Production cost becomes easier to examine when programming, reporting, and the remaining production work are shown separately. The following illustrative production-only model compares $20,000 with $4,150. It is a planning example, not an audited customer P&L, a market average, or a current MX8 Labs Research Platform tariff.
Model reviewed October 2, 2026. This example covers production labor only. It excludes sample, fielding, agency strategy, interpretation, and the client relationship. It specifies no survey length, respondent count, or market; those inputs are needed for an actual study quote.
The breakdown
The model allocates $3,000 to programming and $7,500 to reporting within a $20,000 total. The remaining $9,500 covers QA, data processing, coding, weighting, and coordination: $20,000 − $3,000 − $7,500.
| Production task | Illustrative manual workflow | Illustrative automated workflow |
|---|---|---|
| Programming | $3,000 | $500 |
| Reporting | $7,500 | $1,500 |
| Remaining production allocation | $9,500 | $2,150 |
| Total production | $20,000 | $4,150 |
The automated production allocation is calculated the same way: $4,150 − $500 − $1,500 = $2,150. Sample and fielding are excluded from both columns.
Where the 80% hides
In this particular model, the difference is $15,850. Dividing it by the $20,000 baseline gives 79.25%, rounded to about 80%. That describes the model's production-cost difference. It does not establish that 80% of every study's cost is waste, or that every agency will achieve that reduction.
The useful question is which repeated tasks consume effort in your workflow. Programming, table production, coding, and report preparation can include routine handoffs that automation helps reduce. Methodology, quality review, interpretation, and recommendations still need appropriate expert ownership.
What it costs when the relay collapses
The illustrative comparison assumes less manual effort on programming and reporting while retaining review. Actual effort depends on study complexity, existing templates, required outputs, and the team's review process. Automated output still needs checking; the model is not evidence of identical quality in every implementation.
Use survey automation to examine the supported workflow and evaluate it on a questionnaire your team understands. Record time and corrections at each handoff before estimating a saving. Production automation does not itself guarantee a fieldwork deadline.
The point isn't one study
A smaller production burden may make more client questions feasible. It does not automatically mean more revenue, identical deliverables, or a particular annual multiplier. The Wunderkind utilization example reports 24+ surveys by August 2025 while distinguishing them from the customer's previous annual anchor reports.
Current platform pricing is quote-based, determined by survey length and respondent count. There are no separate seat or feature fees; data ingestion, such as exposure records or pixels, can incur additional charges. Sample-sourcing arrangements and other terms belong in the quote. The pricing page provides the current explanation; the labor model above is separate from those fees.
Run it on your own numbers
Record your programming, QA, data processing, reporting, and coordination effort for a comparable study. Keep sample and fielding separate, use consistent task boundaries, and include reviewer time. Replace these illustrative allocations with your own documented inputs before using the result in a client proposal.
That makes the comparison reproducible and relevant to the agency's workflow, while leaving the study design, interpretation, and client recommendations with the team.