Function point metrics are the most accurate and effective metrics yet developed for software sizing and also for studying software productivity, quality, costs, risks, and economic value.

Unlike the older “lines of code” metric function points can be used to study requirements, design, and in fact all software activities from development through maintenance.

In the future function point metrics can easily become a universal metric used for all software applications and for all software contracts in all countries.  The government of Brazil already requires function points for all software contracts, and South Korea and Italy may soon follow.

However, there are some logistical problems with function point metrics that need to be understood and overcome in order for function point metrics to become the primary metric for software economic analysis.

Manual function point counting is too slow and costly to be used on large software projects above 10,000 function points in size.  Also, application size is not constant but grows at about 2% per calendar month during development and 8% or more per calendar year for as long as software is in active use.

This paper discusses a method of high-speed function point counting that can size any application in less than two minutes, and which can predict application growth during development and for five years after release.  This new method is based on pattern matching and is covered by U.S. utility patent application and hence is patent pending.

Introduction
Function point metrics were invented by A.J. Albrecht and colleagues at IBM’s White Plains development center circa 1975.  Function point metrics were placed in the public domain by IBM in 1978.  Responsibility for function point counting rules soon transferred to the International Function Point User’s Group (IFPUG).  Their web site is www.IFPUG.org.

Function point metrics were developed by IBM due to serious mathematical and economic problems associated with the older “lines of code” metric or LOC.  The LOC metric penalizes high-level programming languages and also cannot be used to evaluate requirements, design, business analysis, user documentation, or any other non-coding activities.

In the current era circa 2013 function point metrics are the major metric for software economic and productivity studies.  At least 50,000 software projects have been measured using IFPUG function point metrics, including more than 5,000 projects that are publically available from the International Software Benchmark Standards Group (ISBSG).  Their web site is www.ISBSG.org.

The Strengths of Function Point Metrics

1. IFPUG function point metrics have more measured projects than all other metrics combined.
2. IFPUG function point metrics are endorsed by ISO/IEC standard 20926:2009.
3. Formal training and certification examinations are available for IFPUG function point counting.
4. Hundreds of certified IFPUG function point counters are available in most countries.
5. Counts of function points by certified counters usually are within 5% of each other.
6. IFPUG function point metrics are standard features of most parametric estimating tools such as KnowledgePlan, SEER, and Software Risk Master.
7. Function points are increasingly used for software contracts.  The government of Brazil requires function points for all software contracts.

The Weaknesses of Function Point Metrics

1. Function point analysis is slow.  Counting speeds for function points average perhaps 500 function points per day.
2. Due to the slow speed of function point analysis, function points are almost never used on large systems > 10,000 function points in size.
3. Function point analysis is expensive.  Assuming a daily counting speed of 500 function points and a daily consulting fee of $1,500 counting an application of 10,000 function points would require 20 days and cost $30,000.  This is equal to a cost of $3.00 for every function point counted.
4. Application size is not constant.  During development applications grow at perhaps 2% per calendar month.  After development applications continue to grow at perhaps 8% per calendar year.  Current counting rules do not include continuous growth.
5. More than a dozen function point counting variations exist circa 2013 including COSMIC function points, NESMA function points, FISMA function points, fast function points, backfired function points, and a number of others.  These variations produce function point totals that differ from IFPUG function points by perhaps + or – 15%.

A Patented Method for High-Speed Function Point Analysis

In order to make function point metrics easier to use and more rapid, the author filed a U.S. utility patent application in 2012.  The utility patent application is U.S. Patent Application No. 13/352,434 filed January 18, 2012.

The high-speed sizing method is embedded in the Software Risk Master ™ (SRM) sizing and estimating tool under development by Namcook Analytics LLC.  A working version is available on the Namcook Analytics web site, www.Namcook.com.  The version requires a password from within the site.

The Namcook Analytics high-speed method includes these features:

1. From more than 200 trials sizing speed averages about 1.87 minutes per application.  This speed is more or less constant between applications as small as 10 function points or as large as 300,000 function points.
2. The sizing method often comes within 5% of manual counts by certified counters.  The closest match was an SRM predicted size of 1,802 function points for an application sized manually at 1,800 function points.
3. The sizing method can also be used prior to full requirements, which is the earliest of any known software sizing method.
4. The patent-pending method is based on external pattern matching rather than internal attributes.  So long as an application can be placed on the SRM taxonomy the application can be sized.
5. The method can size all types of software including operating systems, ERP packages, telephone switching systems, medical device software, web applications, smart-phone applications, and normal information systems applications.
6. The sizing method is metric neutral and predicts application size in a total of 15 metrics including IFPUG function points, the new SNAP metric for non-functional attributes; COSMIC function points, story points, use-case points, logical code statements, and many others.
7. The sizing method predicts application growth during development and for five years of post-release usage.

A Short Summary of Pattern Matching
Today in 2013 very few applications are truly new.  Most are replacements for older legacy applications or enhancements to older legacy applications.  Pattern matching uses the size, cost, schedules, and other factors from legacy applications to generate similar values for new applications.

Software pattern matching as described here is based on a proprietary taxonomy developed by the author, Capers Jones.  The taxonomy uses multiple-choice questions to identify the key attributes of software projects.  The taxonomy is used to collect historical benchmark data and also as basis for estimating future projects.  The taxonomy is also used for sizing applications.

For sizing, the taxonomy includes project nature, scope, class, type, problem complexity, code complexity, and data complexity. For estimating, additional parameters such as CMMI level, methodology, and team experience are also used.

The proprietary Namcook taxonomy used for pattern matching contains 122 factors.  With 122 total elements the permutations of the full taxonomy total to 214,200,000 possible patterns.  Needless to say more than half of these patterns have never occurred and will never occur.

For the software industry in 2013 the total number of patterns that occur with relatively high frequency is much smaller:  about 20,000.

The Software Risk Master tool uses the taxonomy to select similar projects from its knowledge base of around 15,000 projects.  Mathematical algorithms are used to derive results for patterns that do not have a perfect match.

However a great majority of software projects do have matches because they have been done many times.  For example all banks perform similar transactions for customers and therefore have similar software packages.  Telephone switches also have been done many times and all have similar features.

Pattern matching with a good taxonomy to guide the search is a very cost-effective way for dealing with application size.

