Skip to main content

What are you looking for?

Explore our services and discover how we can help you achieve your goals

Client Workspace

Trunk-based development or long-lived branches: delivery trade-offs for release frequency, review load and incidents

Trunk-based development suits a team that has fast automated tests and ships one live version, because small daily merges raise release frequency and shrink each review; long-lived branches still fit products with several supported versions or mandated release approval. This Netbase guide compares both by review load and incident rate, with a decision matrix and switch steps.

Book a solution review See the related solution

By Netbase Commerce and Platform Engineering Desk · Reviewed by David (CEO) · Updated · 10 min read

star

This guide is for a CTO, head of engineering or product owner who decides how a team ships changes, or who reviews how an outsourced team does it. It sits under cloud platform engineering, and the pipeline work it often leads to is DevOps and platform engineering.

In this guide

What is the real difference between the two models?

Both models use branches. The difference is how long a branch lives before it reaches the main line. In trunk-based development, developers merge small changes into one shared trunk at least once a day, and a branch typically lasts a few hours. DORA's capability guide adds three or fewer active branches and no code freezes or integration phases.

In a long-lived branch model, work stays on a feature, develop or release branch for days or weeks and is merged when it is finished. The best-known form is git-flow, published in 2010 with feature, release and hotfix branches around master and develop. Its author noted in 2020 that teams doing continuous delivery should prefer a simpler workflow, while git-flow still fits versioned software that must support several versions.

So the choice is not about tools. It is about batch size: how much unmerged change a team carries at any moment, and how late it learns that two changes do not fit together.

How does each model affect release frequency, review load and incident rate?

The three effects follow from batch size. The table states the mechanism; it does not promise a number, because the result depends on the team's tests, build time and release process.

Effect Trunk-based development Long-lived branches Choose it when
Release frequency Trunk is always releasable, so a release is a decision, not a project Releases wait for a branch to finish and be merged Trunk when the business wants small, frequent releases
Review load Many small reviews, each read in minutes, but reviews must be fast Few large reviews that are slow to read and easy to rubber-stamp Branches only when a formal review board must sign off a whole change
Merge conflicts Small and found the same day Grow with branch age and team size; found at merge time Trunk when several developers touch the same code
Incident rate A faulty change is small and easy to locate and roll back, if tests run before merge A large merge can carry several faults at once, but trunk without tests ships faults faster Trunk only after automated tests gate every merge
Unfinished work Hidden behind feature flags or built behind an abstraction Isolated on its branch until done Branches when flags are not yet managed

The incident row is the one teams misread. Trunk-based development lowers the risk per change, because each change is small. It raises the number of changes that reach production, so without tests and a fast build it can raise incidents rather than lower them. The order matters: tests first, shorter branches second.

Which model fits your team? A decision matrix

Score the six factors below. If most rows point to the same column, that is the default; a single "branches" row on supported versions or mandated approval is enough to keep a release branch, even when the rest points to trunk.

  • Supported versions

    Points to trunk-based
    One live version, as in a web or SaaS product
    Points to long-lived branches
    Several customer-installed versions maintained in parallel
  • Automated tests

    Points to trunk-based
    Tests run on every merge and are trusted
    Points to long-lived branches
    Most testing is manual or happens at the end
  • Build and test time

    Points to trunk-based
    A few minutes
    Points to long-lived branches
    Hours, or only overnight
  • Team size on one codebase

    Points to trunk-based
    Several developers changing shared code
    Points to long-lived branches
    One or two developers per isolated module
  • Release approval

    Points to trunk-based
    Approval per change, or automated checks
    Points to long-lived branches
    A regulator or customer signs off a whole release
  • Type of change

    Points to trunk-based
    Steady product work in small slices
    Points to long-lived branches
    A rare, sweeping change that cannot be split

A common middle path is trunk-based development with short release branches cut just before a release and only receiving fixes. It keeps daily integration and still gives a stable line for a scheduled release.

When should a team not move to trunk-based development?

Do not shorten branches before the safety net exists. A team without automated tests on every merge will ship faults faster, and a team whose build takes hours will queue merges and recreate long branches by accident.

Keep long-lived release branches when customers run several versions you must patch, such as on-premise or customer-installed software. Keep a release branch when a regulator or a contract requires sign-off of a whole release; trunk can still be the development line behind it.

Finally, do not switch in the middle of a fixed release with a hard date. Change the branching model at the start of a cycle, so the team learns it without a deadline on top.

How to move from long-lived branches to trunk

  1. Measure the starting point

    Record active branch count, branch age, review wait time and the delivery metrics the team already tracks.

  2. Make the build fast and trusted

    Get the test suite running on every merge request in a few minutes before changing any branch rule.

  3. Split work into small slices

    Plan stories that can merge within a day; a change too large for a day is split, not parked.

  4. Introduce feature flags with an owner

    Hide unfinished work behind flags, and remove each flag once the feature is live.

  5. Use branch by abstraction for large rewrites

    Build the new part behind an interface on trunk instead of on a long branch.

  6. Set a review time rule

    Agree that small reviews are picked up within hours, so short branches do not wait for days.

  7. Cut short release branches only if needed

    Keep them for fixes only and delete them after the release.

  8. Re-measure after a few cycles

    Compare branch age, review wait, release frequency and change failures with the start.

