Scribd
Upload
45 views

Best Practices On Agile Metrics

The document discusses metrics that a project team at TCS implemented to improve deliverable quality and track progress in an agile project. It outlines 10 metrics measured, like story cycle time and code quality. For each metric, it defines what is measured and the benefit. These metrics allow for tracking, quality monitoring, and continuous improvement across iterations. Measuring and reporting on metrics at different levels helps ensure high reliability and minimal impact of changes.

Uploaded by

Ramesh Madurai
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
45 views

Best Practices On Agile Metrics

The document discusses metrics that a project team at TCS implemented to improve deliverable quality and track progress in an agile project. It outlines 10 metrics measured, like story cycle time and code quality. For each metric, it defines what is measured and the benefit. These metrics allow for tracking, quality monitoring, and continuous improvement across iterations. Measuring and reporting on metrics at different levels helps ensure high reliability and minimal impact of changes.

Uploaded by

Ramesh Madurai
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Best Practices

- Metric Collections in Agile Deliveries

TCS Internal
Title
DC/ SC/ Geo/Practice: Telecom
Name of the project/support fn: BT 21CN TE
Name of the author: Moutusi Sen/Lopamudra
Bhattacharya
Date Created: 10/09/2009

TCS Internal 11 September 2009


Description
Project / Context
Transfer Engineering project deals with the migration of the customers from
Legacy Network to IP based 21CN network. The Project Team is responsible for
solution design, component design, development, testing till the delivery and
implementation support.

The project is following Agile delivery methodology where the business


scenarios are delivered in various iteration cycle.

Purpose
To improve the quality of the deliverables and for better tracking of
the different iteration cycles, this project has identified and implemented some
measuring capabilities or metrics.
In this document those capabilities will be discussed so that the other
projects get benefited by reusing these measuring techniques.

TCS Internal 11 September 2009


How we did it
Following are the specific metrics which are followed exclusively in TE for improving the
quality of the deliverables.

Process adopted
Highlights

1. Story Cycle Time ƒ Metrics


2. Code Quality metrics identified and
3. Automated deployment reported to client
4. Automated Build & ƒFor each metrics,
configuration 5 different levels
are identified.
5. Automated regression
Target is to
testing achieve the
6. Iteration efficiency highest level for
7. Test Driven Development each metrics. An
8. Interface test prior to E2E improvement plan
test will be there to
reach the top of
9. Story Acceptance measure
the ladder.
10. Delivered defect density ƒReusable assets
11. Performance scalability
testing

TCS Internal 11 September 2009


How we did it
Metrics Definition Benefit
Story Cycle Time Avg. [Date difference (Iteration Start Date, Help in better iteration planning
Implementation Start Date) on Story i)]

Code Quality Number of review comments identified during Improve the code review quality
metrics the code review and accepted by developers by doing code review manually
or automatically.

Automated No. of activities automated in deployment * Help in automating the whole


deployment 100 /Total no. of activities can be considered deployment (checkout, sanity
for automated deployment test etc) and save time.

Automated Build & No. of activities automated in ABCM* 100 Help in automating the code
configuration /Total no. of activities can be considered for build and configuration
ABCM management process

Automated No. of e2e Testcases automated * 100 / Total Help in improving the
regression testing no. of possible automated e2e testcases automation in E2E regression
level.

Iteration efficiency Effort spent in dev and test *100/total effort This helps in identifying and
spent in the iteration. improving the actual “Value”
being provided to the customer.

TCS Internal 11 September 2009


How we did it
Metrics Definition Benefit
Test Driven No. of unit level Testcases automated * 100 / This is an important parameter
Development Total no. of possible unit level automated of agile and helps tracking the
testcases test cases executed from the
test pool before the actual code
deployment.
Interface test No of Interface activities stubbed * 100/ Total This helps in tracking and
prior to E2E test no. of Interface activities can be stubbed improving the interface testing
without involving the actual
interfacing team.
Story User Satisfaction Index This feedback is required to
Acceptance identify if there is any gap exists
measure in the implementation and also
to improve the process.
Delivered defect Number of Priority 1 and 2 faults found from This will improve the code
density CIT till the warranty period (during CIT + quality by reducing the number
IVV&T + Early Life Failure (i.e., First 28 days of defects from the E2E testing
in GO LIVE) per 1000 person hours of team.
development effort.
Performance Actual execution time to be noted and any Performance monitoring against
scalability testing deviation from earlier run. the benchmark needs to be
carried out on E2E level to
either tune the benchmark or to
identify issues.

TCS Internal 11 September 2009


Why this is a best practice
Justification
• Apart from individual benefits of each metrics, these
metrics help in better tracking, quality & performance
monitoring and benchmarking of the project.
• These metrics help in preparing an improvement plan to
improve constantly for better delivery efficiency.
• A retrospective needs to be carried out on the existing
improvement plan and next level of improvement to be
planned and monitored.

Benefits
• Idea is that the highest level of the metrics will ensure
high reliability of the delivery.
• Ongoing changes will ensure zero impact on the other
functionalities of the delivery where changes has not
incurred. This can be achieved by automatic build and
automatic regression.

TCS Internal 11 September 2009


How this may be adapted elsewhere
Replication
These all metrics can be discussed with the PEG team and other agile
teams in TCS and can be spread in organization level.

Contact Info
[email protected]
[email protected]

TCS Internal 11 September 2009

You might also like