Pattern matching is new for software sizing but common elsewhere.  Two examples of pattern matching are the Zillow data base of real-estate costs and the Kelley Blue Book of used automobile costs.  Both use taxonomies to narrow down choices, and then show clients the end results of those choices.

Increasing Executive Awareness of Function Points for Economic Studies
Because of the slow speed of function point analysis and the lack of data from large applications function points are a niche metric below the interest level of most CEO’s and especially CEO’s of Fortune 500 companies with large portfolios and many large systems including ERP packages.

In order for function point metrics to become a priority for C level executives and a standard method for all software contracts, some improvements are needed:

1. Function point size must be available in a few minutes for large systems; not after weeks of counting.
2. The cost per function point counted must be lower than $0.05 per function point rather than today’s costs of more than $3.00 per function point counted.
3. Function point metrics must be able to size applications ranging from a low of 1 function point to a high of more than 300,000 function points.
4. Sizing of applications must also deal with the measured rates of requirements creep during development and the measured rates of post-release growth for perhaps 10 years after the initial release.
5. Function points must also be applied to maintenance, enhancements, and total costs of ownership (TCO).
6. Individual changes in requirements should be sized in real-time as they occur.  If a client wants a new feature that may be 10 function points in size, this fact should be established within a few minutes.
7. Function points should be used for large-scale economic analysis of methods, industries, and even countries.

Examples of Software Risk Master ™ (SRM) Sizing and Estimating
Following are some examples of the features of the patent-pending sizing method embedded in Software Risk Master ™:

Sizing Application Growth during Development and After Release

Table 1 shows an example of the way SRM predicts requirements growth and post-release changes:

	Table 1: Software Risk Master™ Multi-Year Sizing
	Copyright © 2011-2013 by Capers Jones.
	Patent application 13/352,434.
	Nominal application size
	in IFPUG function points	10,000
		Function
		Points
1	Size at end of requirements	10,000
2	Size of requirement creep	2,000
3	Size of planned delivery	12,000
4	Size of deferred functions	-4,800
5	Size of actual delivery	7,200
6	Year 1	12,000
7	Year 2	13,000
8	Year 3	14,000
9	Year 4*	17,000
10	Year 5	18,000
11	Year 6	19,000
12	Year 7	20,000
13	Year 8*	23,000
14	Year 9	24,000
15	Year 10	25,000

Note that year’s 4 and 7 show a phenomenon called “mid-life kickers” or major new features added about every four years to commercial software applications.

Multi-year sizing is based on empirical data from a number of major companies such as IBM where applications have been in service for more than 10 years.

Predicting Application Size in Multiple Metrics
There are so many metrics in use in 2013 that as a professional courtesy to users and other metrics groups SRM predicts size in the metrics shown in Table 2.  Assume that the application being sized is known to be 10,000 function points using IFPUG version 4.2 counting rules:

Table 2:  Metrics Supported by SRM Pattern Matching

Alternate Metrics	Size	% of IFPUG
Backfired function points	10,000	100.00%
Cosmic function points	11,429	114.29%
Fast function points	9,700	97.00%
Feature points	10,000	100.00%
FISMA function points	10,200	102.00%
Full function points	11,700	117.00%
Function points light	9,650	96.50%
Mark II function points	10,600	106.00%
NESMA function points	10,400	104.00%
RICE objects	47,143	471.43%
SCCQI “function points”	30,286	302.86%
SNAP non functional metrics	1,818	18.18%
Story points	5,556	55.56%
Unadjusted function points	8,900	89.00%
Use case points	3,333	33.33%

 

Because SRM is metric neutral, additional metrics could be added to the list of supported metrics if new metrics become available in the future.

SRM also predicts application size in terms of logical code statements or “LOC.”  However with more than 2,500 programming languages in existence and the majority of projects using several languages, code sizing requires that users inform the SRM tool as to which language(s) will be used.  This is done by specifying a percentage of various languages from an SRM pull-down menu that lists the languages supported.  Currently SRM supports about 180 languages for sizing, but this is just an arbitrary number that can easily be expanded.

Sizing All Known Types of Software Application
One of the advantages of sizing by means of external pattern matching rather than sizing by internal attributes is that the any known application can be sized.  Table 3 shows 40 samples of applications size by the Software Risk Master ™ patent-pending method:

Table 3:  Examples of Software Size via Pattern Matching

                Using Software Risk Master ™

Application Size in IFPUG Function Points

1. Oracle 229,434
2. Windows 7 (all features) 202,150
3. Microsoft Windows XP   66,238
4. Google docs   47,668
5. Microsoft Office 2003   33,736
6. F15 avionics/weapons   23,109
7. VA medical records   19,819
8. Apple I Phone   19,366
9. IBM IMS data base   18,558
10. Google search engine   18,640
11. Linux   17,505
12. ITT System 12 switching   17,002
13. Denver Airport luggage (original)   16,661
14. Child Support Payments (state)   12,546
15. Facebook     8,404
16. MapQuest     3,793
17. Microsoft Project     1,963
18. Android OS (original version)     1,858
19. Microsoft Excel     1,578
20. Garmin GPS navigation (hand held)     1,518
21. Microsoft Word     1,431
22. Mozilla Firefox     1,342
23. Laser printer driver (HP)     1,248
24. Sun Java compiler     1,185
25. Wikipedia     1,142
26. Cochlear implant (embedded)     1,041
27. Microsoft DOS circa 1998     1,022
28. Nintendo Gameboy DS     1,002
29. Casio atomic watch        933
30. Computer BIOS        857
31. SPR KnowledgePlan        883
32. Function Point Workbench                    714
33. Norton anti-virus        700
34. SPR SPQR/20        699
35. Golf handicap analysis        662
36. Google Gmail        590
37. Twitter (original circa 2009)        541
38. Freecell computer solitaire        102
39. Software Risk Master™ prototype          38
40. ILOVEYOU computer worm          22

This list of 40 applications was sized by the author in about 75 minutes, which is a rate of 1.875 minutes per application sized.  The cost per function point sized is less than $0.001.  As of 2013 SRM sizing is the fastest and least expensive method of sizing yet developed.  This makes SRM useful for Agile projects where normal function point analysis is seldom used.

Function Points for Early Analysis of Software Risks
Software projects are susceptible to more than 200 risks in all, of which about 50 can be analyzed using function point metrics.  As application size goes up when measured with function point metrics, software risks also go up.

Table 4 shows the comparative risk profiles of four sample projects of 100, 1000, 10,000, and 100,000 function points.  All four are “average” projects using iterative development.  All four are assumed to be at CMMI level 1.

