Business Data Analyst

Overview

You'll be the sole data person at an 8-person AI video generation startup, responsible for turning messy early-stage data into insights that drive product and growth decisions. You'll define what metrics matter, build the dashboards to track them, and run ad-hoc analyses when the team needs to make decisions. This isn't a plug-and-play role—you're building the data function from scratch.

Role Snapshot

Company Context

Stage: Seed (€16M raised, profitable)

Size: 8 employees

Growth: Targeting 3x ARR this year (aggressive growth from small base)

Market Position: Early-stage player in AI-powered marketing video generation space—competing with traditional video production, other AI video tools, and DIY solutions

Product: AI tool that generates marketing videos from text prompts, targeting growth teams and creators

Data Landscape Reality

Current State:

• Small team means data is probably scattered across multiple tools (Stripe, product DB, analytics platform, spreadsheets)
• "Messy environments" in the job description = you'll spend time cleaning data before analyzing it
• KPIs probably aren't consistently defined or tracked yet
• No existing data team means no one's built the infrastructure

What Needs Building:

• Core metric definitions (activation, retention, NRR, unit economics)
• Reliable data pipelines so numbers don't change depending on who pulls them
• Dashboards that stakeholders can actually use without bugging you
• Ad-hoc analysis capability when someone asks "why did churn spike last week?"

What You'll Actually Do

Time Breakdown

Data Infrastructure (35%) | Analysis & Insights (40%) | Stakeholder Collaboration (25%)

Key Activities

• Building & Maintaining Dashboards: You'll create the executive dashboard, growth metrics dashboard, and product analytics views. Expect to iterate these constantly as the team figures out what actually matters. You'll probably use tools like Metabase, Looker, or Tableau.

• Defining & Validating KPIs: Right now, "activation" might mean different things to different people. You'll nail down precise definitions for activation rate, onboarding conversion, retention cohorts, upgrade/downgrade patterns, and NRR. Then you'll make sure these numbers are actually correct and consistent.

• Running Ad-Hoc Analyses: The growth team will come to you with questions like "which onboarding flow drives better activation?" or "do users who try feature X retain better?" You'll pull the data, run the analysis, and present findings that inform product decisions.

• Managing Data Quality: You'll spend unglamorous time making sure Stripe data matches product usage data, fixing broken tracking, and investigating why numbers don't add up. At an 8-person startup, someone probably implemented tracking wrong at some point.

• Unit Economics Modeling: You'll build models to understand CAC, LTV, payback periods, and contribution margin by channel or customer segment. This matters because they're profitable but scaling aggressively—leadership needs to know what growth is sustainable.

The Honest Reality

What's Hard

• Solo Role Means No Backup: When something breaks or you're on vacation, there's no one else who knows the data infrastructure. You'll need to document everything clearly and build things that don't require constant manual intervention.

• Competing Priorities: With only 8 people and aggressive growth targets, everyone will want analysis yesterday. You'll need to push back and prioritize ruthlessly, which is uncomfortable when you're new.

• Data Quality Issues: Early-stage startups have messy data. You'll spend more time than you'd like investigating discrepancies, fixing broken tracking, and explaining why the numbers changed (because someone finally fixed the tracking).

• Ambiguous Requirements: "We need better retention metrics" isn't a clear spec. You'll need to ask good questions to understand what decision someone is trying to make, then figure out what analysis actually answers that question.

What Success Looks Like

• Leadership makes major product and growth decisions based on your analyses, not gut feel
• The team trusts your dashboards enough that they stop asking you to manually pull numbers
• You identify a key retention lever that meaningfully improves economics
• Within 6 months, metric definitions are standardized and everyone uses the same numbers

Who You're Working With

Primary Stakeholders:

• CEO/Founder: Wants unit economics, growth trajectory, board-level metrics
• Product Lead: Needs activation/retention data to prioritize roadmap
• Growth/Marketing Lead: Wants channel performance, funnel conversion, cohort analysis

What They Care About:

• Can they trust the numbers you're showing them?
• Can you translate data into actionable recommendations, not just charts?
• Can you move fast enough to support 3x growth targets?

Requirements

• 2+ years of business/product data analysis experience (preferably at an early-stage startup)
• Strong Python and SQL skills—you'll write queries daily and build analysis pipelines
• Comfortable with messy, incomplete data and figuring out how to make it useful
• Math background for unit economics modeling and statistical analysis
• Self-directed enough to define your own priorities when no one tells you what to do
• Comfortable working in English (company operates internationally) and ideally French (team is Paris-based)
• Genuinely excited about building from scratch rather than optimizing an existing system