The Value of Iterative Design Processes
Whether you’re developing software, designing a website, or refining a recipe, you probably rely on an iterative process. The iterative process starts with requirements or assumptions that form the base of the project. In essence, you create the first product, test it, and revise for the next version.
The iterative process is simply a series of steps that you repeat, tweaking and improving your product with each cycle. In practical terms, think of it as practice to make your product perfect.
In this article, you will learn the steps of the iterative process, how it’s used in workplaces from development to design, and why you would use this approach rather than a traditional project methodology such as Waterfall.
In this article
What Is Meant by the “Iterative Process”?
What Is an Iterative Procedure?
Why Use an Iterative Process?
What Is the Iterative Process Model?
How Does the Iterative Process Work?
What’s the Difference Between Iterative and Incremental Development?
What Is the Iterative Development Process?
What Is the Iterative Process in Scrum?
What Is the Iterative Design Process?
What Is a Non-iterative Process?
Iterative Architecture Process
What Does an Iterative Process Mean in Research?
What Is the Iterative Process in Qualitative Research?
What Does the Iterative Process Mean in Psychology?
Challenges of the Iterative Process
Improve the Iterative Design Process with Smartsheet for Project Management
What Is Meant by the “Iterative Process”?
The iterative process is an approach to continuously improving a concept, design, or product. Creators produce a prototype, test it, tweak it, and repeat the cycle with the goal of getting closer to the solution.
Anyone can use the iterative process, but designers, developers, educators, scientists, and mathematicians most often utilize it. The concept and the solution eventually converge, such as in a math function or a scientific discovery, because you progress toward your desired result each time you iterate on the product.
The iterative process is widespread across many industries. Most Agile projects use an iterative approach, incrementally improving the product with each cycle or sprint. The end of one iteration becomes the starting point for the next round.
For example, think of Microsoft or Apple products. Using an iterative approach, these products are regularly updated with new features or benefits, minus some of the problems of previous editions. Even writers, musicians, and cooks use the iterative process to refine their creative work.
You can also use the iterative process when the final output or decision cannot be easily revoked (such as a jury verdict when many votes have to be taken to reach agreement) or if the consequences of revoking the outcome could be emotionally and financially costly (such as a marriage) or have long-term implications (such as war).
What Is an Iterative Procedure?
The iterative process refines a product through repeated adjustment cycles. In math, an iterative procedure identifies the solution to a problem based on an initial guess. Then, you introduce a sequence of algorithms to find the answer.
You can adapt this math procedure for any problem that requires a specific solution. While you may not have the right answer initially, you can start with a guess, review your solution, refine your guess, review the revised solution, and repeat until you get the correct answer.
Imagine you are trying to duplicate a dessert you enjoyed at a restaurant. You know it contains chocolate, but can’t identify the secret ingredient that makes it taste unique. You take a guess, cook the dish, and taste it to see how close you are to the original. Then, you adjust the ingredients and the quantities, and try again. Each iteration gets you closer to your goal: the perfect dessert.
Why Use an Iterative Process?
The iterative model is an alternative to the Waterfall approach, which relies on sequential steps. When using Waterfall, it’s difficult to quickly address development or design inefficiencies, but iteration gives developers more flexibility to swiftly incorporate changes.
NASA adopted the iterative model in the 1960s with Project Mercury and has continued its use through the Space Shuttle project, as well as for working with the U.S. Air Force to develop the X-15 hypersonic aircraft. The approach has also spread to other industries.
In the 1970s, IBM began using the iterative model in computer system design. And, in the 1990s Canada developed its air traffic control system with an iterative approach. Today, businesses ranging from software applications and engineering to design, marketing, and education use an iterative approach for developing products and meeting market needs.
The iterative process gives you the ability to refine and revise a product quickly, especially if you have an initial version of a product but still need to identify detailed features and functions. This approach assumes that you don't already have all the answers or that the environment for your product, as well as the needs of your customers, will change during the development timeline. An iterative model takes into account the changing landscape, allows you to plan for changes as you create your product, and helps you produce deliverables that are tailored to your market. Here are some of the specific benefits:
It's efficient. You can build your product step-by-step, rather than having to rework an entire plan as changes occur. In addition, the workload of the team is spread out more effectively throughout the project's development lifecycle.