Table 4: Average Risks for IT Projects by Size
(Predictions by Software Risk Master ™)
Risks for 100 function points
Cancellation	8.36%
Negative ROI	10.59%
Cost overrun	9.19%
Schedule slip	11.14%
Unhappy customers	36.00%
Litigation	3.68%
Average Risks	13.16%
Financial Risk	15.43%
Risks for 1000 function points
Cancellation	13.78%
Negative ROI	17.46%
Cost overrun	15.16%
Schedule slip	18.38%
Unhappy customers	36.00%
Litigation	6.06%
Average Risks	17.81%
Financial Risk	25.44%
Risks for 10,000 function points
Cancellation	26.03%
Negative ROI	32.97%
Cost overrun	28.63%
Schedule slip	34.70%
Unhappy customers	36.00%
Litigation	11.45%
Average Risks	28.29%
Financial Risk	48.04%
Risks for 100,000 function points
Cancellation	53.76%
Negative ROI	68.09%
Cost overrun	59.13%
Schedule slip	71.68%
Unhappy customers	36.00%
Litigation	23.65%
Average Risks	52.05%
Financial Risk	99.24%

All of the data in Table 4 are standard risk predictions from Software Risk Master ™.  Risks would go down with higher CMMI levels, more experienced teams, and robust methodologies such a RUP or TSP.

Small projects below 1000 function points are usually completed without too much difficulty.  But large systems above 10,000 function points are among the most hazardous of all manufactured objects in human history.

It is an interesting phenomenon that every software breach of contract lawsuit except one where the author worked as an expert witness were for projects of 10,000 function points and higher.

Activity-Based Sizing and Cost Estimating
In order to be useful for software economic analysis, function point metrics need to be applied to individual software development activities.  Corporate executives at the CEO level want to know all cost elements, and not just “design, code, and unit test” or DCUT as it is commonly called.

SRM has a variable focus that allows it to show data ranging from full projects to 40 activities.
Table 5 shows the complete set of 40 activities for an application of 10,000 function points in size:

Table 5:  Function Points for Activity-Based Cost Analysis

	Development Activities	Work	Burdened
		Hours per	Cost per
		Funct. Pt.	Funct. Pt.
1	Business analysis	0.02	$1.33
2	Risk analysis/sizing	0.00	$0.29
3	Risk solution planning	0.01	$0.67
4	Requirements	0.38	$28.57
5	Requirement. Inspection	0.22	$16.67
6	Prototyping	0.33	$25.00
7	Architecture	0.05	$4.00
8	Architecture. Inspection	0.04	$3.33
9	Project plans/estimates	0.03	$2.00
10	Initial Design	0.75	$57.14
11	Detail Design	0.75	$57.14
12	Design inspections	0.53	$40.00
13	Coding	4.00	$303.03
14	Code inspections	3.30	$250.00
15	Reuse acquisition	0.01	$1.00
16	Static analysis	0.02	$1.33
17	COTS Package purchase	0.01	$1.00
18	Open-source acquisition.	0.01	$1.00
19	Code security audit.	0.04	$2.86
20	Ind. Verif. & Valid.	0.07	$5.00
21	Configuration control.	0.04	$2.86
22	Integration	0.04	$2.86
23	User documentation	0.29	$22.22
24	Unit testing	0.88	$66.67
25	Function testing	0.75	$57.14
26	Regression testing	0.53	$40.00
27	Integration testing	0.44	$33.33
28	Performance testing	0.33	$25.00
29	Security testing	0.26	$20.00
30	Usability testing	0.22	$16.67
31	System testing	0.88	$66.67
32	Cloud testing	0.13	$10.00
33	Field (Beta) testing	0.18	$13.33
34	Acceptance testing	0.05	$4.00
35	Independent testing	0.07	$5.00
36	Quality assurance	0.18	$13.33
37	Installation/training	0.04	$2.86
38	Project measurement	0.01	$1.00
39	Project office	0.18	$13.33
40	Project management	4.40	$333.33
	Cumulative Results	20.44	$1,548.68

 

SRM uses this level of detail for collecting benchmark data from large applications.  In predictive mode prior to requirements this much detail is not needed, so a smaller chart of accounts is used.

Function Points and Methodology Analysis
One topic of considerable interest to both C level executives and also to academics and software engineers is how various methodologies compare.  Software Risk Master ™ includes empirical results from more than 30 different software development methodologies; more than any other benchmark or estimation tool.

Table 6 shows the approximate development schedules noted for 30 different software development methods.  The rankings run from slowest at the top of the table to fastest at the bottom of the table:

	Table 6:  Application Schedules
	Application Size =	10	100	1,000	10,000	100,000
	(IFPUG 4.2)
		Schedule	Schedule	Schedule	Schedule	Schedule
	Methods	Months	Months	Months	Months	Months
1	Proofs	2.60	6.76	17.58	45.71	118.85
2	DoD	2.57	6.61	16.98	43.65	112.20
3	Cowboy	2.51	6.31	15.85	39.81	100.00
4	Waterfall	2.48	6.17	15.31	38.02	94.41
5	ISO/IEC	2.47	6.11	15.10	37.33	92.26
6	Pairs	2.45	6.03	14.79	36.31	89.13
7	Prince2	2.44	5.94	14.49	35.32	86.10
8	Merise	2.44	5.94	14.49	35.32	86.10
9	DSDM	2.43	5.92	14.39	34.99	85.11
10	Models	2.43	5.89	14.29	34.67	84.14
11	Clean rm.	2.42	5.86	14.19	34.36	83.18
12	T-VEC	2.42	5.86	14.19	34.36	83.18
13	V-Model	2.42	5.83	14.09	34.04	82.22
14	Iterative	2.41	5.81	14.00	33.73	81.28
15	SSADM	2.40	5.78	13.90	33.42	80.35
16	Spiral	2.40	5.75	13.80	33.11	79.43
17	SADT	2.39	5.73	13.71	32.81	78.52
18	Jackson	2.39	5.73	13.71	32.81	78.52
19	EVO	2.39	5.70	13.61	32.51	77.62
20	IE	2.39	5.70	13.61	32.51	77.62
21	OO	2.38	5.68	13.52	32.21	76.74
22	DSD	2.38	5.65	13.43	31.92	75.86
23	RUP	2.37	5.62	13.34	31.62	74.99
24	PSP/TSP	2.36	5.56	13.11	30.90	72.86
25	FDD	2.36	5.55	13.06	30.76	72.44
26	RAD	2.35	5.52	12.97	30.48	71.61
27	Agile	2.34	5.50	12.88	30.20	70.79
28	XP	2.34	5.47	12.79	29.92	69.98
29	Hybrid	2.32	5.40	12.53	29.11	67.61
30	Mashup	2.24	5.01	11.22	25.12	56.23
	Average	2.41	5.81	14.03	33.90	81.98

Software Risk Master ™ also predicts staffing, effort in months and hours, costs, quality, and 5-years of post-release maintenance and enhancement.  Table 6 only shows schedules since that topic is of considerable interest to CEO’s as well as other C level executives.

