R.D.Fernandez et al - Software rates vs price of function points

http://itconfidence2013.wordpress.com
Software Rates vs Price of Function Points
2°°°°International Conference on
IT Data collection, Analysis and Benchmarking
Tokyo (Japan) - October 22, 2014
Rafael de la Fuente
Founder and CEO of LEDAmc
Raúl Fernández
SW Productivity Consultant
Insert here a picture
An Updated Cost Analysis

2IT Confidence 2014 – October 22, 2014 http://itconfidence2014.wordpress.com
Software Rates vs Price of
Function Points: A cost
analysis
Goals of the presentation
Nowadays, in Software Development Contracts the key elements determining
the price are: the Rate and the Effort.
The possibility to measure the quantity of software produced (the size in
Function Points) allows us to assess whether there is a logical connection
between the price of the projects and the software actually produced.
The main goal of this presentation is to show what happens with the FP
price in several scenarios:
G1. Implementing estimation models.
G2. Implementing productivity models.
G3. Introducing new competitors in a productivity model.
G4. De-localizing providers.

• Background
• Objective
• 2014 vs 2012 benchmarking
• Case studies
• Conclusions
Agenda

Background
Over the last seven years, LEDAmc has managed the productivity of more than 12,000
development projects of 10 significant clients in Spain (mostly telecommunication and
financial companies). They were mainly small enhancement projects.
The main goal of the measures is to control large contracts of Adaptive Maintenances,
which involves the highest percentage of our clients’ development budget.
In 2012 LEDAmc developed a benchmarking study with some of these projects (we
selected 3,405). The purpose of the study was to analyze the relationship between the
rates of the different software providers and the price of the Function Point that they
were offering their clients. This study was presented:
•In the UKSMA conference in October 2012
•An updated study, in the IT Confidence Conference in Rio, in October 2013.

Background
The basic numbers of the 2012 sample were:
•10 Clients
•14 Providers
•3,405 Projects
•196,356 UFP
•2,168,192 Hours
•69,926,907 Million Euros
And the main conclusions were:
Pressure to lower rates ends up in lower productivity and higher FP price.
There is not a logical relationship between Rates and FP Price.
The performance of the providers changes dramatically among clients.
Clients with only one development provider have the highest FP Price.

Background
CLIENT-PROVIDER
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
1 0 0
1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 2 0 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9 3 0 3 1
Pressure to lower rates ends up in lower productivity and higher FP price.

Background
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
1 0 0
1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 2 0 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9 3 0 3 1
The same provider for different clients, offers significant differences in FP price.
CLIENT-PROVIDER

Background
0
10
20
30
40
50
60
70
80
90
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
The price of the FP with only one provider is higher.
CLIENT-PROVIDER
MULTI-PROVIDER CLIENTS
ONE-PROVIDER CLIENTS

Objective
LEDAmc has been continuously working for most of those clients, implementing Software
Development Productivity Models over the last two years.
Our productivity benchmarking data base has grown: It has more data for the same
clients and new clients have been included.
The main goal of this presentation is to show what has happened in these two years with
clients who have worked hard to improve their software development productivity.
We will first explain the general evolution of the Function Point Price for those clients,
comparing 2012 and 2014 benchmarks.
Afterwards, we will focus on three case studies that are especially significant.
Finally, we will draw conclusions on the economics of software management, based on
our results.

2014 vs 2012 benchmarking
The 2014 sample were:
•Clients that were in the 2012 study and have implemented some estimation or
productivity control
•4 Clients
•12 Providers
•883 Projects
•44,897 UFP
•415,768 Man Hours
•15,428,772 Million Euros
What has happened with the FP price?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
CLIENT-PROVIDER
The rates have remained almost the same.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
The FP price is more homogeneous in 2014 than 2012.
CLIENT-PROVIDER

Apparently there is no relation between FP price and rate…
CLIENT-PROVIDER
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Rates vs FP Price in 2014

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
.. but if we group the data by client, the results look different.
CLIENT A CLIENT B CLIENT C CLIENT D

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
There are big differences in the FP price for the same provider with different clients.
CLIENT A CLIENT B CLIENT C CLIENT D
P1 P1P2 P2P3 P3 P3

Conclusions
The results are homogeneous on a client basis because they have implemented
estimation or productivity control models that allow them control the FP price.
Because there is no transparency in software development and the public FP price does
not exist in the market, the FP price stills remains very different among the clients.
The natural differences of the systems architectures and processes should not imply
such FP price differences.
… now we will present some case studies highlighting the change that has been
produced for these clients who have introduced control systems.

