No competition
- Opening
Creates urgency by highlighting lack of competition in elections.
Boston harbor scene with text overlay about uncontested races
Blurry figure with American flag and text on primary ballots
City square with diverse people and text on inclusive voting
High-angle street view showing pedestrians and vehicles with campaign text
Close-up of feet walking toward a ballot box with text on voter participation
Close-up of American flag with 'Vote Yes on Question Three' text
This take — "No competition, no accountability" read by Danielle — held viewers unusually well once they started watching, but the opening failed to stop the scroll at a rate well below average, and almost no one shared, saved, or commented. The overall Resonance score of 44 (range 40–49) sits below the reference average of 50.
What happened
Of every 100 people who saw the ad, about 19 started a 3-second play — low against the reference (component:stop, score 25). Of those who did start watching, the average viewer got through about 35% of the 43-second video, which is well above average (component:hold, score 86; metric:avgWatchSec). But shares, saves, and comments combined came to just over 1% of plays — very low (component:act, score 9; metric:shares, metric:saves, metric:comments). The score moved only one point between the two reporting dates, so it has stabilized (history).
What the score means
Resonance is 25% scroll-stop + 40% watch-through + 35% shares/saves/comments, measured against a frozen reference where 50 is average (scale). A 44 means the hold was strong but the other two components pulled it down — especially Act, which is the heaviest drag. This score does not measure whether anyone was persuaded or changed their view on Question 3; it measures paid attention and engagement only.
Evidence
The overall Resonance score is 44 (range 40–49), sitting just below the reference average of 50. The score is built from three parts: Engage (25% weight) scored 25, Interest (40% weight) scored 86, and Act (35% weight) scored 9. The score measures paid attention and engagement against a frozen reference — it does not measure whether viewers were persuaded or changed their minds about Question 3.
Interest is the clear standout: the average viewer watched about 15 seconds of a 43-second video, which works out to roughly 35% of the runtime — well above the reference average (score 86 out of 100). Scroll-stopping was weak: only about 19 in every 100 impressions turned into a 3-second play, scoring 25. Sharing, saving, and commenting together came to just over 1 action per 100 plays, scoring 9.
Compared with the other takes using the same script, this version (Danielle, calm framing) scores 44, while the Ruth version of the same framing scores 50 and the Stephen version scores 63. Across all nine takes in the reference set, Danielle's scores on the other two scripts are also the lowest for her messenger slot.
One reading of the Interest score is that viewers who did stop found the content worth staying with — the script moves steadily from a concrete statistic through a clear explanation of Question 3 to a direct ask, which may have rewarded the viewers who gave it a chance. This is a reading, not a certainty.
One reading of the low Engage score is that the opening — a Boston harbor shot with a text overlay about uncontested races — did not stand out enough in a fast-moving feed to stop scrolling at scale. The video review notes the opening creates urgency through the statistic, but urgency in the script does not automatically translate to a visual hook. This is a reading, not a certainty.
A short-cut version of this take may improve the Act score: the Stephen short-cut on a different script scored 91 versus 48 for the full-length version. It is worth testing whether trimming this 43-second video — possibly cutting after line:3 at 25.6s — lifts sharing and saves without losing the Question 3 name-check that makes this take distinctive.
Not enough evidence yet — the next useful measurement
- A/B test of the opening line against a faster or more visual hook would show whether the low scroll-stop (component:stop, score 25) is about the framing or the visual (seen:0).
- A comment-sentiment read on the 31 comments (metric:comments) would show whether low Act reflects disengagement or a quiet-but-convinced audience.
- Audience-segment breakdown by age or geography would show whether the calm delivery (take framing) resonates differently across the people this campaign most needs to reach.
Next test
Test a short-cut version of this take, trimming runtime to roughly 25–30 seconds.
Proposed- What prompted it
- The Stephen short-cut version of a different framing scored 91 vs. 48 for the full-length version — the largest single gap in the reference set.
- What remains uncertain
- Whether the runtime effect holds for Danielle's calmer delivery and this specific framing.
- How to tell
- Compare Interest (hold rate) and Act score between full and short-cut versions at matched spend.
Test the same script with Ruth as messenger instead of Danielle.
Proposed- What prompted it
- Ruth outscored Danielle on both other framings: calm (50 vs. not tested) and coin (58 vs. 39) and listen (65 vs. 35).
- What remains uncertain
- Whether the messenger gap persists on this calmer, Question-3-named framing or narrows because the framing itself carries more weight.
- How to tell
- Compare overall Resonance score and Act component between Danielle and Ruth versions at matched spend.
Test a version without captions, matching the no-captions condition already run for Ruth on the coin framing.
Proposed- What prompted it
- The coin-Ruth no-captions version scored 49 vs. 58 with captions — a drop, but the archive research flags text-on-screen as the only coding with a negative confidence interval.
- What remains uncertain
- Whether removing captions hurts reach among sound-off viewers more than it helps by reducing on-screen text clutter.
- How to tell
- Track 3-second play rate (stop) and watch-through rate (hold) with and without captions at matched spend.
Forecast
A guess, written down before the numbers come in: will this video do better or worse than usual? When the numbers arrive, PAX2 checks the guess against them, and every check makes the next guess sharper.
Every round on file is already loaded; forecasting reopens when a new round is expected.