Note that table 6 assumes close to a zero value for certified reusable components.  Software reuse can shorten schedules compared those shown in table 6.

Table 6 also assumes an average team and no use of the capability maturity model.  Expert teams and projects in organizations at CMMI levels 3 or higher will have shorter schedules than those shown in table 6.

SRM itself handles adjustments in team skills, CMMI levels, methodologies, programming languages, and volumes of reuse.

Function Points for Software Quality Analysis
Function points are the best metric for software quality analysis.  The older metric “cost per defect” penalizes quality and also violates standard economic assumptions.  Quality economics are much better analyzed using function point metrics than any other.  Table 7 shows a sample of the quality predictions from SRM for an application of 1000 function points.  The table shows a “best case” example:

Table 7: SRM Quality Estimate			Output Results
Requirements defect potential			134
Design defect potential			561
Code defect potential			887
Document defect potential			135
Total Defect Potential			1,717
Per function point			1.72
Per KLOC			32.20
Defect Prevention	Efficiency		Remainder	Bad Fixes	Costs
JAD	27%		1,262	5	$28,052
QFD	30%		888	4	$39,633
Prototype	20%		713	2	$17,045
Models	68%		229	5	$42,684
Subtotal	86%		234	15	$127,415
Pre-Test Removal	Efficiency		Remainder	Bad Fixes	Costs
Desk check	27%		171	2	$13,225
Static analysis	55%		78	1	$7,823
Inspections	93%		5	0	$73,791
Subtotal	98%		6	3	$94,839
Test Removal	Efficiency		Remainder	Bad Fixes	Costs
Unit	32%		4	0	$22,390
Function	35%		2	0	$39,835
Regression	14%		2	0	$51,578
Component	32%		1	0	$57,704
Performance	14%		1	0	$33,366
System	36%		1	0	$63,747
Acceptance	17%		1	0	$15,225
Subtotal	87%		1	0	$283,845
					Costs
PRE-RELEASE COSTS			1,734	3	$506,099
POST-RELEASE REPAIRS	(TECHNICAL DEBT)		1	0	$658
MAINTENANCE OVERHEAD					$46,545
COST OF QUALITY	(COQ)				$553,302
Defects delivered			1
High severity			0
Security flaws			0
High severity %			11.58%
Delivered Per FP			0.001
High severity per FP			0.000
Security flaws per FP			0.000
Delivered Per KLOC			0.014
High severity per KLOC			0.002
Security flaws per KLOC			0.001
Cumulative			99.96%
Removal Efficiency

Function points are able to quantify requirements and design defects, which outnumber coding defects for large applications.  This is not possible using LOC metrics.  Function points are also superior to “cost per defect” for measuring technical debt and cost of quality (COQ).  Both technical debt and COQ are standard SRM outputs.

Table 7 is only an example.  SRM can also model various ISO standards, certification of test personnel, team experience levels, CMMI levels, and in fact a total of about 200 specific quality factors.

Function Points and Software Maintenance, Enhancements, and Total Cost of Ownership
Software costs do not end when the software is delivered.  Nor does delivery put an end to the need to monitor both costs and quality.  Some applications have useful lives that can span 20 years or more.  These applications are not fixed, but add new features on an annual basis.  Therefore function point metrics need to continue to be applied to software projects after release.

Post-release costs are more complex than development costs because they need to integrate enhancements or adding new features, maintenance or fixing bugs, and customer support or helping clients when they call or contact a company about a specific application.

The need to keep records for applications that are constantly growing over time means that normalization of data will need to be cognizant of the current size of the application.  The method used by Software Risk Master ™ is to normalize results for both enhancements and maintenance at the end of every calendar year; i.e. the size of the application is based on the date of December 31.   The pre-release size is based on the size of the application on the day it was first delivered to clients.  The sizes of requirements creep during development are also recorded.

Table 8 shows the approximate rate of growth and the maintenance and enhancement effort for five years for an application of a nominal 1000 function points when first delivered:

Table 8:  Five Years of Software Maintenance and Enhancement for 1000 Function Points
(MAINTENANCE + ENHANCEMENT)
		Year 1	Year 2		Year 3	Year 4	Year 5	5-Year
		2013	2014		2015	2016	2017	Totals
Annual enhancement %		80	86		93	101	109	469
Application Growth in FP		1,080	1,166		1,260	1,360	1,469	1,469
Application Growth in LOC		57,600	62,208		67,185	67,185	78,364	78,364
Cyclomatic complexity increase		11.09	11.54		12.00	12.48	12.98	12.98
Enhancement staff		0.81	0.88		0.96	1.05	1.15	0.97
Maintenance staff		5.68	5.72		5.85	6.36	7.28	6.18
Total staff		6.49	6.61		6.81	7.41	8.43	7.15
Enhancement effort (months)		9.72	10.61		11.58	12.64	13.80	58.34
Maintenance effort (months)		68.19	68.70		70.20	76.31	87.34	370.74
Total effort (months)		77.91	79.30		81.78	88.95	101.14	429.08
Total effort (hours)		10,283.53	10,467.77		10,794.70	11,741.94	13,350.37	56,638.31
Enhancement Effort %		12.47%	13.37%		14.16%	14.21%	13.64%	13.60%
Maintenance Effort %		87.53%	86.63%		85.84%	85.79%	86.36%	86.40%
Total Effort %		100.00%	100.00%		100.00%	100.00%	100.00%	100.00%
Enhancement cost		$77,733	$84,845		$92,617	$101,114	$110,403	$466,712
Maintenance cost		$331,052	$316,674		$304,368	$315,546	$347,348	$1,614,988
Total cost		$408,785	$401,518		$396,985	$416,660	$457,751	$2,081,700
Enhancement cost %		19.02%	21.13%		23.33%	24.27%	24.12%	22.42%
Maintenance cost %		80.98%	78.87%		76.67%	75.73%	75.88%	77.58%
Total Cost		100.00%	100.00%		100.00%	100.00%	100.00%	100.00%

The original development cost for the application was $1,027,348.  The costs for five years of maintenance and enhancements the cost were $2,081,700 or more than twice the original development cost.  The total cost of ownership is the sum of development and the 5-year M&E period.  In this example the TCO is $3,109,048.

