Estimating with SNAP: IFPUG has released a new counting approach called SNAP that is designed to complement traditional function point based sizing. Unfortunately, work by both Level 4 and Capers Jones indicates that the size of work represented by a SNAP point is significantly different from the size of work represented by a Function Point. While calibration is on-going, our early indicators are that for estimation and productivity purposes, one SNAP point is equal to roughly 0.18 Function Points. ExcelerPlan will handle this adjustment for you automatically when estimating effort for projects using SNAP.
|
|
|
William Roetzheim will be delivering the closing keynote speech at the International Conference on Software Quality and Test Management on October 1st. The topic will be “Seeking Best Practice Universal Truths.”
Level 4 was awarded contracts to do ExcelerPlan based estimation work in support of acquisitions for the States of Washington, California, and Florida.
|
|
|
All times are Pacific Time!
Free WebEx
Estimating with ExcelerPlan
9/4, 11 AM-12:30 PDT
9/18, 11 AM-12:30 PDT
10/2, 11 AM-12:30 PDT
To register for demos, email:
Edward@portal.level4ventures.com |
|
|
Free ExcelerPlan License with Training Purchase
Through September 30th sign up for our 12 module self-paced video training program for $1,995 and receive one free license to ExcelerPlan with six-months of maintenance, support, and license activation. You’ll also receive 24 Professional Development Units (PDU) of training credits (Certificate requires verification of attendance plus passing final exam.)
To register for training and your free six-month license, email:
Edward@portal.level4ventures.com |
|
|
I often hear the argument that “we don’t have a budget to purchase ExcelerPlan right now.” If this is your situation, there may still be options. Here are some ideas that have worked for other organizations:
1. Purchase training and receive a free license, as described above.
2. Use project funds to have Level 4 Ventures provide an independent cost estimate for one of your major initiatives. We’ll develop a comprehensive and accurate estimate using ExcelerPlan, and include an ExcelerPlan license with the delivered report so that you can access and maintain the estimate yourself.
3. Use executive discretionary funds to do a small proof-of-concept engagement now, with a larger initiative possible in a future budget period if the proof-of-concept is successful. We can provide some free brainstorming help to define goals and objectives for a small effort such as this.
Edward
Director of Sales and Marketing
|
|
|
| Free ExcelerPlan with Training Purchase |
|
| Through September 30th you can obtain a free ExcelerPlan license with the purchase of our self-paced estimation training. See column above for details. |
|
|
|
|
|
|
 Internal versus External Benchmarking: Because estimation tools and supporting data rely heavily on historic data, I’m often asked about using these tools for benchmarking. Benchmarking is one of those management buzzwords that is often a good idea and often a terrible idea, it all depends on the purpose and the execution. I find that it’s useful to break benchmarking discussions down into external versus internal benchmarking. For external benchmarking, you’re comparing your organization’s performance with similar organizations. The challenge is making sure that you’re really comparing “like versus like.” I would argue that this is difficult to impossible using unadjusted data. The only way to effectively benchmark against external data is to use normalization factors to adjust the data so that the differences between organizations are including in the compensation equations. And even then, I find that the primary uses for external benchmarking are to fill in gaps in your own metric collection or to identify potential shifts in staff allocation.
With internal benchmarking, on the other hand, you’re comparing your organization with itself over time, looking for trends. Here it’s easy to receive relatively consistent data and you don’t need to adjust for organization specific variables. They’re all baked into the data. What you do need is to find a suitable denominator for the ratio analysis that you’ll want to do, a topic that I’ll cover next month.
william@portal.level4ventures.com
|
|
 |
|
 Estimation Process (part 3), Follow-Up: With the estimation interview behind you, it’s time to roll up your sleeves and get to work. During the interview you’ll normally identify specific quantitative data that is available but that has not yet been provided to you. What you’re looking for is generally lists or counts of things (reports, pages, interfaces, and so on). There may also be some unanswered questions where you’ll need a follow-up interview specifically for that topic. For example, there might be a business requirement to “add a new product type,” but until you talk to someone on the technical side, you’re not sure if that’s a simple configuration change or a major system modification with significant ripple effects. So you’ll find the right person on the technical side, and ask them what’s involved in adding a new product type.
You’ll then prepare and document your estimate, then do internal validation. This might be with other members of the estimation team or with someone else knowledgeable about the project. When it’s just me working on a complex estimate, I’ll often set it aside and then come back and attack the estimate again using a completely different approach to see if the results converge. For example, I might do the initial estimate using Function Point Equivalents, then come back and try analogy with adjustments.
