Senior Marketing Data Analyst

About ApprovalMax

ApprovalMax provides end-to-end accounts payable automation for businesses using cloud accounting systems such as Xero and QuickBooks Online. We have 20,000+ businesses worldwide across 70+ countries, and strong product-market fit built over 8+ years. Our customers, accountants, bookkeepers, and finance teams are risk-averse, non-technical users who rely on us to control how money leaves their business.

Remote — applicants must be based in the UK, Portugal, Serbia or Moldova.

Role Overview:

Reporting to the Head of Analytics, you will own the analytical relationship between the Data Team and the Marketing function. This is a dual role, you will do serious analytical work, building the models, dashboards, and insights Marketing relies on, and you will own how Marketing uses data, setting the rhythms, SLAs, and standards that make analytics a genuine partner to the business rather than a service desk. The measurement layer underneath Marketing (attribution, tracking taxonomy, UTM standards, the HubSpot property schema, and the definitions behind the funnel numbers) has grown organically and needs to be rebuilt deliberately. You will be the person who does that, and then keeps it honest.

This is not a role for someone who wants to be handed a well-defined brief and execute against it. The embedded analytics model at ApprovalMax is being built now, and the person in this role will have real influence over what it becomes, first for Marketing and as a template for how analytics operates across the business over time. You should be as comfortable defining how analytics works as you are doing the analytical work itself.The role is deliberately scoped as an analyst, not a marketing operations generalist. You own the measurement architecture and the analysis, while campaign execution, website engineering, and tool administration sit elsewhere.

How your time is split. Roughly 80% marketing work and 20% wider data-function contribution like mentoring, standards, triage, and the shared practices that keep analytics consistent across the business.

Key Responsibilities:

Marketing Analytics

• Own the analytical relationship with the Marketing function: understand their priorities, anticipate their data needs, and deliver analysis that drives decisions rather than merely informing them.

• Build and maintain the core dashboards, reports, and models Marketing relies on, and take responsibility for their accuracy, freshness, and usability.

• Own the analysis behind the funnel that matters most to the business: visitor → free trial → paid conversion, trial nurture performance, lifecycle-stage progression, and the leading indicators that move them.

• Quantify channel and campaign performance end to end (spend, CAC, payback, pipeline contribution and retained revenue) rather than stopping at platform-reported conversions.

• Translate ambiguous business questions into clear analytical frameworks; make assumptions explicit and deliver useful outputs before perfect data is available.

• Bring cross-functional context to Marketing analysis, combining product usage, subscription revenue, and CRM data to surface insights a channel-level view would miss.

Attribution & Measurement

• Own the rebuild of marketing attribution: agree the model with Marketing and Finance, document its assumptions and known blind spots, and implement it in the warehouse so that one number is produced once and reused everywhere.

• Reconcile platform-reported numbers (ad platforms, GA4, HubSpot) against warehouse truth, and be the person who can explain — credibly and repeatedly — why they differ.

• Build the measurement approach for activity that attribution cannot see cleanly: brand, organic, partner and word-of-mouth. Incrementality and holdout testing where the spend justifies it.

• Design and analyse marketing experiments — landing pages, nurture sequences, pricing and packaging messages — including how success is defined before the test runs.

Tracking Architecture & Data Collection Standards

• Diagnose the tracking we already have, not just the tracking we intend to build: read the existing client-side implementation, cookie and local-storage handling, and referrer behaviour to work out why a channel is misreporting. Our largest single attribution problem today is an oversized Direct segment, and unpicking it is an early priority for this role.

• Own the event taxonomy: what we track, what each event means, what properties it carries, and what it is allowed to be used for.

• Own the Google Tag Manager container (tags, triggers, variables) and the QA process that proves tracking works before and after a release.

• Own the UTM naming standard and the automated compliance checks that enforce it. Per-campaign tagging is done by whoever launches the campaign; you own the standard, the validation, and the report that shows who is out of compliance.

• Specify the dataLayer contract that the website must expose, and the requirements for server-side tracking. You write the spec and verify the result; the implementation in site code and the infrastructure underneath it are built by web engineering.

• Own tracking QA as a standing process, not a one-off project: broken tracking should be detected by a check, not by someone noticing a dashboard looks wrong three weeks later.

Marketing Data Model & Metric Definitions

• Own the custom property schema in HubSpot from a data perspective: which properties exist, what they mean, who writes them, and how they map to the warehouse model. Specify the fields; workflow build sits with Revenue Operations.

• Own the Marketing domain within the Central KPI & Metrics Glossary — MQL, SQL, opportunity, trial, activation, conversion, retention — and bring additions and conflicts to the governance process rather than letting shadow definitions accumulate.

• Make sure that when Marketing, Sales, Finance and Product quote the same metric, they are quoting the same number, and that where they legitimately differ, the difference is documented rather than argued about.

• Ensure new product features, pricing changes, or GTM motions are reflected in metric definitions and dashboards ahead of launch, not after.

Ways of Working with Marketing

• Define and document the working model between the Data Team and Marketing: metric ownership, request intake, turnaround SLAs, escalation paths, and data freshness expectations.

• Establish an explicit fast lane for time-critical marketing requests, so that urgent work can be pulled into the current week rather than waiting for the next planning cycle — and be honest, in writing, about what does not qualify.

