Journal.
On people, choices, and the
evidence in between.
Latest articles
174 articles
Key-driver analysis: choose the improvement behind the score
In Alder’s fictional example, two communication ratings overlap strongly. A confidence-in-the-fix rating adds more information, changing which improvement the team should investigate.
Read article
Turn Sapien research into a useful ChatGPT or Claude brief
Prepare approved findings, product facts and a clear assignment. Follow a manual ChatGPT or Claude handoff from source inventory to a reviewed landing-page brief.
Read article
Churn research: find what has to change before asking customers to stay
Investigate churn with usage cohorts, event records and exit interviews. Compare explanations, retain missing voices and test a useful change.
Read article
NPS vs CSAT vs CES: choose the question your CX team needs answered
A broadband example separates +25 NPS, 80% CSAT and 3.9 out of 5 activation ease. Read each against its own question and respondent base.
Read article
Synthetic research for agencies: make the creative recommendation inspectable
Connect a client brief, audience comparison and creative revision. Use modeled research to make an agency recommendation clear, inspectable and bounded.
Read article
SaaS pricing research: test the package buyers will actually pay for
Test a SaaS package around seats, usage, buyer roles and the real bill. Compare the outside option, trade-offs and economics before a commercial test.
Read article
Data privacy in synthetic market research: what to check before a study
Review source permissions, personal data, processing roles, providers, retention and transfers, with the facts in Sapien's current public privacy policy.
Read article
How Sapien builds a synthetic audience for your research
See how Sapien connects consumer data, audience definition, surveys and comparison tests to an interactive research report and useful follow-up questions.
Read article
Synthetic market research vs ChatGPT: what are you paying to learn?
Compare ChatGPT and managed synthetic research by audience, grounding, study design and validation. Decide what extra evidence your team needs.
Read article
Kano analysis: which features must work, and which could make the difference?
An illustrative support-portal example shows why a 45% must-be plurality needs more investigation before it becomes a roadmap priority.
Read article
Win-loss research: find what changed before rewriting the pitch
Reconcile wins, competitor losses, no-decisions and unknown outcomes before choosing your buyer research program or changing the sales response.
Read article
Financial services market research: when an account opening is not a switch
Choose banking research that distinguishes an account opening from a move of everyday banking, using account roles, payment routines and approved offers.
Read article