CEO’s and other C level executives want to know the “total cost of ownership” (TCO) of software and not just the initial development costs.

Five-year maintenance and enhancement predictions are standard outputs from Software Risk Master ™ (SRM).

Function Points and Forensic Analysis of Canceled Projects
About 35% of large systems > 10,000 function points are canceled and never delivered to end users or clients.  These canceled projects are seldom studied but forensic analysis of failures can lead to important insights.

Because they are terminated, the size of the project at termination and accumulated costs and resource data may not be available.   But Software Risk Master ™ can provide both values.

Tables 9.1 through 9.3 are taken from Chapter 7 of the author’s book The Economics of Software Quality, Addison Wesley, 2011.

Table 9.1: Odds of Cancellation by Size and Quality Level
	(Includes negative ROI, poor quality, and change in business need)
Function	Low	Average	High
Points	Quality	Quality	Quality
10	2.00%	0.00%	0.00%
100	7.00%	3.00%	2.00%
1000	20.00%	10.00%	5.00%
10000	45.00%	15.00%	7.00%
100000	65.00%	35.00%	12.00%
Average	27.80%	12.60%	5.20%
Table 9.2: Probable Month of Cancellation from Start of Project
	(Elapsed months from start of project)
Function	Low	Average	High
Points	Quality	Quality	Quality
10	1.4	None	None
100	5.9	5.2	3.8
1000	16.0	13.8	9.3
10000	38.2	32.1	16.1
100000	82.3	70.1	25.2
Average	45.5	30.3	16.9
Percent	150.10%	100.00%	55.68%
Table 9.3: Probable Effort from Project Start to Point of Cancellation
	(Effort in terms of Person Months)
Function	Low	Average	High
Points	Quality	Quality	Quality
10	0.8	None	None
100	10.0	7.9	5.4
1000	120.5	92.0	57.0
10000	2,866.5	2,110.1	913.2
100000	61,194.7	45,545.5	13,745.5
Average	21,393.9	15,915.9	4,905.2
Percent	134.42%	100.00%	30.82%

When high-quality projects are canceled it is usually because of business reasons.  For example the author was working on an application when his company bought a competitor that already had the same kind of application up and running.  The company did not need two identical applications, so the version under development was canceled.   This was a rational business decision and not due to poor quality or negative ROI.

When low-quality projects are canceled it is usually because they are so late and so much over budget that their return on investment (ROI) turned from positive to strongly negative.  The delays and cost overruns explain why low-quality canceled projects are much more expensive than successful projects of the same size and type.  Function point metrics are the best choice for forensic analysis of canceled projects.

Portfolio Analysis with Function Point Metrics
To be useful and interesting to CEO’s and other C level executives function points should be able to quantify not just individual projects but also large collections of related projects such as the full portfolio of a Fortune 500 company.

Table 10 is an example of the special SRM predictions for corporate portfolios.  This prediction shows the size in function points for the portfolio of a Fortune 500 manufacturing company:

Table 10:  Portfolio Analysis of a Fortune 500 Manufacturing Company

			Number of
			Applications	Function	Lines of
	Corporate Functions		Used	Points	Code
1	Accounts payable		18	26,674	1,467,081
2	Accounts receivable		22	33,581	1,846,945
3	Advertising		32	47,434	2,134,537
4	Advisory boards – technical		6	8,435	463,932
5	Banking relationships		38	131,543	7,234,870
6	Board of directors		4	6,325	347,900
7	Building maintenance		2	3,557	195,638
8	Business intelligence		21	73,972	4,068,466
9	Business partnerships		18	44,457	2,445,134
10	Competitive analysis		30	74,635	4,104,901
11	Consultant management		3	4,609	253,486
12	Contract management		32	94,868	5,217,758
13	Customer resource management		56	140,585	7,732,193
14	Customer support		45	67,003	3,685,140
15	Divestitures		10	15,000	825,000
16	Education – customers		7	11,248	618,663
17	Education – staff		4	6,325	347,900
18	Embedded software		84	252,419	21,455,576
19	Energy consumption monitoring		4	6,325	347,900
20	Energy acquisition		5	7,097	390,350
21	Engineering		79	276,699	20,752,447
22	ERP – Corporate		63	252,982	17,708,755
23	Finances (corporate)		84	210,349	11,569,183
24	Finances (divisional)		63	157,739	8,675,663
25	Governance		10	25,000	1,375,000
26	Government certification (if any)		24	35,571	1,956,383
27	Government regulations (if any)		13	20,003	1,100,155
28	Human resources		7	11,248	618,663
29	Insurance		6	8,935	491,421
30	Inventory management		45	67,003	3,685,140
31	Legal department		24	35,571	1,956,383
32	Litigation		32	47,434	2,608,879
33	Long-range planning		7	18,747	1,031,105
34	Maintenance – product		75	112,484	6,186,627
35	Maintenance – buildings		6	8,435	632,634
36	Manufacturing		178	311,199	23,339,917
37	Market research		38	56,376	3,100,659
38	Marketing		27	39,911	2,195,098
39	Measures – customer satisfaction		4	6,325	347,900
40	Measures – financial		24	35,571	1,956,383
41	Measures – market share		8	12,621	694,151
42	Measures – performance		9	14,161	778,850
43	Measures – quality		10	15,000	825,000
44	Measures – ROI and profitability		32	47,434	2,608,879
45	Mergers and acquisitions		24	59,284	3,260,639
46	Office suites		8	29,449	1,619,686
47	Open-source tools – general		67	100,252	5,513,837
48	Order entry		27	39,911	2,195,098
49	Outside services – manufacturing		24	35,571	1,956,383
50	Outside services – legal		27	66,518	3,658,497
51	Outside services – marketing		15	22,444	1,234,394
52	Outside services – sales		17	25,182	1,385,013
53	Outside services – terminations		9	11,141	612,735
54	Outsource management		32	47,434	2,608,879
55	Patents and inventions		19	28,255	1,554,010
56	Payrolls		21	52,837	2,906,047
57	Planning – manufacturing		42	63,254	3,478,996
58	Planning – products		10	15,000	825,000
59	Process management		12	17,828	980,514
60	Product design		56	140,585	7,732,193
61	Product nationalization		13	30,004	1,650,233
62	Product testing		38	56,376	3,100,659
63	Project offices		32	55,340	3,043,692
64	Project management		10	27,500	1,512,500
65	Purchasing		30	44,781	2,462,941
66	Quality control		13	20,003	1,100,155
67	Real estate		8	12,621	694,151
68	Research and development		106	370,739	20,390,634
69	Sales		45	67,003	3,685,140
70	Sales support		15	22,444	1,234,394
71	Security – buildings		21	31,702	1,743,628
72	Security – computing and software		32	110,680	6,087,384
73	Shareholder relationships		8	29,449	1,619,686
74	Shipping/receiving products		27	66,518	3,658,497
75	Software development		79	238,298	13,106,416
76	Standards compliance		13	20,003	1,100,155
77	Stocks and bonds		21	73,972	4,068,466
78	Supply chain management		47	70,973	3,903,498
79	Taxes		42	84,339	4,638,662
80	Travel		10	25,000	1,375,000
81	Unbudgeted costs – cyber attacks		32	86,963	4,782,945
82	Warranty support		7	10,025	551,384
					–
		Portfolio Totals	2,366	5,192,567	308,410,789

