Thanks to Blue Ink and the Just-In-Time Application Generation
(JAG) Methodology, Headstrong Public Sector realized a
400% increase in productivity on its last software development project compared to traditional development methods.
Below is information on the metric used along with details about Headstrong's time savings.
At right find our time savings calculator, which will provide an estimate of how much time you could save on your next development project. If this page interests you, you may also be interested in our user submitted success stories.
Hours Per Base Entity
The Headstrong consulting company has used the hours per base
entity metric as a measure of efficiency for software development projects for the past seven years.
It is a faster, less rigorous version of the hours per function point metric used by the
International Function Point Users Group (IFPUG).
The sections below describe the two elements of the metric: base entities, and the types of hours
included in the metric.
The hours per entity metric assumes that base entities (also known as entity, base class, or main table) represent a unit of work. A base entity is a non-lookup, non-associative table in a database that generally represents an entity in real life such as an employee, customer or sale. The hours per entity metric assumes a moderate to large amount of work is required to implement the Create, Read, Update, Delete (CRUD) functionality as well as custom business rule functionality for one base entity.
Hours per Base Entity
The hours per base entity metric uses the total cumulative hours
for all resources performing software development on a project.
Hours for analysis, design, testing, management, or development of extensive reporting are not included.
The Headstrong consulting company has used the hours per base entity metric (labeled Hours per Main Table in the chart below) on development projects since 1999. The projects in the chart below used similar architectures (all n-tier, some .Net, some classic ASP), similar complexities, and had similar toolsets. The projects in red (solid line) did not use code generation while the projects in green (dash line) used various Rapid Application Development tools.
The trend is that projects without code generation technology took 121 hours per base entity while the projects that used code generation technology took 77 hours per entity. Headstrong management found these metrics to be consistent and used them repeatedly for project estimation.
Blue Ink Metrics
In early 2005 the Headstrong consulting company used Blue Ink in conjunction with the Just-In-Time Application Generation (JAG) Methodology on an n-tier .Net resource management web application project for a top four consulting company. The project contained 16 base entities (58 total tables). Using traditional metrics the project should have taken 1,936 hours without code generation or 1,232 with code generation.
The resource management project took 368 development hours to complete! Blue Ink and the JAG Methodology increased productivity over 400% compared to hand coding and over 200% compared to other code generation technologies.
A primary developer of the project, Ankush Tewari, believes this metric is repeatable and attainable on other projects: "Depending on the scope of work and the type of work Blue Ink could generate at least 50% time savings, even on a project that is heavily business rule centric. Just getting the CRUDs out there would probably save half the time."