How Clearer Drove AI Adoption and Lifted Weekly Output 50% While Modernizing Legacy Laravel

Clearer.io builds and acquires e-commerce software that helps Shopify merchants boost revenue and streamline operations. The company brings together multiple specialist apps—spanning reviews, search, loyalty, returns, and more—into one connected platform that merchants use daily. With engineering teams distributed across the world, Clearer.io serves thousands of e-commerce businesses worldwide.

Clearer identified a mission-critical legacy codebase within its portfolio that required modernization—a core service in its reviews platform, built on Laravel 4 and PHP 7, that needed upgrading to the frameworks’ latest versions. The Laravel 4 to 12 jump is documented by Laravel’s official guides as requiring a “full rewrite.” This was a large manual effort the team hadn’t resourced. Years of deferred modernization were blocking feature development on an aging legacy stack.

But modernizing one app alone would not be enough. Clearer’s engineering leadership had already identified AI adoption as a strategic priority and was actively exploring how to embed AI workflows across their teams. If they could pair that initiative with a structured modernization approach, they could tackle this migration—and the backlog behind it—at a pace that a manual effort couldn’t match.

DevClarity partnered with Clearer’s engineering leadership and their distributed team on a full AI Coding Jumpstart—assessing baseline adoption, configuring AI context rules across key repositories, delivering hands-on foundations training across the distributed team, and running deep dives on AI-first modernization and QA/testing workflows.

The engagement then put all of that into practice in a focused, hands-on sessions centered on an app modernization. The DevClarity team paired directly with Clearer developers in laptop-open working sessions. Everything was designed to be reusable—structured workflows, shared context, and documented prompts the broader team could own and extend independently.

The work centered on building what the team called a “Modernization Factory”—a set of repeatable AI workflows and context documents committed directly to the codebase:

• • Custom migration commands. Created 7 reusable commands implementing a multi-step workflow: identify files → generate explainer → plan migration → execute. These commands could be adjusted and re-run when issues arose, creating an iterative approach rather than fingers-crossed, one-shot AI generation.
• • Migration context documents. Developed comprehensive docs capturing Laravel and PHP upgrade gotchas, breaking changes, and official guide links. These serve as high-quality, reusable AI context for current and future migrations.
• • Unit test generation framework. Established patterns for validating legacy code comprehension. The AI quickly generated ~30 tests covering a controller, token handling, and invitation flow—tests that passed and self-corrected PHPUnit compatibility issues automatically.

Throughout the engagement, the team learned that building context upfront creates a more reliable, repeatable process. It also helps the team avoid troubleshooting inconsistent AI output. Each solved challenge became a documented solution in the growing modernization playbook.

By the end of the engagement, 65% of the team now uses AI coding tools every day. That adoption translated directly into velocity: developers estimated an increased speed on repetitive tasks of +66% and increased weekly output of +50%.

But the real impact shows in the workflow:

• • Teams now have a repeatable modernization factory living in their repos—7 commands, migration docs, and proven patterns ready to apply to other legacy services.
• • AI successfully self-corrected errors during migration—creating missing database models, fixing view adaptation issues, and adjusting PHPUnit compatibility automatically.
• • AI-assisted testing became part of the workflow, with developers using AI to generate and validate unit tests—accelerating a practice the team had deprioritized during years of feature-focused development on the legacy stack.
• • Artifacts and workflows are committed to version control, enabling knowledge transfer across the distributed team—including Copilot-compatible instructions for engineers using different AI tools.

What Clearer received. A complete Modernization Factory: 7 reusable commands, migration context docs, clean room methodology, and unit test generation patterns. These practices live inside their codebase and are ready to scale across their backlog of legacy applications. Clearer’s entire engineering team walked away with hands-on AI coding training, shared context rules across their repositories, and proven workflows they can apply to any project. What once demanded months of manual effort can now be tackled systematically, by a team equipped to use AI as a force multiplier.