A Work Breakdown Structure (WBS) is a tool to convert requirements from free text, or from a Requirements Breakdown Structure (RBS), into the type and quantity of materials, technology, and/or resources (labor) required to successfully execute the work. It is usually focused on the labor aspect of a requirement and should always include both the labor category and the quantity of labor required. It should never include the names of personnel as that can, and usually does, unduly influence the WBS in a way that disconnects the client’s requirements from your solution. Assigning named resources occurs in a later stage.
Perhaps the best way to explain this is to take an example and work it from start to finish. Here is our simplistic Statement of Work (SOW) and how a WBS helps us solve the problem within the ‘four corners of the contract.’ If we go outside of those four corners and over engineer the solution we negatively impact our price and lower our Probability of Win (pWin).
1.0 Help Desk Operations
1.1 The contractor shall maintain a help desk between the hours of 6 am and 8 pm Monday through Friday and between 8 am and 12 pm on Saturday.
1.2 The average time to triage a help desk ticket is 20 minutes. The average number of tickets per weekday is 300 and on the weekend it is 50.
1.3 On the first workday of each month, due to heavy user traffic, the average number of tickets spikes to
twice the amount stated in 1.2
1.4 The contractor shall maintain an average first response speed of 5 minutes and an average time to close a ticket of 24 hours (not business hours).
2.0 Management
2.1 The contractor shall provide at least one supervisor during times the help desk is operational
2.2 The contractor shall provide monthly reports by the 10th of each month showing Help Desk ticket statistics to include first response speed, average response speed, and any other metrics that are available for which the Government requests.
Note: for purposes of this example we will staff it for just one month which is 4 weeks (starting on a Monday).
Usually a SOW goes on for many pages with additional requirements and extends for a period beyond just 30 days, but for this example we will keep it simple. Looking through the requirements a few things become apparent:
- we need resources (people) to staff the help desk
- we need different types of resources: help desk staff, supervisory staff, and someone to prepare the monthly report
- we need to determine the number of hours by day we will need each type of resource
Aside
Before we solve this specific problem let us explore more about the solutioning part of this. If this was a real solicitation we would need to strategize on a few things such as:
- Would a Knowledge Base (KB) reduce the number of tickets?
- Would an internal knowledge management tool improve the help desk response time over the historical numbers provided?
- Would increased training of the help desk staff improve triage time?
- Is there potential misinformation in the provided metrics? Does the incumbent know those times to be less and thus can propose fewer resources?
- Should we have multiple people with Help Desk knowledge prepare a WBS and we compare (Delphi technique)?
- Do we, or a possible teammate, have experience in help desk operations that would allow us to solution engineer this into an efficient staffing model with either tools or techniques that we propose in our technical response?
- What approaches do we think our competition may take to solve this problem?
- How sensitive to price is the client? -> This may seem like an odd question to ask during the WBS process but our solution, and our competition’s, depends on that answer.
Those are just a few example of the questions we should be asking, but the bottom line is, what is our strategy and resulting approach to solving the Government’s requirements that provides the most value?
So again, for simplicity sake let us set aside, the questions and considerations raised above and focus on what a completed WBS looks like.
In our solutioning we determine that we are going to need just two types of resources which we are going to start referring to as Labor Categories (LCATs):
- Help Desk staff
- Supervisory staff
- No other resources are needed as the reports will be generator by a supervisor
Now that we have the LCATs we can place them along the top of our WBS matrix. We can also begin filling out our WBS Levels on the left. Depending on how complicated the requirement document is you may need 1 to n WBS levels (we use two below). Note that it is easiest to design this in Microsoft Excel.
| WBS Level 1 | WBS Level 2 | Times | Help Desk Staff | Supervisor |
| 1.0 Help Desk Ops | 1.1 Weekday | 6am-8pm | ||
| 1.0 Help Desk Ops | 1.1 Weekend (Sat) | 8am-12pm | ||
| 1.0 Help Desk Ops | 1.3 Surge | 1 day/month | ||
| 2.0 Management | 2.1 Supervisor | All active times | ||
| Total Hours |
At this point we haven’t put in any hours, we just translated text requirements into a tabular format that will later help us solve the math problem. Sometimes a business development, capture, or project manager person with in-depth knowledge will say “I already know the client is looking for X number of people, we should bid that.” That is a good way to throw away time and valuable Bid and Proposal (B&P) dollars. If it is worth time and money to write a proposal, then its worth a little extra time to do a proper WBS to compare to what the intel says.
Next let’s start entering in the quantitative information.
| WBS Level 1 | WBS Level 2 | Times | Help Desk Staff | Supervisor |
| 1.0 Help Desk Ops | 1.1 Weekday | 6am-8pm | 20m/tix*300 tix =6k min / 60m/hr = 100 hrs/day * 5 days/wk * 4 wks = 2k hrs | 14hrs/day * 5 days/wk * 4 wks=280hrs |
| 1.0 Help Desk Ops | 1.1 Weekend (Sat) | 8am-12pm | 20m/tix * 10 tix = 200 min / 60 = 3.3hrs but we need a minimum of 4 hours X 4 weeks = 16 hrs | 4 hrs/sat * 4 wks = 16 hours |
| 1.0 Help Desk Ops | 1.3 Surge | 1 day/month | 20m/tix*300 tix = 6k min / 60m/hr = 100 hrs. This is the doubled part | 0 hrs as one is already on site in 1.1 Weekeday |
| 2.0 Management | 2.1 Supervisor | All active times | Not applicable | see above |
| Total Hours | 2k + 16 + 100 = 2,116 hrs | 280 + 16 = 296 |
We now know how many hours we need for each of the two LCATs for the single month required.
Full Time Equivalent (FTE)
Sometimes the amount of resources required is referred to as Full Time Equivalents (FTEs). A FTE is normally competed as the total hours / a standard workday hours. In this case lets us assume the typical employee workday is 8 hours. Therefore an FTE is 8 hrs/day * 5 days/wk * 4 weeks = 160+ (4hrs/Sat * 4 weeks) = 176hrs. We know that a supervisor is the same 176 hours from the table above and because a supervisor is required on-site for the same time period as help desk staff (requirement 2.1). Therefore the FTE calculation is:
| Item | Help Desk Staff | Supervisor |
| Total Hours | 2,116 | 296 |
| Standard Hours | 176 | 176 |
| FTES | 12.02 | 1.68 |
So generally we can see that we need at least 2 supervisors and a little more than 12 help desk staff or 14 total FTEs. Now we know generally the size of this compared say other help desk work in terms of FTEs. If we know that a Help Desk person is usually priced (not costed) at $50/hr and a supervisor is $55/hr we can now do a quick Rough Order of Magnitude (ROM) price of $122,080 ($105,800 + $16,280). Now we can compare the ROM price to what the client intel says their budget is…all before we have written any part of our technical. If our ROM is say 20% higher then the customer’s budget or where we think the competition will come in at then we need to go back to the drawing board and think about ways we can introduce efficiencies to our process. This is exactly why an early and comprehensive WBS is so important. Additionally, if we did a Delphi method and/or an estimated competitor WBS on their solution we would have even more data points to evaluate again, before we started writing any of our technical approach.
Next up we will look into the GovPTW GAO Protest based PTW product before we bring everything together in the final article.