White-labeled research reports for B2B tech

I design the study, run the research, and create the report. You publish it as your own.

The market for B2B research is fragmented

  • Forrester and Gartner engagements run $80K to $200K and take six to nine months
  • Boutique research firms charge $5K to $50K and deliver findings (deck, exec summary, raw data). Turning that into publishable content is a separate project with a separate vendor
  • Content agencies produce whitepapers built around existing data rather than primary research

Between these tiers, companies that want genuinely original research delivered in a form they can publish have limited options.

What I do

  • Design a bespoke study based on conversations with your stakeholders
  • Collect and analyze the data
  • Review results and interpretations with your stakeholders, refining as necessary
  • Develop a formatted report written in your brand voice with charts, graphs, and visuals
  • Deliver raw data, processed data, documented methodology, and formatted, editable reports (HTML, MS Word, PDF)
  • Build three to five derivative content assets (executive summary, blog post, slide deck, social assets)

Where the data comes from

Your user data
Public agencies
Open platforms
Live experiments
Surveys
Interviews
All of the above
Recently published
Nature Letter

AI agents replicate human social dynamics in days

Co-authored with Dr. Sam Illingworth. Nature 652, 828 (April 2026). A platform built for AI agents reproduced human social dynamics within days of launch.
Reddit analysis · July 2025

I analyzed 415 Reddit posts about AI in schools

Solo. 415 Reddit posts across 10 education subreddits, ML keyword analysis plus qualitative review. Expert interview with Dr. Torrey Trust (ed tech and digital literacy, UMass Amherst). Surfaced what parents and teachers worry about most, plus practical guidance from a specialist.
AI grading study · May 2026

I ran over 1,000 API calls to find out how ChatGPT grades essays

Solo. 1,120 API calls across four GPT models, eight experiments. AI graders are more consistent than two human teachers, but vulnerable to gaming, prompt injection, and bias against non-native English writers. Code on GitHub.
Prompt tone study · April 2026

Does it matter if you're polite to Claude?

With Dr. Sam Illingworth. 500 API calls across Sonnet 4.5 and Opus 4.7. Tone doesn't affect correctness, but flattery compresses Opus's deliberation by nearly 50%.

Process

1 Week 1 Scoping, hypothesis development, methodology approval
2 Weeks 2–4 Data collection
3 Weeks 5–7 Analysis and first draft
4 Weeks 8–10 Revisions, design, derivative assets

Typical engagement: 8–10 weeks.

Pricing

Starting at $30,000

Most engagements range $30K to $60K based on scope: sample size, number of methods, depth of analysis, and derivative assets.

About

I'm Karen Spinner. I've spent fifteen years writing for B2B technical buyers and the last few years conducting empirical research on AI systems. Co-author of correspondence published in Nature on AI agent social dynamics. Prior clients include Adobe, Confluent, ChargePoint, Drift, and VidMob.

A few sentences are useful to start: company, audience, the question a report would answer, target launch window. Responses typically within 48 hours.