PRAPI Research · Experimental finding · Cycle 1

Cycle 1 — what works in B2B PR pitches (preliminary findings)

PRAPI Research Cycle 1 tests three pre-registered hypotheses about pitch framing: external deadline + concrete data point (H1), founder voice (H3), and operational insider data (H5). This page accumulates data toward the 200-pitch-per-arm significance threshold; verdicts appear once the cycle closes.

Published 5/23/2026 · H1, H3, H5

Data accumulating. This cycle is still running. Numbers below refresh as outcomes land; the page will flip to a final report once the n ≥ 200/arm significance threshold clears and the cycle close + 14-day freeze passes.

The cycle in one paragraph

PRAPI Research Cycle 1 tests three pitch-framing hypotheses surfaced from the descriptive survey of 24 practitioners in Report 1. Each hypothesis is pre-registered with locked variant definitions, a 30% minimum-detectable-effect, a sample-size threshold of 200 per arm for significance claims, and a 10-item confounder disclosure list. The analysis pipeline is open-source and the aggregate dataset publishes alongside this page for independent replication.

What we're testing

Full variant definitions and treatment-isolation rules: pre-registration.

Status

This cycle is accumulating. The auto-filled methodology footer below shows the live per-arm sample size from the PRAPI corpus. Numbers refresh each time you load the page. Once the cycle clears the 200-pitch-per-arm threshold AND the 14-day outcome freeze, this page flips to the final report.

What ships when the cycle closes

When the threshold clears, this page becomes the Part 2 findings report:

What does not ship

How to participate

If you're a PRAPI user, you can opt in to the experimental track at app.prapi.dev/pr-pitch/settings#research. Opt-in is reversible at any time; opting out preserves historical rows in the corpus (per the pre-registration) but stops future capture. Per-pitch exclusion is also supported — you can flag any specific pitch as not-for-research from the outcome capture queue.

If you're a journalist or researcher who wants to look at the data, the aggregate CSV is published alongside this page on every cycle.

References

Methodology

Pre-registration

This report tests hypotheses pre-registered at prapi.dev/research/preregistration/cycle-1 before any data was collected. Pre-registration includes hypothesis variant definitions, primary outcome metric, minimum-detectable-effect, sample-size thresholds, and the 10-item confounder list. Any amendments are logged at that URL with timestamps.

Hypotheses tested

Sample

Cycle 1 corpus as of 5/27/2026, 4:00:33 PM 6 pitches across 3 hypotheses.

Hypothesisn (A treatment)n (B control)Composite mean (A)Composite mean (B)n unique usersn unique brands
H1110.000.0022
H310.0011
H5120.000.0022

Primary outcome

Composite engagement score per pitch: 1·replied + 3·quoted + 5·published. See methodology for the secondary outcomes, MDE (30% relative lift), and statistical-discipline rules.

Statistical method

Confounder disclosure

Per-arm distribution across the 10 confounders is captured at outcome write time (per the snapshot rules) and disclosed in the per-hypothesis appendix of the full PDF. Confounders that hit material imbalance (any bucket ≥ 30% of the arm) trigger a stratified sub-analysis reported alongside the headline figure.

Limitations

Data availability

Aggregate dataset for independent replication is published at https://marketing.startvest.ai/api/v1/research/cycles/1/aggregate.csv. Row-level pitch data is never released.

External review

No external review for this report. Per Open Decision #4, external review is optional from day one and disclosed when used.

Contributor citation

Users whose pitches contributed to this cycle's dataset (with opt-in consent) are cited at /research/cycle-1/contributors. Each cited contributor has an opt-in profile in the PRAPI directory.

Replication

The hypothesis variant generator, statistical analysis pipeline, and aggregation cron are open-source at github.com/Startvest-LLC/marketing-agent under src/lib/research/experiments/. Anyone with the published aggregate dataset can reproduce the headline statistics.