Case studies
High FP price
Only one provider
No estimation control
Case Study 1 Competition among providers + Estimation Model
INITIAL SITUATION OF THE CLIENT (2012)

Case studies
During 2013, the implementation of a centralized mechanism of estimation control has
allowed to objectively contrast the costs offered by providers with corporate cost
strategies.
It includes functional sizing, as well as other sizing methods (for parametrization,
technical changes, test support,..)
Benchmarking studies helped establish a cost strategy and a target price for FP .
Introducing a new development provider fosters competitiveness among providers and
helps establish a FP price within market values.
INPUTS (2013)

TOTAL 1 2PROV 1 PROV 2
INPUTS (2013)
Case studies
1. The rate was only
reduced by 5%
2. New development
provider worked with
a similar rate
21

TOTAL 1 2PROV 1 PROV 2
INPUTS (2013)
Case studies
TOTAL 1 2
The FP price was
reduced by 44 %
Similar FP Price
among providers
1. The rate was only
reduced by 5%
2. New development
provider worked with
a similar rate
21
PROV 1 PROV 2

Case studies
A good estimating process was used for all projects
44% reduction in FP price
More than 10 Million euros saved
FP Price within market values
Two competing providers in 2013 – Four in 2014
FINAL SITUATION (2014)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Case studies
Specialized providers
On-site development
High rates and very low FP price
Estimation and productivity control implemented
Case Study 2 De-localization of providers

Case studies
The client strategy was to proceed to the de-localization of the providers to a
typical outsourced model introducing new development providers with lower rates.
The client also decided to create an in-house development team in order to
preserve and extend the know-how.
INPUTS (2013)

Case studies
TOTAL 1 2 3 4
1. New provider with
lower rate
2. In-house development
rate
1 2
INPUTS (2013)
PROV 1 PROV 2 PROV 3 PROV 4

TOTAL 1 2 3 4
Case studies
TOTAL 1 2 3 4
1. New provider with
lower rate
2. In-house development
rate
1 2
Logical behavior of FP
price
In-house
developments at lower
FP cost than in
outsourced scenario
INPUTS (2013)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Case studies
FP price controlled
Outsourcing process controlled
2014 situation

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Case studies
Ratio of the rates: 1.1
Ratio of the FP Price: 2.3
Case Study 3 Controlling large maintenance contracts

Case studies
During 2012 the client implemented a productivity control for maintenance
development.
An annual productivity SLA for all the providers was established.
The client also implemented a productivity and quality providers ranking.
Every two months the client presented to the providers their position in the ranking.
INPUTS (2013)

TOTAL 1 2 3 4 5 6
Case studies
Renegotiation of rates
Ratio 2012: 1.1
Ratio 2014: 1.0
INPUTS (2013)

TOTAL 1 2 3 4 5 6
TOTAL 1 2 3 4 5 6
Case studies
Renegotiation of rates
Ratio 2012: 1.1
Ratio 2014: 1.0
FP cost control
Ratio 2012: 2.3
Ratio 2014: 1.7
INPUTS (2013)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Case studies
Uniformity of FP price
9.5 million euros saved
2014 situation

Conclusions
1. It is relatively easy to influence FP price. The main levers are:
>>> A centralized estimation control
>>> A development productivity control
>>> Several providers in competition
2. Working on those key elements the results are much better than working only
on the rates:
>>> The behavior of the unit cost of production (FP price) is more homogeneous.
>>> The cost of the development process can be controlled.
3. Diversification of providers with joint management results in a better uniformity
of the FP price.

R.D.Fernandez et al - Software rates vs price of function points

Recommandé

Recommandé

Contenu connexe

En vedette

En vedette (16)

Similaire à R.D.Fernandez et al - Software rates vs price of function points

Similaire à R.D.Fernandez et al - Software rates vs price of function points (20)

Plus de International Software Benchmarking Standards Group (ISBSG)

Plus de International Software Benchmarking Standards Group (ISBSG) (8)

Dernier

Dernier (20)

R.D.Fernandez et al - Software rates vs price of function points