The ROI of Continuous Delivery
This article discusses how to calculate the ROI of Continuous Delivery and links to a free Continuous Delivery ROI Calculator for your convenience. Calculating ROI is often tricky when numerous variables are involved. In this post, we stick to a practical method of calculating the ROI of Continuous Delivery by using the latest available research on Continuous Delivery and sticking to only the tangible benefits.
What is Continuous Delivery?
First, let’s make sure you understand what Continuous Delivery is. Continuous Delivery is a process that allows software teams to routinely, build, test and deploy software on a regular basis. This process is accomplished by completely automating each step in the Software Development Life Cycle (SDLC) and implementing the agile software methodology. You can get a deeper definition of Continuous Delivery.
Calculating the ROI of Continuous Delivery
It is safe to assume, that automating anything provides benefits, but the proof is in the ROI. While there are many less tangible but impressive benefits of Continuous Delivery, were going to focus on what we can calculate in this post.
There are four main advantages of Continuous Delivery. We use the latest industry measurements and surveys as a basis for the calculations. See the end of this article for the sources.
Productivity and Efficiency Returns of the Development Team
Continuous Delivery works best when it is automating the agile software process. But agile has short iteration cycles (sprints) that introduce a lot of overhead. The entire build, test and deploy process is performed in only 2-4 weeks, so manual tasks performed are amplified. In other words, some of your most expensive resources are performing tedious, manual, repetitive tasks instead of working on new features.
A major area of time consumption for dev teams is problem resolution. Production problems are notoriously difficult to pin down and resolve, due to lack of access and dissimilarity of environments.
The majority of teams spend around 40% of their time on problem resolution, but can be as high as 75-90% in some cases. Continuous Delivery addresses these issues by improving the time for root issue determination of the problem by making it easier to deploy a production environment replica resulting in a 50-70%-time reduction in problem resolution.
ROI of Continuous Delivery Formula:
[12] Cost Savings Factor = .18 (40% x 75% x 60%) x Dev Team Salary
[6] Time majority of developers spend on problem resolution. = 30-40% (40%)
[9] Out of 40% of time, 75% of that time is spent on determining root cause (75%)
[12] Teams with Continuous Delivery eliminate problems resolution time by 50-70% (60%)
Productivity and Efficiency Return for IT Operations Team
Depending on the size and structure of your IT department, you may not have separate resources for operations (DevOps). If not, your developers are playing a prominent role in IT operations. Either way, there is a cost associated with these activities. The automation will save your developers or ops team valuable time.
Automated deployments of Continuous Delivery establish a consistent technology stack for deployment. Variations between environments are responsible for 40% of delays in deploying code to production. Problem resolution can be difficult to determine when facing environmental differences.
ROI of Continuous Delivery Formula:
[12] DevOps Gains Factor = .16 (7.2 hours/45 weekly hours) x DevOps Team yearly salary or Dev team salary (if no DevOps team)
[5] IT & DevOps productivity gains measured total of 7.2 hours savings per week.
[12] Platform inconsistencies account for 40% of delays in deployment to production
Return on Quality Improvements
To stick with the tangible, the best way to measure quality is via downtime. Quality issues that don’t result in downtime can still have an impact on your revenue and retention. Improved quality will also improve the time spent by the team with problem resolution discussed earlier.
The automation from Continuous Delivery and the use of unit test, acceptance test, smoke test, and integration test have a significant impact on quality. Downtime however, is measurable and can be considerable. It is a relatively easy calculation, if you know the amount of revenue your software application or platform generates yearly. You can then calculate the amount produced per minute.
ROI of Continuous Delivery Formula:
[12] Failures per year x Avg. minutes to recover x revenue per minute
[5] Average number of failures per month=2.13 with the median 1 per month
[3] Teams with DevOps resulted in 50% fewer failures and restored the issues 12x faster.
Return on accelerated features time to market
Software does not have value, unless it is in production. Continuous Delivery is the practice of consistently delivering incremental business value with less risk. It’s the commitment of keeping your software in a working state and deploying frequently.
The speed at which you can deliver high-quality code to production should accelerate revenue. New applications, features, enhancements allow you to innovate and deliver to your customers. The value will vary depending on your business of course, but we can use some industry data to establish some reasonable numbers, and you can adjust appropriately.
ROI of Continuous Delivery Formula:
[12] Formula: Gain = 15% x Revenue per year
[1] 19% increase in revenue from accelerated time to market
[12] Conservative average of 15%
Total ROI% on Continuous Delivery
To conservatively calculate the ROI of Continuous Delivery, you add the gains from all four main areas of benefit – the Cost of Investment to get the ROI%:
ROI% = (Total Gains – Cost of Investment) x 100/Cost of Investment
How much should you use for the cost of investment? The actual cost of investment can vary widely depending on the size and complexity of your team and what solution you select. For SMB’s, an estimate is easier to approximate as the organizational complexity is less than a large company.
The good news is that you could use a large number for investment and still have a generous ROI. With Continuous Delivery, the ROI is almost always a winning number.
To help you out, we created an online ROI Calculator to make it easier. Go ahead and have fun with it and see what your ROI could be.
[supsystic-social-sharing id=’1′]
References and resources:
[1] DevOps Driving 20 Percent Faster Time-to-Market for New Services, Global IT Study Reveals, http://
www.ca.com/us/news/press-releases/na/2013/devops-driving-20-percent-faster-timetomarket-fornew-
services-global-it-study-reveals.aspx – from TechInsights Report: What Smart Businesses Know
About DevOps, CA Technologies, September 2013
[2] Jon Jenkins, Amazon.com, “Velocity Culture”, Velocity Conference June 2011
[3] The State of DevOps Report (by Puppet Labs), 2013, http://info.puppetlabs.com/2013-state-ofdevops-
report.html
[4] http://www.incomediary.com/20-websites-making-the-most-money
[5] IT Ops & DevOps Productivity Report 2013: Tools, Methodologies and People, RebelLabs
[6] 2013 Cost of Data Center Outages, Ponemon Institute Research Report, December 2013
[7] http://static.zend.com/topics/Zend-Developer-Pulse-report-Q2-2013-0523-EN.pdf
[8] http://static.zend.com/topics/zend-developer-pulse-survey-report-Q2-2012-0612-EN.pdf
[9] An ISV’s Guide to Achieving Business Advantage through Faster Application Problem Resolution,
BMC Software
[10] Michael Krigsman, “Worldwide cost of IT failure (revisited): $3 trillion” http://www.zdnet.com/blog/
projectfailures/worldwide-cost-of-it-failure-revisited-3-trillion/15424
[11] Accenture Technology Vision 2014, From Digitally Disrupted to Digital Disrupter, www.accenture.
com/technologyvision.
[12] Zend Blueprint for Continuous Delivery Technology, https://pages.zend.com/CDRoi.html.
Mr. Barbato brings over 35 years of experience in the software and technology industries to Third Wave. He has invested over 20 years in banking and highly regulated environments. He guides Third Wave’s expert team in applying Continuous Delivery Solutions to reduce risk and increase ROI. Follow on twitter: @frankjbarbato
