Use these KPIs for the best results when outsourcing software development

KPIs for outsourcing software are essential for improving productivity, software quality, and customer satisfaction in outsourced development projects. When developing software with an external IT partner, maintaining control throughout the process is crucial. What KPIs can you apply to ensure the success of outsourced software development? Nearshore software developer NetRom and Software Improvement Group (SIG) explain in this blog post how to effectively manage this process.

NetRom and SIG have been working together for many years for various clients and share a common vision on KPIs for outsourcing software development. This article will help guide you in thinking about KPIs when managing both the process and the final outcome of software development by an external party. We will cover the productivity of the development process, code quality, the health of the development process, and customer satisfaction.

First of all: KPIs are never an end in themselves, but should provide insight into the impact of your actions. To what extent have you achieved the intended goals? Useful KPIs give you the ability to make adjustments. Making IT-related KPIs practically useful sometimes requires extra thinking about the actual end result you want to achieve. Instead of measuring “application uptime,” you might also think of “no flight delays due to IT,” as mentioned by Marco Gianotten of Giarte in the IT Xperience monitor 2021.

Metrics for productivity

We begin with the productivity of a software development project – the speed at which software is created. Productivity is crucial for more strategic business goals, such as a quick time-to-market. You can also monitor what is delivered compared to the plan and keep track of how much functionality is being completed. These variables are relevant to both the product owner and the development team.

How do you measure productivity?

When measuring productivity, you’ll get the most benefit from three KPIs in practice: sprint burndown, team velocity, and release burndown.

• Sprint burndown is a graphical representation of the estimated remaining time for a specific period during a sprint. This KPI helps predict when all work will be completed and also helps track scope creep.
• Team velocity refers to the average number of user stories completed by the team over multiple sprints. This KPI provides insight into the entire project: does it fit within the planned number of sprints? Note: this KPI is too subjective to compare teams against each other.
• Release burndown provides insight into the progress of the release by comparing the remaining work at the end of each sprint against the planned workload. The remaining work can be expressed in story points, ideal days, team days, etc.

Metrics for code quality

The second group of KPIs focuses on the quality of the software. Some managers believe that high software quality conflicts with functionality and speed, but the opposite is true. Poor software quality leads to high maintenance costs and delayed time-to-market. On the other hand, high-quality code is easier to manage and maintain, safer, and more cost-efficient; it results in a lower total cost of ownership (TCO). Many metrics are general measurements, unlike KPIs, which are specifically focused on strategic objectives. Metrics can be measured automatically, allowing developers to receive alerts if threshold values, such as those related to quality, are exceeded.

How to measure software quality?

ISO 25010 is a global standard for software quality that measures the quality characteristics of software across eight different aspects. While ISO 25010 describes what quality is, is ISO 25002 the standard that outlines how to assess that quality. SIG is the first fully certified company in the world to measure according to the ISO 25010 standard. Here, we present the key variables (and sub-variables) that play a role in measuring and evaluating software quality.

Manageability

Manageability (the ease of maintaining software) is divided into five components: analyzability, modifiability, testability, modularity, and reusability. These components can be seen as a representation of the stages involved in performing maintenance tasks.

Modularity improves manageability in the other four stages.

• Availability refers to the degree to which a system is operational and accessible when needed for use.

• Maturity refers to how well a system meets the need for reliability under normal use.

• Fault tolerance shows whether a system continues to function as intended despite the presence of errors.

• Transferability (portability) is important for how effectively and efficiently software can be adapted for new forms of hardware or software.

• Replaceability concerns the degree to which a product can replace another specified software product.

Performance in relation to software quality

Performance plays a crucial role in software quality. Consider the following factors:

• Response time: Does the speed and processing capacity of the system meet the required standards?
• Capacity: What is the maximum growth that can occur within the response time requirements?
• Resources: How many resources are needed for the software to meet the performance requirements for response speed and capacity?

Part of ISO 25010 also includes the usability of software. This refers to the extent to which a solution can be used by users to achieve specific goals. Along with effectiveness and efficiency, user satisfaction is a key aspect of usability.

Additionally, essential are the security features. Key components include the integrity of the software (the degree to which software can prevent unauthorized access to applications and data), as well as non-repudiation and accountability. These concepts refer to the ability to demonstrate actions or events and the irrefutable ability to trace actions back to the entity performing them.

Metrics for the health of your software development process

In addition to the productivity of development teams and the quality of their delivered products, it’s also essential to evaluate the software development and implementation process.

• Lead time refers to the time between the start and execution of a process. It also reveals something about the partnership, highlighting areas where delays may occur or need to be discussed and resolved.
• Cycle time indicates how much time a team spends working on a specific task. It’s a measure of how quickly the team delivers new functionality to the client.
• The third and final health indicator is flow efficiency, which refers to the percentage of time spent actively working on your backlog items compared to the total available time. A low flow efficiency may indicate bottlenecks in the process – such as a developer having to wait before completing a task.
  

Metrics for customer satisfaction

In our opinion, there are four key factors that determine customer satisfaction in a partnership.

“Do what you promise” is the starting point. The first law of software development is: “You can do it well, fast, or cheaply, but you can only choose two.” Taking responsibility for commitments at the sprint level is how you ensure you deliver what you promised. To continuously monitor this, you can use a metric like sprint burndown.

The second success factor for customer satisfaction is employee retention. High turnover can be detrimental to the continuity of your projects and the retention of knowledge within your teams.

The third factor is domain expertise. Having and applying domain knowledge contributes to the creation of software with an excellent user experience. Our tip: ask your development partner for proposals to test the domain expertise of developers.

And finally: never compromise on quality. This contributes to future-proof products that are reliable, secure, and easily adaptable.

Getting started with KPIs? Use them in dialogue with your partner

Ultimately, in a successful partnership, you should jointly decide which KPIs to apply. Your product owners should be able to work with the selected metrics and KPIs. The team must also be able to validate and interpret the metrics related to code quality and architectural decisions in collaboration with an architect. It’s wise to make agreements with your development partner regarding the standards applied for code quality. You can supplement this with external audits or certifications like SIG TÜViT certification. Lastly, make use of KPIs in ongoing dialogue, sprint retrospectives, and sprint cycle evaluations.

Want to learn more? Download our comprehensive whitepaper!

For those who want to dive deeper into the use of KPIs in software development, NetRom Software offers a detailed whitepaper. Enter your name and email address and receive the whitepaper “KPIs for outsourced software development”.

Looking for reliable software development services?
See how we can help.