It is obvious that sizing a full portfolio with more than 2,300 applications and more than 5,000,000 function points cannot be accomplished by manual function point counting.  At an average counting rate of 500 function points per day, counting this portfolio would take 10,385 days.  At a cost of $1,500 per day the expense would be $5,192,197.

This fact alone explains why faster and cheaper function point analysis is a critical step leading to interest in function points by chief executive officers (CEO’s) and other C level executives.

Industry Studies using Function Point Metrics
One of the high-interest levels of CEO’s and other C level executives is in the area of how their companies compare to others in the same business sector, and how their business sectors compare to other business sectors.  Function point metrics are the best choice for these industry studies.

Table 11 shows approximate productivity and quality results for 68 U.S. industries using function points as the basis of analysis:

Table 11:  Approximate Industry Productivity and Quality Using Function Point Metrics

					Software	Defect	Removal	Delivered
					Productivity	Potentials	Efficiency	Defects
	Industry				2013	2013	2013	2013
1	Government – intelligence				7.20	5.95	99.50%	0.03
2	Manufacturing – medical devices				7.75	5.20	98.50%	0.08
3	Manufacturing – aircraft				7.25	5.75	98.00%	0.12
4	Telecommunications operations				9.75	5.00	97.50%	0.13
5	Manufacturing – electronics				8.25	5.25	97.00%	0.16
6	Manufacturing – telecommunications				9.75	5.50	96.50%	0.19
7	Manufacturing – defense				6.85	6.00	96.25%	0.23
8	Government – military				6.75	6.40	96.00%	0.26
9	Entertainment – films				13.00	4.00	96.00%	0.16
10	Manufacturing – pharmaceuticals				8.90	4.55	95.50%	0.20
11	Smartphone/tablet applications				15.25	3.30	95.00%	0.17
12	Transportation – airlines				8.75	5.00	94.50%	0.28
13	Software (commercial)				15.00	3.50	94.00%	0.21
14	Manufacturing – automotive				7.75	4.90	94.00%	0.29
15	Transportation – bus				8.00	5.10	94.00%	0.31
16	Manufacturing – chemicals				8.00	4.80	94.00%	0.29
17	Banks – investment				11.50	4.60	93.75%	0.29
18	Open source development				13.75	4.40	93.50%	0.29
19	Banks – commercial				11.50	4.50	93.50%	0.29
20	Credit unions				11.20	4.50	93.50%	0.29
21	Professional support – medicine				8.55	4.80	93.50%	0.31
22	Government – police				8.50	5.20	93.50%	0.34
23	Entertainment – television				12.25	4.60	93.00%	0.32
24	Manufacturing – appliances				7.60	4.30	93.00%	0.30
25	Software (outsourcing)				14.00	4.65	92.75%	0.34
26	Manufacturing – nautical				8.00	4.60	92.50%	0.35
27	Process control				9.00	4.90	92.50%	0.37
28	Stock/commodity brokerage				10.00	5.15	92.50%	0.39
29	Professional support – law				8.50	4.75	92.00%	0.38
30	Games – computer				15.75	3.00	91.00%	0.27
31	Social networks				14.90	4.90	91.00%	0.44
32	Insurance – Life				10.00	5.00	91.00%	0.45
33	Insurance – medical				10.50	5.25	91.00%	0.47
34	Public utilities – electricity				7.00	4.80	90.50%	0.46
35	Education – University				8.60	4.50	90.00%	0.45
36	Automotive sales				8.00	4.75	90.00%	0.48
37	Hospitals				8.00	4.80	90.00%	0.48
38	Insurance – property and casualty				9.80	5.00	90.00%	0.50
39	Oil extraction				8.75	5.00	90.00%	0.50
40	Consulting				12.70	4.00	89.00%	0.44
41	Public utilities – water				7.25	4.40	89.00%	0.48
42	Publishing (books/journals)				8.60	4.50	89.00%	0.50
43	Transportation – ship				8.00	4.90	88.00%	0.59
44	Natural gas generation				6.75	5.00	87.50%	0.63
45	Education – secondary				7.60	4.35	87.00%	0.57
46	Construction				7.10	4.70	87.00%	0.61
47	Real estate – commercial				7.25	5.00	87.00%	0.65
48	Agriculture				7.75	5.50	87.00%	0.72
49	Entertainment – music				11.00	4.00	86.50%	0.54
50	Education – primary				7.50	4.30	86.50%	0.58
51	Transportation – truck				8.00	5.00	86.50%	0.68
52	Government – state				6.50	5.65	86.50%	0.76
53	Manufacturing – apparel				7.00	3.00	86.00%	0.42
54	Games – traditional				7.50	4.00	86.00%	0.56
55	Manufacturing – general				8.25	5.20	86.00%	0.73
56	Retail				8.00	5.40	85.50%	0.78
57	Hotels				8.75	4.40	85.00%	0.66
58	Real estate – residential				7.25	4.80	85.00%	0.72
59	Mining – metals				7.00	4.90	85.00%	0.74
60	Automotive repairs				7.50	5.00	85.00%	0.75
61	Wholesale				8.25	5.20	85.00%	0.78
62	Government – federal civilian				6.50	6.00	84.75%	0.92
63	Waste management				7.00	4.60	84.50%	0.71
64	Transportation – trains				8.00	4.70	84.50%	0.73
65	Food – restaurants				7.00	4.80	84.50%	0.74
66	Mining-coal				7.00	5.00	84.50%	0.78
67	Government – county				6.50	5.55	84.50%	0.86
68	Government – municipal				7.00	5.50	84.00%	0.88
	TOTAL/AVERAGES				8.95	4.82	90.39%	0.46