It's timely. The first iteration allows you to develop the top priority in functionality. Each iteration is based on any improvements identified in the past cycle, and continual testing gives you a clear picture of the status of your project. You can see the results early and often because each iteration is a managed milestone.
It's cost-effective. Any changes to the project's scope or requirements, which are common in most projects, are less costly than with a Waterfall approach.
It's collaborative. You can present the results of each iteration to stakeholders and clients. They can see the evolution of the project and ensure that you are meeting their requirements.
It improves usability. Testing and debugging are easier with smaller iterations, as you can identify defects early in the process and users and customers are involved with each iteration.
It eliminates confusion. You can detect inconsistencies or flaws in requirements, design, code, and other implementations with each iteration, so you can avoid misunderstandings.
It's easier to manage risk. You tackle the high-risk portions of a project first, and every iteration lets you identify and resolve risks.
It offers continual improvement. Each iteration allows the team to easily incorporate any lessons learned from previous runs and continually improve the development process.
What Is the Iterative Process Model?
The iterative process model is a cyclical process in which you make and test incremental adjustments. It is popular in technology, engineering, software development, design, qualitative research, project management, and other industries.
Here is some information on how companies use the iterative process model:
Website Wikis: These types of websites allow users to aggregate, update, or improve on content. For example, Wikipedia relies on user-generated content that allows reviewers to add relevant information or delete content that is outdated or no longer useful. A wiki is never finished. In the iterative process model, the site continually evolves with new information.
Human-Computer Interfaces: Many consumers had difficulties using early computers. Programmers had a specific set of skills and knowledge, but they did not represent everyday consumers. Using the iterative process model, designers gathered feedback, adjusted the interface based on that feedback, and released an updated version. These refinements continue today. Computer interfaces and software continue to evolve based on consumers’ feedback and needs. The process of improving usability now extends to a range of mobile devices and smartphones.
Computing and Computer Programming: Iteration is a key computer science technique for creating algorithms and developing software. With the iterative process model, a computer program runs a sequence of instructions or math operations over and over. The program loops to repeat some part of the program or code until it reaches a predetermined event or completes the desired number of repetitions. For example, a website could use code that repeatedly refreshes a page until the user clicks a button. Or, a computer program could run an algorithm to rearrange letters in a word (or the entire alphabet) until it forms all the possible combinations. Iteration is similar to recursion in computer programming. Both repeatedly follow a set of instructions and can be used for the same effect.
Project Management: The iterative approach helps organizations (and project managers) handle the uncertainty inherent in many projects. Rather than assuming that all requirements, use cases, and business needs can be identified before the project starts (typical in a Waterfall approach), the iterative model assumes that user expectations and business needs will change during the development timeframe. By creating a product in chunks that are refined and adapted in subsequent cycles, the iterative process responds to the unpredictable nature of projects, as well as customer needs and engagement.
Education: The iterative process model focuses on students learning from their mistakes as they solve problems. Think of this approach as “practice makes perfect.” Students reflect, refine, and revise their work as part of the learning process. Unlike computing and math, educational iterations are not focused on a single, predetermined “right” solution. The iterative approach in education creates a culture that fosters a continual cycle of learning.
Legal Systems: The foundation of common law is stare decisis, the concept that rulings are based on previous legal decisions. Stare decisis is Latin for “to stand by things decided.” Following the iterative process model, this means that any legal precedent builds and rests upon previous decisions.
Evolution Science: The fossil record shows the iterative process record. Repeated evolution over generations shows a change in characteristics of biological species. Some characteristics become more common or rarer as parents pass genes to offspring.
How Does the Iterative Process Work?
The iterative process involves a continuous cycle of planning, analysis, implementation, and evaluation. Each cycle produces a segment of development that forms the basis for the next cycle of iterative improvement.
You’ll start with initial planning and defining overall requirements. Implement your initial development work, and then refine it by trial and error. Once you complete the first cycle, this work segment forms the next chunk of the project. Each cycle should ideally improve the overall product.
There is no standard pace of development; rather, the product requirements and your resourcing will define how efficiently you move through the iterative cycle.
