ROBATZ editorial · · 6 min read

The short answer. A Rails upgrade costs more the further your app is from a supported version, the less its critical journeys are covered by tests, the more abandoned or patched gems it depends on and the more tangled its most-changed code is. Application size matters less than most people expect. Ask for an assessment that measures these factors before you accept a fixed estimate.

The version gap

The official Rails upgrade guide recommends moving “one minor version at a time,” fixing deprecations at each step. So the first cost driver is simply the number of steps. An app on 7.2 is two steps from 8.1; an app on 6.1 is five (6.1 → 7.0 → 7.1 → 7.2 → 8.0 → 8.1). Each step means updating dependencies, running the update task, fixing what breaks, testing and releasing.

Not every step is equal. Some versions carry larger changes: for example, Rails 7.0 requires applications to use the Zeitwerk autoloader, and changes the default key generator digest, which needs a cookie rotation plan to avoid logging users out. FastRuby, a company that publishes data from its own upgrade projects, reports that upgrade costs generally go down the closer the target is to the latest version, with the 6.1 → 7.0 jump as an exception. That is one vendor’s dataset, but it matches the shape of the guide: older starting points carry more accumulated change.

Support status adds urgency rather than effort. As of November 2026, only Rails 8.1 receives security fixes (maintenance policy), so any older app is on its own for new vulnerabilities until it gets there.

Test coverage of critical journeys

Tests are how an upgrade team knows that a change did not break anything. The Rails guide is blunt about the alternative: without good automated tests, you’ll “need to spend time manually exercising all the parts that have changed,” which in a Rails upgrade means every piece of functionality.

What matters is not a single coverage percentage but whether the journeys that make money or carry risk are protected: signup and login, checkout and billing, permissions, data exports, integrations. If they are not, the estimate should include writing those tests first. That work is not overhead; it is what makes the rest of the upgrade predictable, and it keeps paying off after the project ends.

Dependencies and local patches

Every gem has to work with each new Rails version. The cost rises with:

  • Abandoned gems that need a replacement, a fork or removal.
  • Forked or pinned gems that the team maintains itself.
  • Monkey patches that change framework or gem behavior, especially untested ones. They tend to break silently when the code they patch changes.
  • Gems tightly woven into the domain, such as an old authorization or state machine library used across hundreds of files.

FastRuby lists monkey patching, forked and outdated gems, and incompatible dependencies among the factors that raise upgrade effort. A dependency inventory with a status for each gem is usually the single most useful page of an upgrade assessment.

Code complexity, not just size

A large, well-structured app can be cheaper to upgrade than a small one with a few very complex, poorly tested files. FastRuby makes the same point from its project data: complexity doesn’t necessarily relate to application size. Tools that combine complexity and coverage, such as Skunk, help locate the files most likely to cause trouble. Ask a supplier where the risk is concentrated, not only how many lines of code you have.

Ruby, infrastructure and frontend

A Rails upgrade often brings other upgrades with it:

  • Ruby. Rails 8.0 and 8.1 require Ruby 3.2 or newer. The guide recommends upgrading Ruby and Rails separately, so an old Ruby adds its own step.
  • Infrastructure. Base images, database versions, background job systems and deployment tooling may need updates to support the new versions.
  • Frontend. Apps that rely on older asset pipelines or JavaScript build setups may need changes. New Rails 8 apps default to Propshaft, while existing apps can keep Sprockets and migrate later, which keeps that work optional.

An estimate should say which of these are in scope and which are assumed to be handled by your team.

How the work is organized

The delivery approach changes both cost and risk:

  • Incremental releases. Releasing each version step to production keeps every change small and easy to roll back. It can mean more release cycles but fewer surprises.
  • Dual boot. Running the current and next Rails versions side by side in CI (for example with next_rails) lets upgrade work merge continuously instead of living on a long branch.
  • Feature work alongside. Continuing product development during the upgrade is usually possible, but it needs coordination; a feature freeze is cheaper on paper and more expensive for the business.
  • Your team’s involvement. Access to people who know the business rules, QA support and quick code review all shorten the project.

Questions to ask about an upgrade proposal

  • Which Ruby and Rails versions are the start and target, and how many steps are planned?
  • Which critical journeys are covered by tests today, and is writing missing tests included?
  • Which gems need replacement or upgrades, and which are the riskiest?
  • Is each step released to production, or is everything delivered at the end?
  • What is assumed about infrastructure, frontend and your team’s availability?
  • What happens to the estimate if an assumption turns out to be wrong?
  • What will be documented and handed over when the work is done?

If two proposals differ a lot, compare these answers before comparing totals. Often one of them includes test writing, dependency replacement or infrastructure work that the other leaves to you. For the broader question of whether to upgrade at all, see Rails upgrade vs. rewrite: how to decide; for how custom software estimates work in general, see what drives the cost of custom software development.

Frequently asked questions

How much does a Rails upgrade cost?

It depends mainly on how many versions you need to move through, how well critical journeys are covered by tests, how many dependencies need replacing and how complex the most-changed code is. A short assessment that measures these factors is the reliable way to get an estimate for your application.

How long does a Rails upgrade take?

The duration follows the same factors as cost: the number of version steps, test coverage, dependencies and complexity, plus how quickly changes can be reviewed and released. Releasing each step separately makes progress visible from the first weeks.

Is it cheaper to upgrade Rails regularly?

Generally yes. Staying close to the latest version means fewer steps, smaller changes per step and fewer abandoned dependencies. Vendor data also suggests that upgrades get cheaper as the starting version gets closer to the latest one.

Can we upgrade Rails without good test coverage?

Yes, but the uncertainty and the manual testing effort go up. Most plans add tests for the critical journeys before the first version step, which reduces risk for the rest of the project.

Does application size determine upgrade cost?

Less than people expect. A large, well-tested, well-structured app can cost less to upgrade than a small app with complex, untested code and many patched dependencies.

Apply this to your product

Explore our Ruby on Rails development and software modernization services for scope, deliverables and the decisions to make before starting.

Further reading