Product Analysis Lead

Our client is a well-established charity that mobilises volunteers to support individuals, local communities and public health services across the UK.

They have developed a volunteering platform. Your role's purpose is to:

• lead the Product Analysis function for this platform
• be the driving force behind a specific shift in how the team works: from "AI helps" to "AI-first"

Our client's product practice today is "AI helps" — analysts still write tickets and specs largely by hand, with AI used to assist. You are expected to bring in "AI-first," and be the example for your small team to see. In "AI-first" the idea should always be prompt-engineering first — you develop detailed prompts and context, then AI produces the first draft of tickets, specs and Confluence pages, and you and your team own the reviewed output. You and your team are still fully accountable for everything AI produces under this role's direction — a well-crafted prompt doesn't reduce ownership of the output, it's simply how the output now gets made.

You have already lived this way of working elsewhere — building prompt-first habits into your own practice and having taught it to others — and will bring that direct, hands-on experience to this organisation rather than learning it here.

You will line-manage two Product Analysts and a Designer working on the platform, actively coaching and retraining them from their current habits into prompt-first production, working closely with the CTO to embed the change.

Central to the client's shift to AI-assisted + TDD delivery: ensures specs, acceptance criteria and architecture are agreed once, upfront, so decisions aren't relitigated mid-sprint. The Engineering team is similarly making the change to "AI-first."

What you will be doing

Leading by Example for the Shift to Prompt-First Practice

• Introduce and embed a prompt-first working model for product analysis, where AI drafts tickets, specs and Confluence pages from detailed prompts and context, replacing today's largely manual, AI-assisted approach
• Work with the CTO to define what "good" looks like for the organisation's prompt-first practice, using your own prior experience as the reference point
• Retrain the existing Product Analysts from their current way of working, moving them from occasional AI assistance to prompt-first production as their default
• Track and report progress on the team's adoption, addressing resistance or bad habits directly

Prompt-First Ticket, Spec & Documentation Production

• Build detailed prompts and context packages (requirements, constraints, prior decisions, design links, existing Confluence content) for AI to draft tickets, specs and Confluence pages
• Direct AI to produce the first draft of feature specs, acceptance criteria and story tickets, rather than writing them from a blank page
• Review, correct and finalise AI-drafted output before it goes to Engineering, Design or QA — you own it, not the AI
• Continuously refine your prompts and context packages as you learn what produces strong first drafts, and share what works with the team

Feature & Story Delivery Ownership

• Own Product's stages of the Feature delivery flow: idea formulation, kick-off, and breakdown of features into stories
• Own Product's stages of the Story delivery flow: ticket and feature writing, product review, and planning with Engineering and QA
• Fit AI-drafted tickets, specs and documentation into the organisation's existing JIRA/Confluence process and delivery cadence

Specification & Requirements Discipline

• Ensure specs, acceptance criteria, architecture and data structure are agreed before coding starts, not revisited mid-sprint
• Give early warning of next-quarter product features and keep the sprint backlog scoped to what's actually being built next
• Use AI-assisted feasibility checks to gauge whether tickets are achievable within a sprint before presenting them

Team Leadership & Coaching

• Line-manage the Product Analysts, setting priorities and reviewing their output
• Personally teach the team how to write prompts and assemble context that get AI to a strong first draft — not delegate this training to anyone else
• Set the review bar: nothing AI-drafted goes out to Engineering, Design or QA without the analyst's review and sign-off
• Be available for quick check-ins and give complete, thought-through feedback

Sprint Planning & Cross-functional Collaboration

• Work closely with Engineering, Design and QA through planning, sprint build and review
• Accept and act on feedback from Engineering and Design; don't avoid conflict where something doesn't add up

What you need to know

Knowledge

• Product analysis and requirements-gathering for an in-house built SaaS/web application
• How to construct detailed, effective prompts and context packages that get AI tools to a strong first draft of a ticket, spec or Confluence page
• How to move a team from ad hoc, assistive AI use to a disciplined, prompt-first way of working, including what typically goes wrong in that transition
• Writing acceptance criteria, user stories and specifications to a standard that supports test-driven, AI-assisted development
• Agile/Scrum delivery, including sprint planning and backlog management
• How to critically review and correct AI-drafted output before it's relied on by Engineering, Design or QA
• JIRA and Confluence, ideally with AI tooling integrated into either

Skills

• Prompt-writing and context-assembly for AI-drafted tickets, specs and documentation — this is a primary, not secondary, skill for the role
• Coaching and retraining a team from an established way of working into a new one, with patience for the transition but clarity about the destination
• Precise editorial judgement — able to spot gaps, ambiguity or errors in AI-drafted specs and tickets before they reach the team
• Stakeholder management with Engineering, Design and QA
• Line management
• Prioritisation and backlog grooming
• JIRA, Confluence, and AI drafting tools
• Facilitation of product reviews and cross-functional planning sessions
• Clear written and verbal communication, including giving direct, constructive feedback

Experience

• 5+ years in product analysis, business analysis or product ownership roles, ideally on an in-house built product
• Direct, hands-on, prior experience working in a prompt-first model — personally using structured prompts and context to have AI draft tickets, specs and documentation, with accountability for the reviewed output — not just familiarity with AI tools in general
• Experience leading a team through the transition from manual or AI-assisted work to prompt-first practice, not just operating within an already-established one
• Experience leading or mentoring a small team of analysts
• Experience operating in a spec-first, AI-assisted delivery environment
• Track record of writing specs/tickets that reduce mid-sprint scope churn