Branch by abstraction is described on trunkbaseddevelopment.com, and Martin Fowler's branching patterns article explains why integration frequency, not the branch itself, drives merge pain. To judge the switch with numbers, the guide on measuring engineering delivery health explains which metrics to pick and how a buyer reads them.

Where AI fits

AI helps most with the review load. An AI assistant can summarise a small merge request, point out missing tests and flag a change that touches a risky module, so the human reviewer starts from a prepared view. Netbase works with the major commercial and open-source models and chooses them per project.

Engineers still read every change, decide whether it merges and own the release. An AI summary does not replace the automated tests that make trunk-based development safe.

What delivery record exists, and what does not?

Netbase manages source code in GitHub and runs projects in a four-step workflow of investigation, planning, coding and testing, and delivery and maintenance, with progress that clients can check at any time. Its listed security practices include secure code review and version control.

Netbase works as offshore development partner on a multi-tenant cloud ERP delivered as SaaS for a US client, not named, and builds and runs its own SaaS products, such as Cloodo Workspace. Delivery follows a six-step lifecycle with weekly reviews and KPI dashboards. Most Netbase projects are delivered on fixed-price contracts.

No published Netbase record states which branching model a given project uses, and no release-frequency, review-time or incident result is claimed for any project.

Alternatives: other ways to reduce merge and release pain

GitHub flow is the lightest branch model: one main branch, short feature branches with a pull request, and a deploy after merge. It is close to trunk-based development when branches stay short, and it fits a small team with one live version.

Git-flow keeps develop, release and hotfix branches. It fits versioned products with several supported releases and costs more merge work on a product with a single live version.

If the real pain is risky releases rather than merges, the fix may sit elsewhere: a release gate, as in the DevOps maturity assessment scorecard, or a tested rollback plan, as in the cloud migration cutover plan.

What are the limits of this guide?

  • The matrix fits web, SaaS and business applications; embedded, game and mobile store releases have other constraints.
  • The effects are mechanisms, not measured results; a team should compare its own numbers before and after a switch.
  • Feature flags add their own debt; a flag inventory with owners is part of the cost of trunk-based development.

Plan the next step with a Netbase consultant

Frequently asked questions

No. Most teams still open short branches for a pull request and review. The rule is that a branch merges into trunk within a day or so, and no long-running develop or feature branch sits beside it.

No. Trunk stays releasable, but the team can still choose when to release, and unfinished work stays hidden behind feature flags until it is ready.

Yes, if both teams merge small changes through the same pipeline and the same review rules. Ask the vendor to show its branch age and review times before the switch.

Not for every product. Its author recommends a simpler workflow for continuous delivery, and git-flow still fits software that must support several released versions at the same time.

Next step

If you want a second reader for your branching and release rules, book a solution review and bring your branch age and review times.

AI-ready cloud platform engineering for SaaS and commerce workloads AI-ready cloud platform engineering for SaaS and commerce workloads

Netbase provides cloud platform engineering for SaaS and commerce teams that need a secure, repeatable home for their software and its AI features, on AWS, Google Cloud, DigitalOcean or Cloudflare. We design the architecture, write the environments as code, connect managed and AI services and hand over a platform your team can run, built the way we run ours.

Learn More
line
AI-assisted cloud migration with a cut-over plan and a way back AI-assisted cloud migration with a cut-over plan and a way back

Netbase provides cloud migration services for companies running business applications on ageing servers or hosting to move them to AWS, Google Cloud, DigitalOcean or Cloudflare without losing data or control. AI speeds up the code and dependency assessment; every migration still has an inventory, a rehearsed cut-over, cost controls from day one and a tested rollback point.

Learn More
line
AI-assisted DevOps and platform engineering for routine releases AI-assisted DevOps and platform engineering for routine releases

Netbase provides DevOps and platform engineering for SaaS and commerce teams to make releases routine instead of risky. We set up CI/CD pipelines, repeatable environments and observability on AWS, Google Cloud, DigitalOcean or Cloudflare, with AI assistance for tests, reviews and incident triage, so every change ships the same way and problems surface before customers notice.

Learn More
line
SaaS product accelerator: launch an AI-ready SaaS on proven Netbase modules SaaS product accelerator: launch an AI-ready SaaS on proven Netbase modules

A SaaS product accelerator is a set of reusable Netbase modules for accounts, billing, roles and integrations that helps founders and product teams launch subscription software faster, with room for in-product AI from the first release. Reusing these modules can shorten development time, and the approach is proven on Printcart, Netbase's own web-to-print SaaS.

Learn More
line

Work with Netbase remotely

Every Netbase engagement is delivered remotely, so clients in any country and any industry can work with us; Netbase has worked with clients in 80 countries. Contracts and invoices are issued by Netbase JSC, and payment is made by bank transfer to Netbase JSC's company account or PayPal Business. Talk to the CEO, Huy Nguyen (David), on WhatsApp or by email.

Contact Netbase

Discuss a project

Netbase JSC helps organizations design, build, modernize, and operate digital products and AI-enabled business systems.
Project enquiries

project@netbasejsc.com

WhatsApp

+84 937 869 689

Office address

91 Nguyen Chi Thanh, Dong Da, Hanoi, Vietnam

Get in touch

Tell us what you want to build, modernize, or operate.

Tell us what you want to build, modernize, or operate.

Contact Netbase