Calibration - Battleship Potemkin (1925)
Screen this public-domain film excerpt with the standardized 12-agent calibration panel. Focus on immediate viewer response to the visual and audio material in the attached clip.
Main insights: Willingness to continue was low across segments; a minority reported being “pulled in” despite distress, and craft‑aware viewers admired the visual composition yet still opted out. Credibility split: many labeled it stylized/theatrical with no true host to trust, while a smaller group found the scenario believable enough to feel credible.
Takeaways:
- Add a neutral content advisory and brief context slate, and overlay/translate intertitles to reduce surprise and comprehension friction
- Reposition under Film History/Education rails and de-target family/comfort cohorts; aim distribution at cinephile/education/history segments
- Rapid A/B framing tests; track CR3, Watch‑More Intent, and early‑exit to validate packaging before scaling
Overall appeal
n=12"It was hard to watch. Too much killing and panic, and that woman and child made it worse. Not something I found appealing."
"Watching it, I found the content very unappealing. The violence felt brutal and chaotic, especially with defenseless people involved, and it gave me a strong sense of disorder and danger rather than anything engaging or..."
"I found it hard to call appealing. It was visually powerful, yes, but the suffering and violence made it more upsetting than enjoyable for me."
"Here's the thing... I found it really upsetting to watch. The violence and the scenes with the child and baby carriage were not appealing to me at all, and it felt like the opposite of calm, manageable content."
Watch intent
n=12"What I watched felt very distressing and chaotic, especially with a child caught in the violence. I can respect its seriousness, but for me it creates too much emotional friction to want to keep watching."
"Not very likely. What I watched was rough - too much shooting, panic, and people dropping, especially women and kids. That's not something I'd choose to sit with much longer."
"I probably would not keep going. What I watched had some punch to it, but the silent black-and-white format and untranslated title cards make it feel like work, not something I'd choose to sit with much longer."
"What I watched had real force, and I can respect how plain and effective the storytelling is without a lot of fuss. But the violence against innocent people is hard to sit with, so for me it lands more as something I adm..."
Host credibility
n=12"What I watched did not really present a host, speaker, or outside source for me to evaluate. It was just the film excerpt itself, so my reaction lands neutral on credibility."
"What I watched felt intentionally dramatic and staged, with very theatrical expressions and those forceful silent-film title cards. It may be artistically effective, but as a source it did not come across as especially c..."
"Didn’t feel very credible to me. What I watched was clearly staged and dramatic, with those silent-film title cards, so it came off more like a story being pushed than a plain, trustworthy source."
"What I watched felt serious and believable. The black-and-white footage, the intertitles, and the scenes of panic and people being shot all came across as harsh in a way that felt real, not dressed up."
Sam Norstrom
62 · Rural, NE, USA · Driver
Peace Evangelista
31 · Somerville, MA, USA · Human Resources Specialist
Sandra Falcinelli
61 · Rural, PA, USA · Designer
Kaila Smith
29 · Ann Arbor, MI, USA · Business Operations Specialist
Daniel Sassaman
55 · Rural, LA, USA · Engineer
Precious Rai
40 · Rural, IL, USA · Medical Records Specialist
Brent Guevara
52 · Fort Myers, FL, USA · Personal Care Aide
Gregory Cumbo
60 · Rural, OH, USA · Brokerage Clerk
Brianna Chapman
32 · Rural, WV, USA · Hairdresser and Cosmetologist
Maribel Miller
35 · Rural, NH, USA · Retail Sales Supervisor
John Grimm
62 · Rural, IA, USA · Civil Engineer
Mario Bockus
58 · Rural, VA, USA · Retail Sales Supervisor
Sam Norstrom
62 · Rural, NE, USA · Driver
Peace Evangelista
31 · Somerville, MA, USA · Human Resources Specialist
Sandra Falcinelli
61 · Rural, PA, USA · Designer
Kaila Smith
29 · Ann Arbor, MI, USA · Business Operations Specialist
Daniel Sassaman
55 · Rural, LA, USA · Engineer
Precious Rai
40 · Rural, IL, USA · Medical Records Specialist
Brent Guevara
52 · Fort Myers, FL, USA · Personal Care Aide
Gregory Cumbo
60 · Rural, OH, USA · Brokerage Clerk
Brianna Chapman
32 · Rural, WV, USA · Hairdresser and Cosmetologist
Maribel Miller
35 · Rural, NH, USA · Retail Sales Supervisor
John Grimm
62 · Rural, IA, USA · Civil Engineer
Mario Bockus
58 · Rural, VA, USA · Retail Sales Supervisor
| Age bucket | Male count | Female count |
|---|
| Income bucket | Participants | US households |
|---|
Brianna Chapman
32 - Rural, WV
"Oh my God, that poor woman and baby, it just breaks your heart to see something like that."