Step One: Planning and Requirements: In this stage, map out the initial requirements, gather the related documents, and create a plan and timeline for the first iterative cycle.
Step Two: Analysis and Design: Finalize the business needs, database models, and technical requirements based on the plan. Create a working architecture, schematic, or algorithm that satisfies your requirements.
Step Three: Implementation: Develop the functionality and design required to meet the specifications.
Step Four: Testing: Identify and locate what’s not working or performing to expectations. Stakeholders, users, and product testers weigh in with their experience.
Step Five: Evaluation and Review: Compare this iteration with the requirements and expectations.
After you complete these steps, it’s time to tackle the next cycle. In the iterative process, the product goes back to step one to build on what’s working. Identify what you learned from the previous iteration. This iterative development, sometimes called circular or evolutionary development, is based on refining the first version through subsequent cycles, especially as you gather and include requirements. It allows you to remain flexible as you identify new needs or unexpected business issues.
What’s the Difference Between Iterative and Incremental Development?
Iterative development relies on a series of repetitive or cyclical loops in which you revise and improve a product based on feedback from users and stakeholders. While the term can be used interchangeably with incremental development, there’s a fine difference.
With incremental development, you layer in new functionality in small sections with each iteration. But it’s more than simply improving the product through iteration. You methodically add new features with each cycle in a step-by-step process, develop different sections of the product or system at various times and rates, and add them to the full product as you complete them.
Many projects use both approaches at the same time to complete a project. For example, Agile and Scrum methodologies are based on an iterative and incremental approach. They are iterative because one version is refined in subsequent runs. They are incremental because sections of work are delivered throughout the project.
What Is the Iterative Development Process?
Many software and web applications rely on the iterative development process because it gives developers more flexibility. Typically, the system development life cycle (SDLC) uses iterative development alongside incremental development. As the system is developed, SDLC becomes increasingly complex.
This approach offers several benefits. It allows project managers to divide the overall project into smaller chunks that build upon one another, and start with the issues or components that are most critical to the project. It also allows the development team to focus on one part of the project at a time, shielding them from business issues that could divert or distract them. And stakeholders can see concrete progress through the lifecycle of the software application or system.
In the iterative model, development starts with a small set of requirements for a small section of the project, rather than a complete list for the entire project. The team designs, develops, tests, and refines the code in repeated cycles. They can add more features in subsequent cycles until the complete software application is ready to go to market.
Step One: Requirements. Because you are working on one section — or iteration — of the product, you need to identify the software or hardware requirements for this section. Be clear about who is going to use the system and how will they use it. Identify the objective or task of the product you’re developing, especially if you are working from a project control list. Define what qualifies as a successful outcome for this new section of software.
Step Two: Analysis. Review the requirements to identify and resolve any incomplete, ambiguous, or contradictory expectations. Know what data you should include to produce the software and what data the software will output.
Step Three: Design. Use the final requirement specifications to design the software. Construct the schematics that describe the software behavior, business process, interface layout, and functionality.
Step Four: Coding. Using the software design documents, produce the code that creates the output you identified in the requirements and analysis.
Once you complete the first section, seek user feedback to identify any bugs, glitches, or deficiencies. Each cycle (or iteration) relies on feedback and, in some ways, failure — learning from your mistakes. This heuristic planning, or trial-and-error learning, lets you tweak the results until you get the performance you want. Then you can move to the next iteration to add features and functions. Follow this process until the software, application, or system is completed.
The key to success in iterative software development is to be rigorous in validating the requirements and testing. As you develop new iterations, conduct a systematic review to ensure the new versions are compatible with previous iterations. Be prepared for any backward engineering to integrate new iterations with previous versions.
What Is the Iterative Process in Scrum?
The iterative process drives the Scrum methodology of development, where the goal is to deliver maximum business value in the minimum amount of time.
In Scrum, each iteration is called a sprint and typically lasts 2-4 weeks. Each sprint starts with a planning meeting that evaluates the list of tasks, prioritizes them, and identifies the work to be done in the iteration. Then the team finalizes requirements, analysis, and design. They implement and test the work. The iteration ends with a sprint review and retrospective that form the foundation for the next sprint. Scrum blends all the iterative process steps into each iteration.
