What Is Agile? When It Fits Your Project and When It Doesn't
What is agile? Learn how sprints work, compare Scrum, Kanban and waterfall, and see with a worked example when agile fits your software project.
What is agile? Agile is a way of building software in short cycles, showing real working output to users after each cycle, and letting what you learn change the plan. The point is to find out sooner when an assumption is wrong. It is not a promise of faster delivery or a licence to skip planning.
I’m Aditya S, Business Development Manager at Synetica, a tech-based business consultancy combining strategy and execution for Dreamers and Growing Businesses. This guide explains agile through a fictional 12-week project you can check line by line. The numbers are teaching assumptions, not Synetica client results or industry benchmarks.
You’ll learn what agile means, how a sprint works, how Scrum and Kanban differ, when agile is the wrong tool, and what to ask a vendor who says “we work agile.”
TL;DR — the whole guide in five lines
- Agile is a feedback habit, not a ceremony set: the value is users seeing working software every few weeks.
- It moves the first honest answer forward: the project is not shorter, but wrong assumptions surface in week 2 rather than week 12.
- Scrum and Kanban are two ways to run it: Scrum fixes time, Kanban limits work in progress.
- Fixed scope and fixed price are a poor match: agile flexes scope, so agree the budget and the review rhythm instead.
- Ask to see the last review: a vendor who cannot show one is probably running short waterfalls.
What is agile, and what does the Agile Manifesto say?
Agile is an approach that prefers working software, frequent customer feedback and adapting to change over following a fixed plan to the end. The word comes from the Agile Manifesto, written in 2001 by 17 software practitioners. It lists four values: individuals and interactions over processes and tools, working software over comprehensive documentation, customer collaboration over contract negotiation, and responding to change over following a plan.
Read the “over” carefully. The Manifesto does not say documentation, contracts or plans are worthless. It says that when they conflict with working software and real feedback, the second group wins.
Think of it as the difference between mailing a tailor your measurements and getting a fitting every two weeks. Both can produce a suit. Only one lets you say “the shoulders are wrong” before the cloth is cut.
| Term | What it is | Example |
|---|---|---|
| Agile | A set of values and a feedback habit | “We review working software with users often.” |
| Scrum | One framework for practicing agile with fixed-length sprints | Two-week sprints, a sprint review, a retrospective |
| Kanban | A flow method that limits work in progress | A board where “In progress” holds at most three items |
| Waterfall | A plan-first sequence: specify, build, test, release | Users first see the product at release |
| Sprint | A fixed-length cycle, often one to four weeks | Two weeks ending in a working increment |
| DevOps | Practices for releasing and running software reliably | Automated tests and deployment |
Agile and DevOps often appear together, but they answer different questions. Agile asks what to build next; DevOps asks how to ship it safely.
Decide what you need to learn first with our MVP guide.
Why does agile matter for your business?
Agile matters because the most expensive software mistake is building the wrong thing confidently. A plan-first project hides that mistake until release. A sprint-based project gives it several chances to appear while it is still cheap to correct.
Consider a fictional field-service scheduling app for an air-conditioner repair company. The team has four people and 12 weeks. The plan assumes technicians have a stable internet connection on every job. In reality, many work in basements and new buildings with no signal.
In the plan-first version, the team finds this at release in week 12. In the sprint-based version, the first technician who tries the app after sprint 1 finds it in week 2.
Assume 25% of the work built so far depends on the wrong connectivity assumption. A four-person team builds 4 person-weeks per week. By week 12 that is 48 person-weeks built, so 48 × 25% = 12 person-weeks of rework. After sprint 1 it is 8 built, so 8 × 25% = 2.
The same mistake costs six times less because it surfaced earlier. That is the business case. It ignores lost customers and delayed launch, which would widen the gap, and it is an invented example rather than a measured rate.
Open the rework diagram at full size. If your project has no assumption you are unsure about, you need agile far less.
How does agile work in practice?
Agile works as a loop: rank the work, build a small slice, show it to real users, then change the plan. A sprint is one turn of that loop, usually two weeks long.
- Backlog: a ranked list of work, ordered by user value rather than by who asked loudest.
- Sprint planning: the team picks only what fits the fixed time.
- Build and test: each item must work end to end before it counts as done.
- Review: real users or their representatives see working software and react.
- Retrospective: the team fixes how it works, such as slow hand-offs or unclear tickets.
The review is the step people skip. Without it, you have a series of short waterfalls: planning, building and releasing in two-week boxes with nobody reacting to the output.
In the repair-company example, sprint 1 delivers one thing: a technician can open a job and mark it complete on a phone. It is small and a little ugly, and it lets a real technician try it in a basement. Testing flows this early is also where UI/UX design pays off most.
Open the sprint loop at full size. Notice that the review feeds the backlog. The next sprint starts from evidence, not from the original wish list.
What types of agile are there: Scrum, Kanban or something else?
Scrum and Kanban are the two most common ways to run agile, and they suit different kinds of work. Pick by the shape of your work, not by what sounds modern. For the roles, events and costs of the most common one, see what Scrum is and how it works.
| Approach | How it works | Best for | Watch out for |
|---|---|---|---|
| Scrum | Fixed sprints, defined roles, review and retrospective each cycle | New products with changing scope | Ceremonies without real user review |
| Kanban | Continuous flow, a work-in-progress limit per column | Support, maintenance and steady request streams | No natural moment to step back and re-plan |
| Extreme Programming (XP) | Engineering habits such as pair programming and automated tests | Teams where code quality is the main risk | Treating it as a process rather than a discipline |
| Hybrid (“Scrumban”) | Sprint rhythm with a Kanban-style board | Teams that need both planning and flow | Becoming a vague label for “whatever we do” |
For a new product build, Scrum-style sprints are a sensible default because the fixed review gives you a rhythm. For a team that handles a constant flow of fixes, Kanban usually fits better.
