressed as the sum total of weekly and monthly hours available to work (typically 83% x 160 hours/month equal to 133 hours/month.

o??Identify each type of worker necessary for I&O or DEVOPs project depending on you organization’s methodology and express them as the cost of the specific type of worker.

3.???We can develop a forecast using the worker data type cost from 1 & 2, and factor the known portfolio expressed as estimated hours to completion of each of the projects.

Here’s an example:

SCOPE: A 90-day Agile SCRUM project to replace a MS ACCESS DB with a .NET/Java DB to provide a Risk evaluation tool:

???????????Resource??????????????????????????????Monthly Cost ?x 3 Months = Worker Type Cost

1.???Product Manager????????$9,800 x 3 = $29,400

2.???SCRUM Manager ???????$10,000 x 3 = $30,000

3.???Business Manager ?????? $9,800 x 3 = $29,400

4.???Tech Dev Lead??????????????$8,300 x 3 = $24,900

5.???2 Developers?????????????????$7,400 x 3 = $22,200

6.???QA/Release Analyst???? $5,400 x 3 = $16,200

_______________________________________________________

Total Resource Cost ????$50,700 x 3 = $152,100

Benefits Factor 22%????$11,154 x 3 = $33,462

+ Project Contingency ?$6,185 x 3 = $18,555

_______________________________________________________

_______________________________________________________

Forecast MS ACCESS DB Project = $204,117

Great, now we have a forecast of $204,117 for a simple portfolio project but what about the 90-day schedule? All good projects include slippage. How you handle slippage will be organizationally unique, but for this example let’s assume we program the start and finish of the MS ACCESS DB Project to work only on workdays with no weekends, no holidays, and no more than two 3-hour test windows on Tuesdays and Thursdays with compensatory time off in exchange. ?This will provide a reasonable work life balance for the team to avoid burnout.

I don’t want to oversimplify the 10,000 foot level view for portfolio management but aforementioned forecasting process can produce daily capacity or scale up to the portfolio level. In project management we produce the WBS, or in other instances I have produced a project forecast for 20,000 hours using this simple approach. To give credit where credit is due, USAA was the first PMO I worked with that used an hourly wage per type of human resource to forecast and re-forecast.

All good budget forecasts should have 10-20% contingency built in. The flexibility is that an IT organization should be able to meet monthly based on the CPI and SPI for each project and re-forecast the portfolio. This is made easy using ADOBE Workfront?, eHPPM?, or Planview? as examples or can be hand done if necessary. The Portfolio Manager should be able to shift dollars from and to projects and re-purpose unused project budgets.

So, the moral of this article is capacity planning to develop a forecast for portfolio management! If you need more information on how to implement this capability call on me at (210) 618-6566M or as always provide me feedback in the comments.

?#CIO, #CEO #CFO #PMO #Valuestream