Like all iterative approaches, Scrum recognizes that the goals and scope of a project evolve, because customers may not be able to fully identify what they need at the beginning of the project and the team will face challenges that they could not predict at the start of the project. Scrum acknowledges that you cannot identify the full scope at the beginning and offers a framework that allows the team to nimbly adapt to changes.
What Is the Iterative Design Process?
A variation of the iterative model, the iterative design process allows designers to create, test, analyze, and refine ideas quickly during any phase of the design process.
Designers start with a solution for their product based on initial observations and research. They can then explore and validate their assumptions through the process of ideation, prototyping, testing, analyzing, and refining. This process, called rapid prototyping or spiral prototyping, allows designers to find success more quickly and involve stakeholders and clients more effectively.
Step One: User Observation and Research. The first question designers ask: "What is the problem we are trying to solve?" You must understand users and analyze their habits to know what you will design and develop. User research, such as focus groups, interviews, or A/B testing, will help you identify user needs and behavior, as well as offer insight into the emotional reactions to your product. This provides the user context you need to begin brainstorming solutions.
You'll learn more about how the customer engages with the product, environmental information (such as where and with whom they use it), and other resources they might have around them. You might consider creating flowcharts, diagrams, or other visual helpers to capture information. Words may not be able to describe the problem the way a picture can.
Step Two: Ideate. In this phase, you and your team can start generating ideas that address the problem you identified through your research. Armed with information about users, their emotional approaches to your product, and the context in which they use it, try brainstorming to explore solutions that delight your users. As you develop ideas, make sure you are thinking about the meaning — the "why” — of your product. What will speak to users? What will delight them? You want to articulate the soul of your product using thoughtful, strategic design.
Step Three: Prototype. In this "build" phase, your team creates an early example of your product. You will use this to test concepts or systems that you are considering for the final design. The prototype should outline how the product will be used, whether it's basic software functionality, wireframes, or even paper mock-ups of your visual design. These are easier and cheaper to produce than an entire system, and they allow you to test and refine your design quickly.
Step Four: Analyze. It's time to gather feedback about your design. Ask users to test the prototype, gather feedback on what works and what doesn't, and evaluate how well your assumptions met their needs. Usability testing will help validate your assumptions about your design and guide the next iteration. Take what you've learned, amend your design, and start the next design cycle.
Research proves the iterative design process works. Peter Skillman, of Palm, Inc., invented the Marshmallow Challenge, which started as a team-building exercise to offer lessons in collaboration and creativity. In the challenge, teams are give 20 sticks of spaghetti, one yard of string, one yard of tape, and one marshmallow. They have 18 minutes to build a tower that will hold the marshmallow on top. The team that builds the tallest tower wins.
A group of kindergarteners and a group of business school graduates faced off in the first challenge. The kindergarteners won. Time after time, numerous teams have repeated this challenge. he kindergarteners continue to win. It seems they are more inclined to start with a simple design, test the prototype, and iterate until they find a solution that works. The adults tend to spend their time competing for leadership of the project, arguing about the right plan, and then running out of time to build the tower. Tom Wujec from Autodesk talks about the research in this TED talk.
By focusing on how users think and behave, iterative design helps design teams reduce usability issues, especially in the early stages of a project when the cost of eliminating mistakes is modest. You can improve your effectiveness when you clearly organize assets, protocols, and documentation for each iteration. But the iterative design process has some limitations, and product development should include other skill sets.
Iterative design focuses on incremental or cyclical improvement. If you are looking for true innovation, consider using the strategic design process to help you discover the unmet needs of users.
The user feedback you gather in the iterative design process can help you identify usability problems. But users rarely can tell you how to fix those issues because they may not know the answer. Iteration based on user feedback will help you tweak, but may not solve the problem. As you iterate, include interaction design experts to help you create products that feel intuitive.
Some aspects of product design, such as color, form, typography, and wording create an emotional connection with users. Designers who specialize in these features know how to create a delightful product experience. This is a unique skill that cannot be duplicated through the iterative design process.