Next month I’ll talk about the contents of the estimation report. |
|
|
|
|
Table 1: Software Application Costs by Size
|
| |
|
|
|
Size in
|
Application Cost
|
Application Cost
|
|
Function
|
at $10,000 per
|
per Function
|
|
Points
|
Staff Month
|
Point
|
| |
|
|
|
1
|
$307
|
$307
|
|
10
|
$5,900
|
$590
|
|
100
|
$68,750
|
$688
|
|
1,000
|
$837,008
|
$837
|
|
10,000
|
$27,002,794
|
$2,700
|
|
100,000
|
$383,661,586
|
$3,837
|
| |
|
|
|
Average
|
$68,596,057
|
$1,493
|
[This guest column reprinted with permission from: “RETURN ON INVESTMENT (ROI) IN SOFTWARE PROJECT MANAGEMENT TOOLS AND SOFTWARE QUALITY CONTROL,” Capers Jones, July 2014. Full article available on http://Namcookanalytics.com]
It is obvious from table 1 that the value of quality goes up with overall application. For small projects below 100 function points both marginal quality and marginal management are survivable. Above 1,000 function points excellence in quality and excellence in project management are mandatory for success, and also to ensure that the projects are actually completed and not terminated due to massive cost and schedule overruns. Although software project failures outnumber successes there are successful large system built. From assessment and benchmark studies, two critical factors differentiate successful software applications from failures: 1) project management; 2) quality control. To understand the significance of table 2 (below) it is necessary to know why software projects run late or are canceled. Large systems usually run late because they contain so many defects or errors that they don’t work, so testing stretches out towards infinity. Unfortunately, if this situation is not detected until testing begins it is too late for corrective actions to be fully successful.
Capers Jones
Table 2: Differences between Successful and Unsuccessful Software Projects In the 10,000 Function Point Size Category
|
| |
| |
|
| Successful Projects |
Unsuccessful Projects |
| Software cost estimating tools |
Manual estimating methods |
| Project planning tools |
No project planning tools |
| Quality estimating tools |
No quality estimates performed |
| Project management tools |
Partial use of project management tools |
| Project cost tracking tools |
Partial use of cost tracking tools |
| Project progress tracking tools |
No progress tracking tools |
| Accurate progress reports |
Progress reports that conceal problems |
| Formal project milestone tracking |
Informal milestone tracking |
| Defect prevention such as JAD |
No defect prevention |
| Effective requirements gathering |
Ineffective requirements gathering |
| Formal requirements inspections |
No requirements inspections |
| Formal architecture inspections |
No architecture inspections |
| Formal design inspections |
No design inspections utilized |
| Formal code inspections |
No code inspections utilized |
| FOG readability scores |
No readability scores |
| Prototypes of key features |
No prototypes |
| Pre-test static analysis |
No pre-test static analysis |
| Formal testing |
Informal testing |
| Certified test personnel |
No certified test personnel |
|
|
|
|
 Dear Tabby:
I’ll admit that I’m a physical guy, but my girl friend is driving me nuts with her head in the cloud. She says I’m the problem and I’m too expensive, but I say that only a real physical guy can deliver the performance she wants. Whose’ right?
signed, Physical Pete
Dear Physical Pete:
Sorry, but I’m going to come down on the side of your girlfriend in this case. In the old days you were correct that performance suffered unless the relationship was physical. But the performance hit with today’s virtual environments is under 5%, and that can be easily overcome with hardware allocations. Dedicated virtualization will save you about 10% overall, but virtual-shared and cloud will save you roughly 60% on overall hardware costs. Engineering labor costs are also lower, although the drops are not as dramatic.
signed, Tabby
|
|
|
Hardware Environmental Factors: The project specific environment adjustments in ExcelerPlan will vary for the different ExcelerPlan configurations and installations, but all configurations that support infrastructure estimation will have four key environmental settings that are used as primary adjustment factors for estimating the cost of infrastructure hardware and software. These are;
Tier Level: This setting defines the mission criticality of the application ranging from core down to administrative support. Tier Level is important because it drives the Recovery Time Objective (RTO) and Recovery Point Objective (RPO) plus the organizational/mission risk profile. These in turn determine the required hardware and software approach for things like redundancy and mirroring.
Hardware Loading: This setting specifies the volume and performance requirements for user interaction with the system. It’s primarily based on user driven peak transaction volume.
Database Loading: This setting specifies the “horsepower’ needed to support the database layer, so it’s a combination of the volume of queries, the complexity of those queries, and the size of the data that must then be delivered in response. It is not directly or necessarily related to the size of the database, as data storage is relatively cheap and it is no longer a major driver of infrastructure costs.
Infrastructure Integration: This setting specifies the infrastructure workload to support queries from other systems. An example might be balance inquiries from an ATM network. |
|
|
|
|
|
