Asks for

The ad says“Autonomy, and fluency with AI tooling. You scope your own work with little guidance, and you use tools like Claude and Cursor to compress your cycle time rather than working around them.”
The ad says“Aggregation and privacy discipline. Experience publishing or sharing findings derived from customer data, with anonymization and minimum-cohort rules you can defend to a security reviewer. This one is non-negotiable.”
The ad says“Autonomy, and fluency with AI tooling. You scope your own work with little guidance, and you use tools like Claude and Cursor to compress your cycle time rather than working around them.”
The ad says“Real statistical grounding. Sampling, confidence intervals, significance, cohort and time-to-event analysis. You know when a difference is noise and you understand what selection bias does to a benchmark drawn from one vendor’s customer base.”
The ad says“Real statistical grounding. Sampling, confidence intervals, significance, cohort and time-to-event analysis. You know when a difference is noise and you understand what selection bias does to a benchmark drawn from one vendor’s customer base.”
The ad says“Autonomy, and fluency with AI tooling. You scope your own work with little guidance, and you use tools like Claude and Cursor to compress your cycle time rather than working around them.”
The ad says“Data engineering ownership. You have worked on a warehouse project end-to-end. This includes dimensional or semantic modeling, a transformation framework such as Dataform or dbt with models, tests, docs and CI, orchestration, and the discipline that keeps it from rotting. We run BigQuery on Google Cloud.”
The ad says“Data engineering ownership. You have worked on a warehouse project end-to-end. This includes dimensional or semantic modeling, a transformation framework such as Dataform or dbt with models, tests, docs and CI, orchestration, and the discipline that keeps it from rotting. We run BigQuery on Google Cloud.”
The ad says“Real statistical grounding. Sampling, confidence intervals, significance, cohort and time-to-event analysis. You know when a difference is noise and you understand what selection bias does to a benchmark drawn from one vendor’s customer base.”
The ad says“Data engineering ownership. You have worked on a warehouse project end-to-end. This includes dimensional or semantic modeling, a transformation framework such as Dataform or dbt with models, tests, docs and CI, orchestration, and the discipline that keeps it from rotting. We run BigQuery on Google Cloud.”
The ad says“Real statistical grounding. Sampling, confidence intervals, significance, cohort and time-to-event analysis. You know when a difference is noise and you understand what selection bias does to a benchmark drawn from one vendor’s customer base.”
The ad says“Production SQL, at depth. Window functions, incremental models, and query optimization at warehouse scale over large event and journey tables. You write and tune the query yourself rather than briefing someone else to write it, and you reason about cost as well as correctness.”
The ad says“Real statistical grounding. Sampling, confidence intervals, significance, cohort and time-to-event analysis. You know when a difference is noise and you understand what selection bias does to a benchmark drawn from one vendor’s customer base.”
The ad says“Clear technical writing. You can write a methodology note that a marketer understands and an analyst cannot poke holes in, and state plainly what a number does and does not show.”
The ad says“Real statistical grounding. Sampling, confidence intervals, significance, cohort and time-to-event analysis. You know when a difference is noise and you understand what selection bias does to a benchmark drawn from one vendor’s customer base.”

About Dreamdata

Dreamdata is the industry leader in B2B marketing attribution and activation. We’re on a mission to help B2B marketing leaders finally connect their efforts directly to real revenue — solving a complex problem that has plagued the industry for years.

Founded in 2018 by ex-Trustpilot (TRST:LSE) product and engineering leaders, Dreamdata was born from a real-world need, building the platform they always wished they’d had.

It’s working. We have achieved strong product-market fit and are experiencing explosive, triple-digit YoY growth. This momentum is now backed by a recent $55,000,000 Series B funding round set to fuel our rocketship as we scale globally.

We are expanding rapidly and looking for exceptional, top-tier talent to help us build the future of B2B marketing.

The Opportunity

The Dreamdata Platform is used by leading B2B companies to do marketing attribution and activation. The product is something rare: a uniquely complete view of how B2B buying actually happens, across every channel a modern B2B buyer touches.

That data is the raw material for the best benchmarks in the industry. Our LinkedIn Ads Benchmarks report has been downloaded thousands of times and routinely shows up in third-party newsletters, decks, and LLM answers when people ask how B2B marketing actually performs. We want Dreamdata to be the canonical, cited source of truth for what’s working in B2B marketing.

We are hiring a Staff Insights Engineer to build the and scale data foundation that makes that possible. This is a deeply technical role: analytics engineering, data science, and warehouse ownership are the core of the job.

The data foundation is the starting point. Before Dreamdata can publish anything the market will cite, someone has to model the underlying dataset, get the aggregation and anonymization right, ensure a methodology holds up to scrutiny, and build the tooling that allows us to publish quality answers to questions within hours. That is what you are here to do. If the data underneath a benchmark isn’t right, nothing built on top of it is worth publishing.