Large-company frameworks exist for coordinating many teams. For a business building its first or second custom system, one small team and one board are usually enough.
Check a vendor’s way of working with our software house guide.
Which projects does agile fit, and which does it not?
Agile fits work where you cannot be sure what users need; it fits poorly where requirements are fixed by law, contract or physics. Be honest about which you have.
| Situation | Agile fit | Why |
|---|---|---|
| New product, unproven demand | Strong | You will learn what matters by showing real software |
| Internal tool replacing a spreadsheet | Strong | Staff can react each sprint and shape the workflow |
| Regulated system with a fixed specification | Partial | Plan the compliance parts upfront, iterate the rest |
| Small fixed-scope task, like a landing page | Weak | Little uncertainty, so the overhead outweighs the benefit |
| Fixed price and fixed scope contract | Poor | Agile flexes scope; the contract forbids it |
The last row causes most vendor disputes. If you sign for a fixed list of features at a fixed price, there is no room for the learning agile exists to produce. Either accept a plan-first approach and spend the time on a thorough specification, or agree a budget and review rhythm and let scope flex inside it.
Validate it in 2 weeks. Blueprint & Prototype puts a working prototype in front of real users and turns what you learn into agreed scope and a cost estimate. The two-week promise covers prototype validation; it does not guarantee a production launch.
At Synetica, Blueprint & Prototype takes 2 weeks, Production Grade Software takes 8 weeks, and Launch & Grow runs as an ongoing monthly retainer. The prototype step is where most scope decisions happen, so the build stage starts from evidence rather than assumptions.
Review how the 8-week build is run.
What does an agile project look like in the real world?
A healthy agile project is easy to recognize by what you can inspect, not by what the team calls itself. Using the repair-company example, here is what you would see in weeks 2, 6 and 12.
| Week | What users see | What changes |
|---|---|---|
| 2 | A technician can open and complete a job | Offline use moves to the top of the backlog |
| 6 | Jobs sync after the technician regains signal | A planned reporting screen is pushed back |
| 12 | Scheduling, offline work and completion reports | Release candidate approved by the dispatch lead |
This is a generic illustration, not a Synetica client result. Notice that the scope at week 12 is not the scope on day one. Offline support was added, and a reporting screen was deferred. That is agile working as intended, provided the sponsor knew that scope would move and agreed the total budget up front.
Compare that with a project that only shows you slides each sprint. A real review involves working software, someone who will actually use it, and a visible change to the backlog afterward. Ask to attend one before you sign.
Bring your rough scope and the assumptions you’re least sure about to a scoping conversation.
What mistakes do teams make with agile?
The most common mistake is adopting the vocabulary without the feedback. Daily stand-ups and story points do not help if users never see the software.
Here are the errors I see most often:
- Skipping the review. Sprints become internal deadlines and nobody outside the team reacts.
- Treating the backlog as a promise. It is a ranked list that should change after each review.
- Using agile to avoid deciding. “We’ll figure it out as we go” is not a strategy. Agree the goal, the budget and the stopping conditions first.
- No single decision-maker. If five stakeholders rank the backlog differently, every sprint ends in debate.
- Fixed scope and fixed price, called agile. Either the scope or the label has to change.
- Never stopping. Iteration is a way to find out when a product is not working, so agree conditions for ending it. Use the product stop framework.
Agile also does not remove the need for documentation. Teams still need to record decisions, data definitions and how the system runs. They write what is useful, not everything possible.
What would I tell a founder deciding whether to work agile?
Bring your biggest unknown, one decision-maker and a total budget. If you can name an assumption that would change the product when proven wrong, agile will probably pay for itself. If you cannot, you may need a clear specification more than a sprint rhythm.
For a Growing Business replacing a manual process, involve the staff who do the work in every review. For a Dreamer testing a new idea, keep the first cycles small and measure what users actually do. In both cases, ask any vendor to show you their last sprint review before you commit.
My next step would be to book a scoping conversation with your riskiest assumption, your budget range and the person who will make decisions between sprints.
FAQ — Understanding Agile
What is agile in simple terms?
Agile is a way of building software in short cycles. After each cycle, real users see working output and their reaction changes the plan. It trades a fixed upfront plan for earlier learning.
What is an example of agile?
A team builds a scheduling app in two-week sprints. After sprint 1, a technician tries a basic version and reveals that the app must work offline. The team reprioritizes immediately rather than discovering the problem at release.
What is the difference between agile and Scrum?
Agile is a set of values and a feedback habit. Scrum is one specific framework for practicing it, with fixed sprints and defined roles. A team can be agile without using Scrum, for example by running Kanban.
Is agile faster than waterfall?
Not necessarily. The same project can take the same time. Agile’s advantage is that wrong assumptions appear earlier, when they are cheaper to fix. Waterfall can be more efficient when requirements are stable and well understood.
Can I use agile with a fixed price?
Rarely well. Agile flexes scope, while a fixed price with fixed scope does not allow it. A more workable arrangement is a fixed budget and review rhythm, with scope adjusted inside it after each review.
Related Posts
Aditya S
Business Development Manager, Synetica.
Put it into practice
Need help putting this into practice?
We turn the ideas in this article into a prototype tested with real users in 2 weeks (from Rp49jt)—or production-grade software in 8 weeks (from Rp100jt).