Data & Intelligence Analyst

We are hiring a Data & Intelligence Analyst to own our data work end to end: getting hold of the data we need, turning it into clean and reliable datasets, producing the analysis, and delivering the reports and dashboards our clients commission.

This is a hands-on ownership role, not a support-only one. You will be given the question that matters. From there it is yours: work out where the data lives, go and get it, make it trustworthy, work out what it says, and deliver it as something a senior decision-maker can act on. You own the method, the definitions, the quality, and the finished product.

The bar is high. This work goes in front of senior people who will ask exactly how each number was produced, and you need to be able to answer.

Your core advantage is that you work fluently with AI. You use AI tools not as a novelty but as the way you operate: to gather, classify and analyse data, to move faster, and to produce more, better. That fluency is central to the role.

This is a fully remote role. We are open to candidates based anywhere.

Working hours
Monday to Friday: 9am to 6pm UK time.
Saturday: 9am to 1:30pm UK time.

The Role: AI-Native, used centrally throughout

You will:

• Define exactly which records are in scope for each piece of analysis, source them from licensed data providers, APIs and public sources, and verify how much of the intended set you actually managed to capture before you rely on it.
• Acquire supporting data from wherever it exists: public registers, official statistics, published datasets, international government sources, and formal information requests.
• Assemble everything into clean, structured, reliable datasets, including the unglamorous work of reconciling the same entity appearing under four different names across four different sources.
• Normalise high volumes of inconsistent free-text values into a clean, consistent and defensible set of categories, and keep that mapping stable over time.
• Work out what the data implies where it does not state it directly, inferring what you need from the fields you do have. Document the method and measure how often it is right.
• Own the classifications and definitions: how records map to categories, how comparison groups are defined, how each metric is calculated. Keep them stable so this quarter's analysis is comparable with the last one.
• Define every metric precisely and stand behind it. Where a figure is directly observed, say so. Where it is estimated, build the model, document it, validate it against external benchmarks, and label it on the output.
• Own the end-to-end production of our client-facing reports, from raw data to a finished, polished document, delivered inside the committed window.
• Produce the analysis, the comparisons and the charts, and write the interpretation and recommendations that go with them: concise, evidence-led, and in plain English a busy reader understands in thirty seconds.
• Handle the awkward realities of large record-level datasets: missing fields, inconsistent naming, partial dates, duplicates, and knowing when a small sample cannot carry a percentage.
• Explain and defend your methodology when someone challenges it, including its limitations, honestly.
• Build and maintain our recurring datasets and tracked indicators, refreshed to a calendar, so the picture is always current rather than assembled in a panic.
• Compare data across different countries and reporting systems, and know when two numbers that look comparable are not.
• Specify the dashboard versions of our products precisely enough for our engineers to build them: the data model, the filters, the comparison logic, and the refresh cycle.
• Turn each piece of work into a repeatable process. The second time should be much faster than the first, and the tenth should be close to routine.
• Use AI across the whole workflow to work faster and better than a traditional analyst.
• Flex onto other data work across the business when needed, from marketing and funnel analysis to commercial modelling and market research.

The work spans everything the business needs evidence for, wherever a sharp and reliable analyst is most useful. You are the person who makes sure the data exists, the numbers are right, the method is defensible, and the finished product is excellent.

What we're looking for

• 2 to 6 years in a data, analytics or research role where you owned an analytical output and delivered it to someone who relied on it, not just supported someone else's reports.
• You go and get data rather than waiting for it. You have worked with datasets you had to assemble yourself, not only ones that were handed to you clean.
• Genuinely good with data: you can take a large, messy, inconsistent dataset and make it tell the truth.
• Comfortable with data far too large for a spreadsheet. Working SQL and/or Python, and strong in Excel.
• Experience with large record-level datasets, including data licensed from third-party providers, and the judgement to work out whether a provider's coverage is genuinely what they claim.
• Experience classifying and standardising messy real-world data, for example mapping thousands of free-text values into a clean set of categories.
• Comfort working out what a dataset implies where it does not say so explicitly, and honesty about the error rate when you do.
• Strong data visualisation: Power BI, Tableau, Looker, or an AI-built equivalent. Whatever you use, you produce a clear, polished result and can explain exactly why you chose that view.
• Excellent written English, with the ability to turn analysis into a clear, concise piece of writing that reads as yours. This is essential and will be assessed.
• Good judgement about certainty. You define your metrics before you calculate them, you know the difference between a real finding and a coincidence, you label estimates as estimates, and you will not let a number ship that you cannot defend to someone who challenges it.
• Genuine, demonstrable fluency with AI tools as a core part of how you work, with concrete examples, not just "I use ChatGPT sometimes".
• Strong ownership: you take an ambiguous problem, work out the steps yourself, and come back with a finished answer rather than a list of questions.
• Organised and reliable: commissioned work runs to a deadline and recurring work runs to a schedule, and you hit both.
• Comfortable working fully remote.

Nice to have

• Experience in a role where the core task was assembling a target list or dataset from third-party and public sources, for example market mapping, competitive intelligence, or data enrichment for sales or marketing.
• Experience collecting data from web sources and APIs.
• Experience building a benchmarking product, an index, a scoring model, or any analysis that compares organisations against a peer set.
• Experience evaluating and buying third-party data.
• Experience with public records research, freedom of information requests, or government and official statistics sources.
• A statistics or economics background, or a track record of self-taught rigour.
• Comfort with marketing and funnel data, or with commercial and financial modelling.
• Building automations and data pipelines (Python, SQL, n8n, or AI-driven workflows).