The Doyle Group is a proven partner for Placement and Consulting services, headquartered in Denver, CO. Our core mission is to forge genuine partnerships with our clients who seek strategic talent solutions and to assist highly skilled candidates looking for their next career opportunity. With over 30 years of industry experience, our consultative approach allows us to provide a higher level of guidance and insight, empowering our clients to secure top IT talent that fits seamlessly into their team and culture. We look forward to collaborating to help you achieve your career goals.
| POSITION SUMMARY |
Our client is a fast-growing, private-equity-backed, tech-enabled marketing agency that runs direct response advertising campaigns across TV, radio, digital, and direct mail for clients spanning performance, general media, and retail verticals. The team's current analytics lead is departing, and the company needs a hands-on Data Science & Analytics Lead to backfill the technical core of that role: owning attribution and measurement so the media, account, and leadership teams can trust the numbers behind every budget decision.
This is not a people-management position — the team already has a leadership structure in place. The company is looking for someone who leads through analysis, modeling, and example rather than headcount, and who can step in and contribute immediately because they've already built this kind of expertise inside marketing, an ad agency, or a data science/analytics vendor. Media modeling and media data have a language of their own, and this role is built for someone who already speaks it.
You'll join a lean, four-person analytics team that reports up through a VP of Analytics, Automation and Insights, working as one of two senior peers alongside a junior analyst. The role is broad and hands-on: roughly half building and validating attribution and measurement models (multi-touch attribution, incrementality and lift testing, and marketing mix modeling), a third building the BI dashboards and self-service reporting that non-technical stakeholders rely on daily, and the remainder spent raising the analytical bar for the team through peer review and shared standards. You'll also work directly with media buying, account, finance, and data engineering teams — and if you're excited about bringing AI and agent-assisted tooling into analytics work, you'll have real room to shape how the team adopts it.
This is a full-time, hybrid role based out of the company's Shelton, Connecticut office. The company would love a candidate who can be in the Shelton office regularly, but is also open to strong candidates within a reasonable drive of the tri-state area (CT/NY/NJ/PA/MA) who can travel in for a few days roughly once a month — travel and lodging are covered. Candidates must be authorized to work in the United States without current or future visa sponsorship.
| RESPONSIBILITIES |
- Own the company's attribution and measurement strategy — decide which method (multi-touch attribution, incrementality/lift testing, or marketing mix modeling) answers which business question, and build and maintain each.
- Design and run incrementality and lift tests, including holdouts, geo experiments, and matched-market tests, defining the test design, sample size, and readout methodology the team can trust and repeat.
- Build and maintain marketing mix models that translate media spend, seasonality, pricing, and outside factors into models that inform budget allocation across channels and markets.
- Push attribution toward true 1:1, cross-funnel measurement — moving beyond last-click/last-action views into multi-touch attribution across the full customer journey.
- Build and own BI dashboards and self-service reporting (Power BI, Looker/Looker Studio, or similar) that let media, account, and finance teams use your models without an analyst in the loop.
- Partner directly with data engineering to define what your models need from the data platform — schema, freshness, and pipeline reliability for the tables that feed your work.
- Translate model output into a media and budget point of view, working directly with media buying and account teams to turn results into channel, campaign, and budget recommendations.
- Validate and stress-test your own models — document assumptions, define what “working” means for each method, and hold your work to a standard that would survive a challenge from finance or a client.
- Review other analysts' models, dashboards, and statistical approaches, and help set the shared standards the team analyzes to (technical leadership through practice, not people management).
- Bring AI and agent-assisted workflows into the team's analytics work — using LLMs and agent tooling to accelerate exploratory analysis, QA models, and prototype new measurement approaches — while holding that work to the same rigor as fully manual analysis.
- Support light operational reporting tied to your work, such as finance or revenue-reconciliation checks that keep the team's numbers accurate.
- Mentor analysts through model review and pairing, and weigh in on technical hiring decisions for the team, without owning performance management.
| MINIMUM EXPERIENCE |
- 6+ years of professional experience in marketing analytics, data science, or a measurement-focused role, ideally including time acting as a de facto technical lead or setting analytical direction.
- Hands-on, provable experience across all four of the following — the team's non-negotiables:
- Marketing mix modeling (MMM) — building, operationalizing, or substantially modifying an MMM, whether off-the-shelf/third-party, open-source (e.g., Google Meridian), or custom-coded
- Multi-touch attribution (MTA), including cross-funnel/cross-behavior attribution beyond simple last-click models
- Incrementality testing (holdouts, geo experiments, matched-market tests)
- Lift studies
- Background in marketing, an ad agency, or a data science/analytics vendor — media modeling and media data are specific enough that this context matters more than years of generic data science experience.
- Strong statistical foundation, including:
- Experimental design and causal inference (e.g., familiarity with tools like Google's Causal Impact)
- Regression and time-series methods
- Judgment for what data volume and controls a given model actually needs (for example, a reliable MMM needs a long enough historical window — not six months of data)
- Strong SQL for production analysis: complex joins, window functions, and query optimization against a modern cloud data warehouse. No specific platform is required — BigQuery, Snowflake, Postgres, or similar all count — but you should understand why a platform like Snowflake exists and have worked with something comparable.
- Working proficiency in Python or R for statistical modeling and analysis, with clean, reproducible, documented code rather than one-off notebooks. Python is preferred.
- Proficiency in a modern BI platform (Power BI, Looker or Looker Studio, or similar), building dashboards that non-technical stakeholders actually use.
- Experience with media and marketing data: platform-level spend and performance data, CRM/conversion data, call tracking, and the quirks specific to direct response advertising.
- Clear communicator who can explain methodology and trade-offs to media, account, and finance stakeholders and defend a model under scrutiny.
- Bachelor's degree or equivalent professional experience.
- Must be legally authorized to work in the United States without sponsorship now or in the future.
| ADDITIONAL PLUS |
- Prior experience as a technical lead or “player-coach” who set standards through peer review rather than formal people management
- Genuine enthusiasm for AI and agentic tooling in analytics — willing to bring LLM/agent workflows into modeling and QA work
- Experience across a wide range of media verticals (performance, general/brand media, retail) rather than a single category
- Exposure to Google Meridian or another open-source/code-based MMM framework
- Experience presenting modeling results directly to finance or client-facing stakeholders
- Comfort working in a fast-paced, growing agency environment with a short path from analysis to decision