Many companies experiment with design Darwinism, the idea that products, like evolution, will cycle through a series of incremental changes and nearly design themselves. Iterative design supports rapid and responsive design. But without other layers in the process, such as strategic design and interaction design, these products may fail because they don’t address the real needs of users. Companies such as Google and Microsoft experiment with creating new products solely through design Darwinism. Your company will benefit if product planning includes strategic design to create something that is meaningful to users.
What Is a Non-iterative Process?
The non-iterative process takes an unmodified Waterfall approach to product design and development. Progress flows linearly to deliver one complete product. The planning, requirements, specifications, and documentation during the project are highly controlled.
The non-iterative process also assumes that requirements can be locked down, which makes it difficult (and costly) to handle changes. It doesn’t contain a feedback loop that uses testing and feedback to refine the product. Instead, progress is measured by the completion of intermediate work products. By comparison, the iterative approach produces smaller cycles or iterations, is highly flexible and adaptable, and regularly delivers work products.
For example, most construction and architectural projects rely on a non-iterative process. First, you gather requirements, and then plan the design and break it up into phases. After you complete construction, you review the structure for safety and maintain it. The plan is fully mapped out at the beginning of the project, and most changes are costly, either in time or money.
Iterative Architecture Process
Rather than focusing on a final, completed project, work in iterations that focus on completed subparts. Designers can apply the lessons learned in each subpart to subsequent iterations. In effect, the overall architecture will respond better to changes.
The iterative architecture process allows designers to overcome one of architecture’s major challenges: creating a complete plan at the beginning of the project. Creating a finished plan can be difficult because project plans cannot shift easily to accommodate changing market forces or business needs. That being said, when applied to architecture, the iterative process helps to defeat this obstacle.
What Does an Iterative Process Mean in Research?
Research methods are inherently iterative. Scientists design a hypothesis, test it, note their conclusions, and then use those insights to influence the next hypothesis. Researchers begin with an observation and verify their findings, which lead to new research questions.
What Is the Iterative Process in Qualitative Research?
Qualitative research is a process that gathers data through observations, interviews, surveys, and documents to find the "human" side of a topic. Because this approach involves people's opinions, values, and behaviors, it requires the flexibility of the iterative process.
Researchers adjust their methodologies, data collection processes, and questions during the project based on what they learn. For example, a researcher may revise interview questions based on the responses of previous subjects. The researcher may categorize responses using one framework early on, but may adjust the categories as data reveal more context or shades of meaning. Qualitative researchers in social sciences, healthcare, education, and business typically utilize the iterative process.
What Does the Iterative Process Mean in Psychology?
Psychological assessments are iterative. Information is gathered and evaluated, symptoms are identified, and outcomes or treatments are suggested. Clinicians later evaluate their patients’ changes and adjust treatments. Iteration occurs by reviewing the data and connecting it to emerging insights.
The iterative reprocessing model hypothesizes that emotional episodes (such as anger or joy) are created as information flows through certain centers of the brain. The brain integrates a person’s goals, values, desires, and intentions to shape an emotional instance. Different patterns of neural activity can create different emotional events.
At the heart of all science is the iterative process, with the goal of getting closer to the truth through research over time. Research relies on the credibility of previous findings so that iteration can occur. According to recent article in Nature, psychology faces a replicability crisis. Researchers tried to reproduce the findings of 21 experiments, but had limited success. Other factors may be at work, however. Because it involves human emotion, psychology is extremely complicated and experiments can be difficult to duplicate.
Challenges of the Iterative Process
One major threat to the iterative process is scope creep, which can occur due to the lack of firm deadline or defined solution. To avoid scope creep, project managers should maintain a list of priorities and benchmarks.
Timelines can also be a challenge. Team members want to keep tweaking to improve a product. This can make stakeholders and clients restless. Project managers may need to pay more attention to the team, be firm about the requirements that need to be addressed in the iteration, or be willing to say a product is good enough for testing and evaluation.
If you are working with vendors on any project, you need to be clear about the requirements and timeline. In the iterative model, clear contracts and expectations are essential. At the beginning of any contract, ensure that you have a hourly or market rate if the project veers too far off scope or time. Even if the project changes for good reasons, both you and the vendor must be willing to be flexible rather than stick to a schedule or scope that defeats the whole point of iterating.
originally seen on SmartSheet.com
selected, reviewed and vetted by