You will not do this alone and you will not be doing it all. Our marketing team will run the editorial calendar, the narrative, and the brand. You own everything upstream of that, and you work in close partnership with them; they help you understand the questions the market cares about, and you make sure the answers are right, fast, and properly caveated.

You own

The self-service layer and the internal tooling marketing runs on
The warehouse models, pipelines, and tests behind every published figure
Aggregation, anonymization, and minimum-cohort thresholds
Statistical methodology, and the caveats that travel with each number
Analytical correctness: the right sample, denominator, and scale
Technical sign-off on published claims

Marketing owns, you support

Editorial calendar, publishing cadence, and channel mix
Headlines, narrative, and visual polish
Dreamdata’s social presence and brand voice
Journalist, newsletter, and analyst relationships
Report packaging, launches, and events
Demand generation targets

You will be measured on whether the data foundation is correct, fast, and trusted as well as how little the rest of the team has to wait for you.

This role reports into our engineering organization, works day-to-day with the marketing team, and grows into more ownership over time. It is a rare combination: warehouse-deep technical work whose output an entire industry sees, rather than another internal dashboard.

Key Responsibilities

Your first six months: building the foundation

Learn the Dreamdata data model end-to-end, embedded with our engineering team, and model the benchmark dataset on top of it.
Improve the aggregation and anonymization layer.
Stand up data transformations properly: version-controlled models, tests, CI, and documentation.
Define the metric definitions and the semantic layer that marketing can query without you in the room.
Set the methodology standard: sample sizes, confidence intervals, and how we handle and disclose selection bias, given that our data describes companies that buy Dreamdata rather than the whole market.

Ongoing — run it at speed

Turn an interesting question into a defensible answer quickly, and keep driving that cycle time down. Today marketing waits on engineering to run a query; your job is to remove that bottleneck permanently.
Own the analysis behind quarterly benchmark reports and the recurring data marketing publishes.
Act as technical reviewer to ensure we don't make claims the data does not support.
Keep the self-serve layer good enough that marketing answers most of its own questions without you.

Requirements

Must have

Production SQL, at depth. Window functions, incremental models, and query optimization at warehouse scale over large event and journey tables. You write and tune the query yourself rather than briefing someone else to write it, and you reason about cost as well as correctness.
Data engineering ownership. You have worked on a warehouse project end-to-end. This includes dimensional or semantic modeling, a transformation framework such as Dataform or dbt with models, tests, docs and CI, orchestration, and the discipline that keeps it from rotting. We run BigQuery on Google Cloud.
Real statistical grounding. Sampling, confidence intervals, significance, cohort and time-to-event analysis. You know when a difference is noise and you understand what selection bias does to a benchmark drawn from one vendor’s customer base.
Aggregation and privacy discipline. Experience publishing or sharing findings derived from customer data, with anonymization and minimum-cohort rules you can defend to a security reviewer. This one is non-negotiable.
Clear technical writing. You can write a methodology note that a marketer understands and an analyst cannot poke holes in, and state plainly what a number does and does not show.
Analytically correct charts. You can produce a chart that is right — correct cut, correct denominator, honest axes — and hand it to marketing and design for polish. You do not need to be a designer, and you are not expected to write the copy that goes around it.
Genuine curiosity about B2B go-to-market. You want to know what the data says about how B2B buying actually works. We can teach you the domain; we cannot teach you the interest.
Autonomy, and fluency with AI tooling. You scope your own work with little guidance, and you use tools like Claude and Cursor to compress your cycle time rather than working around them.

Nice to have

BigQuery and Google Cloud specifically.
dbt run at scale, or another transformation framework you have owned in production.
Having been the analyst behind a public benchmark or data report.
Python for analysis: pandas, notebooks, and visualization in matplotlib, Plotly, Vega or Tableau.
B2B martech, RevOps, or adjacent: attribution, CDPs, ABM, growth analytics.
A working point of view on B2B SaaS, GTM, and how buyers actually behave.
Comfort being the technical voice in the room: a podcast segment on methodology, a conference talk, a call with an analyst. Welcome, never required.
An interest in sharing your own work publicly, alongside what marketing publishes.

Benefits

Location: Our office is located in the heart of Copenhagen with a harbor view
Ownership: Equity in a growing company
Lunch: Delicious lunch available in our canteen
Inclusive Culture: Be part of a supportive and dynamic team making an impact in the B2B marketing space
Private insurance: Private health insurance that comes with our pension plan
Office dogs

At Dreamdata, you’ll have the opportunity to make a real impact. Join us in shaping the future of B2B marketing analytics and activation while developing your skills and growing your career.

We are committed to creating an inclusive, diverse workplace where everyone can thrive. If you build data foundations you would stake your name on, and you want that work seen by an entire industry, we’d love to hear from you.

Asks for

Values