This guide helps survey programmers and research operations teams compare their Forsta Decipher workflow with the MX8 Labs Research Platform. It covers questionnaire setup, QA, fielding, data processing, and a parallel evaluation before migration.
The core difference in one paragraph
On MX8 Labs, your questionnaire document supplies the questions and instructions used to program the survey. Specify logic, quotas, and randomization in the document, then inspect the generated build before fielding. Edits can be requested in plain English and reviewed after they are applied.
How familiar concepts map across
Questionnaire and survey logic. Skip patterns, display logic, piping, and masking are written in your survey document as plain, unambiguous instructions ("Ask Q5 only if Q3 = Yes"). The platform programs them and they are reviewable before fielding. See survey logic best practices for examples of clear instructions and the checks to make before fielding.
Randomization and rotation. Statement randomization, block rotation, and concept rotation (monadic and sequential monadic designs) are specified in the document and handled automatically in the build.
Quotas. Quota definitions — including nested and interlocked structures — are set from your specification, with least-fill balancing managed by the platform. The quota management methodology documents the behavior.
Link testing and QA. Generate simulated responses to test survey paths, then review the transcripts for logic, piping, and routing problems. Check coverage and manually review any paths the simulation did not exercise. Simulated data is free and can be regenerated after edits.
Fielding and sample. Respondent sources connect directly: 70+ integrated panels, plus conversational SMS, AI voice interviewing, and first-party lists from CRM. Fraud detection and attention checks run in real time, before a response is accepted. Review the data quality results before analysis. See how to set up respondent sources.
Data processing. Weighting (two-stage IPF raking), significance testing (with effective-sample-size adjustment), and open-end coding run automatically as responses land. The weighting methodology and stat testing reference document exactly what runs, and results are deterministic and reproducible — see reproducing and auditing results.
Exports and downstream tools. SPSS and tabular exports let you retain compatible downstream tools. During the evaluation, check variable names, labels, coding, and formats against your tab specifications and archival requirements. See downloading data and reports.
Advanced methods. Hierarchical Bayes conjoint and MaxDiff are built in, with design tools available for both. Existing conjoint designs can be imported — see importing a completed conjoint design.
What your team stops doing, and what it keeps
Automation reduces manual programming, tabulation, and report formatting. Your team still needs to specify the questionnaire, review the generated survey, check data quality, and approve the analysis and recommendations. Include that review time when comparing workflows.
Running a parallel evaluation
The recommended first step is one study run in parallel with your current process:
- Pick a live or recently completed study with a known cost and timeline.
- Use the same questionnaire document and the same sample frame and quotas.
- Program, QA (simulated data plus your own review), and field it on MX8 Labs alongside your normal process.
- Compare directly: elapsed time per stage, internal hours consumed, external vendor cost, and — most importantly — the data and deliverables themselves, side by side.
Nothing is migrated, no client work depends on the result, and the comparison is yours to keep either way.
Migrating a tracker
Map the existing questionnaire question by question and field a bridge wave on both platforms with comparable samples and overlapping questions. Document where results agree or differ, investigate the differences, and agree any adjustments before joining the series. A bridge wave helps assess continuity; it does not guarantee that every measure remains comparable. Consider changes to cadence separately from the platform migration.
Getting help
Dedicated switching packages cover instrument mapping, bridge waves, team onboarding, and programming support for your first studies — get in touch and mention what you run today. For the strategic context around the Decipher platform's ownership, see our guide to switching from Decipher.