Global Studies Using Function Point Analysis
In today’s world software development is a global business.  About 60% of Indian companies and more than 80% of Indian outsource companies use function points in order to attract outsource business, with considerable success.   As already mentioned Brazil now requires function points for all government outsource contracts.

Clearly global competition is a topic of critical interest to all C level executives including CEO’s, CFO’s, CTO’s CIO’s, CRO’s, and all others.

Function point metrics are the best (and only) metric that is effective for very large scale global studies of software productivity and quality.  Table 12 shows approximate results for 68 countries.

	Table 12: Approximate Global Productivity and Quality in Function Points
		Approximate	Approximate	Approximate	Approximate
		Software	Defect	Defect	Delivered
		Productivity	Potentials	Removal	Defects
		(FP per Month)	in 2013	Efficiency	in 2013
			(Defects per FP)		(Defects per FP)
1	Japan	9.15	4.50	93.50%	0.29
2	India	11.30	4.90	93.00%	0.34
3	Denmark	9.45	4.80	92.00%	0.38
4	Canada	8.85	4.75	91.75%	0.39
5	South Korea	8.75	4.90	92.00%	0.39
6	Switzerland	9.35	5.00	92.00%	0.40
7	United Kingdom	8.85	4.75	91.50%	0.40
8	Israel	9.10	5.10	92.00%	0.41
9	Sweden	9.25	4.75	91.00%	0.43
10	Norway	9.15	4.75	91.00%	0.43
11	Netherlands	9.30	4.80	91.00%	0.43
12	Hungary	9.00	4.60	90.50%	0.44
13	Ireland	9.20	4.85	90.50%	0.46
14	United States	8.95	4.82	90.15%	0.47
15	Brazil	9.40	4.75	90.00%	0.48
16	France	8.60	4.85	90.00%	0.49
17	Australia	8.88	4.85	90.00%	0.49
18	Austria	8.95	4.75	89.50%	0.50
19	Belgium	9.10	4.70	89.15%	0.51
20	Finland	9.00	4.70	89.00%	0.52
21	Hong Kong	9.50	4.75	89.00%	0.52
22	Mexico	8.65	4.85	88.00%	0.58
23	Germany	8.85	4.95	88.00%	0.59
24	Philippines	10.75	5.00	88.00%	0.60
25	New Zealand.	9.05	4.85	87.50%	0.61
26	Taiwan	9.00	4.90	87.50%	0.61
27	Italy	8.60	4.95	87.50%	0.62
28	Jordan	7.85	5.00	87.50%	0.63
29	Malaysia	8.40	4.65	86.25%	0.64
30	Thailand	7.90	4.95	87.00%	0.64
31	Spain	8.50	4.90	86.50%	0.66
32	Portugal	8.45	4.85	86.20%	0.67
33	Singapore	9.40	4.80	86.00%	0.67
34	Russia	8.65	5.15	86.50%	0.70
35	Argentina	8.30	4.80	85.50%	0.70
36	China	9.15	5.20	86.50%	0.70
37	South Africa	8.35	4.90	85.50%	0.71
38	Iceland	8.70	4.75	85.00%	0.71
39	Poland	8.45	4.80	85.00%	0.72
40	Costa Rica	8.00	4.70	84.50%	0.73
41	Bahrain	7.85	4.75	84.50%	0.74
42	Ukraine	9.10	4.95	85.00%	0.74
43	Turkey	8.60	4.90	84.50%	0.76
44	Viet Nam	8.65	4.90	84.50%	0.76
45	Kuwait	8.80	4.80	84.00%	0.77
46	Colombia	8.00	4.75	83.50%	0.78
47	Peru	8.75	4.90	84.00%	0.78
48	Greece	7.85	4.80	83.50%	0.79
49	Syria	7.60	4.95	84.00%	0.79
50	Tunisia	8.20	4.75	83.00%	0.81
51	Saudi Arabia	8.85	5.05	84.00%	0.81
52	Cuba	7.85	4.75	82.50%	0.83
53	Panama	7.95	4.75	82.50%	0.83
54	Egypt	8.55	4.90	82.75%	0.85
55	Libya	7.80	4.85	82.50%	0.85
56	Lebanon	7.75	4.75	82.00%	0.86
57	Iran	7.25	5.25	83.50%	0.87
58	Venezuela	7.50	4.70	81.50%	0.87
59	Iraq	7.95	5.05	82.50%	0.88
60	Pakistan	7.40	5.05	82.00%	0.91
61	Algeria	8.10	4.85	81.00%	0.92
62	Indonesia	8.90	4.90	80.50%	0.96
63	North Korea	7.65	5.10	81.00%	0.97
64	Nigeria	7.00	4.75	78.00%	1.05
65	Bangladesh	7.50	4.75	77.00%	1.09
66	Burma	7.40	4.80	77.00%	1.10
	AVERAGE/TOTAL	8.59	4.85	86.27%	0.67

 

Some of the data in table 12 is provisional and included primarily to encourage more studies of productivity and quality in countries that lack effective benchmarks circa 2013.  For example China and Russia are major producers of software but seem to lag India, Brazil, the Netherlands, Finland, and the United States in adopting modern metrics and function points.

Summary and Conclusions
Function point metrics are the most powerful metrics yet developed for studies of software economics, productivity, risks, and quality.  They are much better than older metrics such as “lines of code” and “cost per defect.”  They are also much better than alternate metrics such as “story points” and “use-case points”.

