From survey programming and quality-screened fielding to crosstabs with significance testing, coded open-ends, live dashboards and index scoring. One partner, end to end, with 30+ years of market research behind every deliverable.
Hire one piece or the whole pipeline. Each stage is built to the same standard: traceable numbers, documented decisions, and deliverables a research lead can put in front of a client without a second read.
Instrument design review, programming on Voxco and comparable platforms, vendor and panel management, and soft-launch relaunch discipline after any instrument change.
Every complete is checked, not sampled. Speeders, straight-liners, duplicate devices, junk verbatims and out-of-population respondents are removed with a documented, billing-defensible cause.
Hyperlinked marginals workbooks and two-book crosstabs (percentages plus frequencies with significance testing) against a banner your research lead approves before anything is built.
Formal NET and sub-net codebooks, built with your research lead and applied consistently across marginals, crosstabs, dashboards and the report deck.
Password-protected web dashboards that move with the study: sample, pre-test, live fielding with a real-time completes counter, and a final results portal with every chart carrying its base.
Index and scoring studies, segmentation, weighting sensitivity checks and insight-led report decks with in-line "by segment" tables after every major question.
Every step exists because skipping it once cost a study time or credibility. Clients see progress at every stage through their dashboard.
Questionnaire review, screener hardening, quota plan, and a banner drafted for your research lead to bless.
Programmed instrument, end-to-end vendor link tests, and a small pre-test wave before anything ships.
15 to 20 completes and one reconciliation cycle after any instrument change, then full release.
Real-time completes counter, quota fill, and quality flags visible to the client while sample is in the field.
Every complete screened. Accepted, rejected or review, each with a cause. Land exactly on contracted n.
Marginals, two-book crosstabs, coded open-ends, final dashboard and deck, all traceable to one final data file.
A working example of a Carter Data Collection client dashboard. The numbers below are illustrative, but the layout, states and controls are exactly what ships. Click the segment filters to see the results views re-cut.
Any methodology section we write can answer, on demand, how gross completes became the published n. These are the standing rules applied to every study.
Interviews under 180 seconds are removed automatically. Under 300 seconds with straight-lining or a junk open-end is removed with cause. Everything between is flagged for human review, never auto-rejected.
Duplicate-IP clusters are investigated on device, geography, timing and answer overlap before a decision. Overlapping sessions on one device profile are rejected together.
Respondents who reveal themselves outside the target population in an "Other (specify)" are removed. Client-priority segments are protected from borderline calls and only rejected on hard evidence.
The final scoring or data file drives every downstream artifact. After any rescoring, every tab, chart and narrative claim is re-derived and diffed against recomputed values.
Before a tab book ships, every cell is recomputed independently and diffed against the built workbook, including derived banner columns at tier cutoffs.
Every vendor invoice reconciles to this. Every row has a reason a billing manager can read.
Trap questions and quality flags stay out of client-facing documents; methodology describes them generically as attention checking.
Founder, Carter Data Collection
30+ years in market research, 25+ years leading research and data collection teams. Based in Salt Lake City, Utah, working with clients nationally.
Carter Data Collection is an independent market research firm led by Luis Carter, a research executive who spent three decades managing fielding operations, tabulation and reporting before building the firm around a simple idea: the research lead should never have to re-check the numbers.
Recent work includes a new-college concept study for CFSI, creator commerce research for LTK, and a composite index and scoring study for MINTY, each delivered with live dashboards, two-book crosstabs, coded open-ends and executive report decks.
Share the audience, the rough n, and the deliverables you need. You'll get a scoped proposal with a fielding plan, timeline and quality rules, usually within two business days.