• Document domain knowledge, stakeholder preferences, and analytical edge cases so marketing analytics can be handed off or scaled without starting from scratch.

Self-Service, Reporting & AI-Assisted Workflows

• Build and maintain dashboards and automated reports (Lightdash, Amplitude, GA4) so Marketing can answer recurring questions without an analyst in the loop.

• Write clear documentation others can verify, extend, and build on: metric definitions, methodology notes, and data caveats.

• Actively shift recurring analytical work to self-service, freeing your capacity for higher-value analysis.

• Use LLM-assisted tools and AI coding agents as a default part of your analytical workflow — for data exploration, query generation, insight synthesis, and documentation. You should be faster and more thorough because of these tools, not merely aware of them.

• Help embed agentic workflows within the Marketing team, showing stakeholders how to get answers from data directly.

• Contribute as a practitioner and user of the natural-language and text-to-SQL tooling being built by the Data Platform Lead.

Data Quality, Governance & Mentorship

• Flag data quality issues upstream to the Data Platform Lead; understand how your analyses depend on specific pipelines and contracts, and raise risks proactively.

• Treat marketing data quality as your own problem where it originates in taxonomy, tagging, or CRM property design — that is your surface, not the platform team's.

• Participate in the intake and weekly triage process; help classify and scope incoming requests from Marketing.

• Provide technical guidance, analytical framing, and code reviews for junior analysts.

• Contribute to agile workflows (Kanban/Scrum) and planning sessions; keep your work visible and priorities explicit.

• Help shape the documentation standards and analytical norms that will apply as the team scales.

What We're Looking For

Must Have's:

• 4+ years of hands-on experience in a data analyst role, including significant time analysing marketing, growth, or GTM funnel performance for senior stakeholders.

• Owned a marketing measurement stack end to end: attribution modelling, event taxonomy, UTM governance, and the reconciliation between platform-reported and warehouse numbers.

• Diagnostic depth in existing tracking implementations. You can read a site's tracking code, cookie and local-storage logic, and referrer handling well enough to explain why a channel is misreporting, and then specify the fix. Concretely: given an unexplained Direct segment absorbing a large share of conversions, you know how to establish the cause rather than assume one.

• Demonstrated use of AI coding agents and LLM-assisted tools as a core part of your analytical workflow, not occasionally but by default. You should be able to show how these tools have changed the speed and depth of your work. We are building an agentic-ready data function, and this role needs to model that approach.

• Strong SQL skills and proficiency in Python for data manipulation, automation, and analysis, with experience writing production-quality code, not just ad hoc scripts.

• Hands-on with the marketing data stack: GA4, Google Tag Manager, a CRM (HubSpot strongly preferred), and the major paid channels' reporting.

• Experience with modern BI and product analytics tools — Lightdash, Looker, Amplitude, or equivalent — including building dashboards that non-technical stakeholders actually use.

• Track record of delivering analysis that combines data from multiple source systems (CRM, product, finance, marketing) to answer questions no single team could answer alone.

• Ability to communicate findings clearly and concisely to senior non-technical audiences, in writing, verbally, and through visualisations — including telling a marketing stakeholder that a number they like is wrong, and making it stick.

• Experience working within a metric governance process: definitions, ownership, and resolving conflicts between competing numbers.

• Comfortable operating with ambiguity, able to scope an unclear question, make reasonable assumptions explicit, and deliver something useful before the perfect data exists.

Nice to Have:

• Experience in a SaaS or B2B product company, with exposure to subscription metrics (ARR, NRR, churn, expansion), product-led trial funnels, and self-serve conversion analysis.

• Experience with product-led growth measurement: free-trial to paid conversion, activation, and in-product behaviour as a leading indicator of revenue.

• Familiarity with dbt or similar transformation frameworks; understanding of how semantic layers and data models underpin self-service analytics.

• Experience specifying server-side tracking or a dataLayer contract for engineers to implement.

• Experience with incrementality testing, geo experiments, or media-mix modelling.

• Experience working in or helping to establish an embedded analytics model (business partner, spoke, or equivalent).

• Exposure to LLM application patterns, prompt design, text-to-SQL, or the use of language models for data exploration and insight summarisation.

• Experience mentoring junior analysts or setting analytical standards within a team.

Environment You'll Work In:

Databricks and dbt as the central warehouse and transformation layer, Lightdash for reporting and the semantic layer, Amplitude for product analytics, HubSpot as the CRM, GA4 and Google Tag Manager for web measurement, and the usual paid and SEO tooling. The data platform is maintained by a dedicated Data Platform Lead, so you will spend your time on measurement and analysis rather than on keeping pipelines alive.

We are mid-migration onto Lightdash and expect to retire our legacy Power BI reporting during Q3–Q4. A small number of recurring marketing reports still run there and are exported manually today; moving them onto the new stack, and automating them on the way, is part of the early work in this role.

What We Offer:

• Health & Wellbeing Benefit Stipend

• Home Office Setup - one off £500 reimbursement, post probation

• 26 days of paid time off

• 1 additional day off for your birthday

• Service-years recognition financial reward

• Regular performance-based compensation reviews

• Growing international business with 20,000+ subscribers