However the slow speed and high costs of manual function point analysis has caused function points to be viewed by top-executives such as CEO’s as a minor niche metric.  In order to be useful to C level executives function point metrics need:

1. Faster counting by more than an order of magnitude from today’s averages
2. Lower costs down below $0.05 per function point counted
3. Methodology benchmarks for all known methods
4. Quality benchmarks for defect prevention, pre-test removal and testing
5. Maintenance, enhancement, and total cost of ownership (TCO) benchmarks
6. Portfolio benchmarks for many companies and government groups
7. Industry benchmarks for all software-intensive industries
8. Global benchmarks for all countries that produce software in large volumes

This paper discusses a patent-pending method of sizing software projects in less than 2 minutes, with function points as one of the default metrics produced.  Software Risk Masters ™ (SRM) produces development estimates in about 3 minutes; quality estimates in 4 minutes; and maintenance estimates in 4 minutes.

The Software Risk Master ™ tool can do more than size.  The SRM tool can also predict the results and risks of any methodology, any level of team experience, any CMMI level, and programming language (or combination), and any volume of reusable materials.

References and Readings

Jones, Capers; “A Short History of Lines of Code Metrics”; Namcook Analytics Technical Report; Narragansett, RI; 2012.

This report provides a mathematical proof that “lines of code” metrics violate standard economic assumptions.  LOC metrics make requirements and design invisible.  Worse, LOC metrics penalize high-level languages.  The report asserts that LOC should be deemed professional malpractice if used to compare results between different programming languages.  There are other legitimate purposes for LOC, such as merely measuring coding speed.

Jones, Capers; “A Short History of the Cost Per Defect Metrics”; Namcook Analytics Technical Report; Narragansett, RI 2012.

This report provides a mathematical proof that “cost per defect” penalizes quality and achieves its lowest values for the buggiest software applications.  It also points out that the urban legend that “cost per defect after release is 100 times larger than early elimination” is not true.  The reason for expansion of cost per defect for down-stream defect repairs is due to ignoring fixed costs.   The cost per defect metric also ignores many economic topics such as the fact that high quality leads to shorter schedules.

Jones, Capers: “Sizing Up Software;” Scientific American Magazine, Volume 279, No. 6, December 1998; pages 104-111.

Jones, Capers; “Early Sizing and Early Risk Analysis”; Capers Jones & Associates LLC;

Narragansett, RI; July 2011.

Jones, Capers and Bonsignour, Olivier; The Economics of Software Quality; Addison Wesley Longman, Boston, MA; ISBN 10: 0-13-258220—1; 2011; 585 pages.

Jones, Capers; Software Engineering Best Practices; McGraw Hill, New York, NY; ISBN 978-0-07-162161-8; 2010; 660 pages.

Jones, Capers; Applied Software Measurement; McGraw Hill, New York, NY; ISBN 978-0-07-150244-3; 2008; 662 pages.

Jones, Capers; Estimating Software Costs; McGraw Hill, New York, NY; 2007; ISBN-13: 978-0-07-148300-1.

Jones, Capers; Software Assessments, Benchmarks, and Best Practices;  Addison Wesley Longman, Boston, MA; ISBN 0-201-48542-7; 2000; 657 pages.

Jones, Capers;  Conflict and Litigation Between Software Clients and Developers; Software Productivity Research, Inc.; Burlington, MA; September 2007; 53 pages; (SPR technical report).

The literature on function point metrics is quite extensive.  Following are some of the more useful books:

Abran, Alain and Dumke, Reiner R; Innovations in Software Measurement; Shaker-Verlag, Aachen, DE; ISBN 3-8322-4405-0; 2005; 456 pages.

Abran, Alain; Bundschuh, Manfred; Dumke, Reiner; Ebert; Christof; and Zuse, Horst; Software Measurement News; Vol. 13, No. 2, Oct. 2008 (periodical).

Bundschuh, Manfred and Dekkers, Carol; The IT Measurement Compendium; Springer-Verlag, Berlin, DE; ISBN 978-3-540-68187-8; 2008; 642 pages.

Chidamber, S.R. & Kemerer, C.F.; “A Metrics Suite for Object-Oriented Design”; IEEE Trans. On Software Engineering; Vol. SE20, No. 6; June 1994; pp. 476-493.

Dumke, Reiner; Braungarten, Rene; Büren, Günter; Abran, Alain; Cuadrado-Gallego, Juan J; (editors); Software Process and Product Measurement; Springer-Verlag, Berlin; ISBN 10: 3-540-89402-0; 2008; 361 pages.

Ebert, Christof and Dumke, Reiner; Software Measurement: Establish, Extract, Evaluate, Execute; Springer-Verlag, Berlin, DE; ISBN 978-3-540-71648-8; 2007; 561 pages.

Garmus, David & Herron, David; Measuring the Software Process:  A Practical Guide to Functional Measurement;  Prentice Hall, Englewood Cliffs, NJ; 1995.

Garmus, David and Herron, David; Function Point Analysis – Measurement Practices for Successful Software Projects; Addison Wesley Longman, Boston, MA; 2001; ISBN 0-201-69944-3;363 pages.

International Function Point Users Group (IFPUG); IT Measurement – Practical Advice from the Experts; Addison Wesley Longman, Boston, MA; 2002; ISBN 0-201-74158-X; 759 pages.

Kemerer, C.F.; “Reliability of Function Point Measurement – A Field Experiment”; Communications of the ACM; Vol. 36; pp 85-97; 1993.

Parthasarathy, M.A.; Practical Software Estimation – Function Point Metrics for Insourced and Outsourced Projects; Infosys Press, Addison Wesley, Upper Saddle River, NJ; 2007; ISBN 0-321-43910-4.

Putnam, Lawrence H.; Measures for Excellence — Reliable Software On Time, Within Budget; Yourdon Press – Prentice Hall, Englewood Cliffs, NJ; ISBN 0-13-567694-0; 1992; 336 pages.

Putnam, Lawrence H and Myers, Ware.;  Industrial Strength Software – Effective Management Using Measurement; IEEE Press, Los Alamitos, CA; ISBN 0-8186-7532-2; 1997; 320 pages.

Stein, Timothy R; The Computer System Risk Management Book and Validation Life Cycle; Paton Press, Chico, CA; 2006; ISBN 10: 1-9328-09-5; 576 pages.

Stutzke, Richard D; Estimating Software-Intensive Systems; Addison Wesley, Upper Saddle River, NJ; 2005; ISBN 0-201-70312-2; 918 pages.