Mario Bockus
58 - Rural, VA
"My god, it's happening, they're just killing everyone, even with that poor baby in her arms."
Sandra Falcinelli
61 - Rural, PA
"This scene is just heartbreaking, seeing folks on the ground and soldiers aiming their rifles like that."
Brent Guevara
52 - Fort Myers, FL
"Just reading those words, 'my boy is very sick,' it keeps that bad feeling right in my stomach."
Daniel Sassaman
55 - Rural, LA
"It's just a terrible thing, seeing innocent people like that, staring down rifle barrels."
Gregory Cumbo
60 - Rural, OH
"This poor woman, it's just too much. I don't know why they're showing this."
John Grimm
62 - Rural, IA
"Seeing all those people down, and that woman holding a child - that's a hard thing to look at."
Kaila Smith
29 - Ann Arbor, MI
"This is just awful, especially with the baby right there."
Maribel Miller
35 - Rural, NH
"It's just sickening to see them pointing rifles at that mother and her child like that, with all those bodies around."
Peace Evangelista
31 - Somerville, MA
"It is extremely upsetting to see such a dire situation involving a child and what appears to be widespread conflict."
Precious Rai
40 - Rural, IL
"It just looks so awful, a mother with her child and all those people fallen like that."
Sam Norstrom
62 - Rural, NE
"That's a real hard thing to look at."
Overview
Key Segments
| Segment | Attributes | Insight | Supporting Agents |
|---|---|---|---|
| Rural adult women | Ages ~29–61; occupations varied (HR specialist, designer, cosmetologist, retail supervisor, medical records); lower-to-mid incomes; renters/homeowners in rural areas | Consistently report the clip as upsetting and unappealing due to scenes featuring women/children; aesthetic recognition does not overcome strong aversive reaction, yielding low willingness to continue watching. | Peace Evangelista, Sandra Falcinelli, Kaila Smith, Precious Rai, Brianna Chapman, Maribel Miller, Gregory Cumbo |
| Older rural men | Ages ~55–62; occupations include drivers, engineers, retail supervisors; mid-to-higher incomes; predominantly White | Describe the excerpt as grim or tiring; silent-film format sometimes read as 'work' or formal exercise, contributing to low interest in continuing; several note stylization reduces perceived credibility. | Sam Norstrom, John Grimm, Daniel Sassaman, Mario Bockus |
| Higher-education / craft-aware respondents | Bachelor’s or graduate degrees; roles in design, engineering, management; generally older | Articulate the film’s composition and dramatic staging and often admire its formal qualities, yet that craft awareness coexists with emotional aversion that suppresses viewing intent; credibility tends to be judged as theatrical/stylized. | Sandra Falcinelli, Daniel Sassaman, John Grimm, Mario Bockus |
| Lower-income, Spanish-speaking respondent | Male, 52, Hispanic/Spanish language speaker, low income, personal care aide | Strong affective response to violence focused on the baby carriage and caregiving imagery; expressed low willingness to continue and foregrounded emotional impact over formal considerations. | Brent Guevara |
| Younger/professional urban women | Ages ~29–31; bachelor’s degrees; early-career professionals in urban areas | Also find the clip distressing and unlikely to continue; some retain a higher sense of credibility despite upset, signaling that younger urban viewers may separate emotional impact from perceived realism. | Kaila Smith, Peace Evangelista |
| Engaged despite distress (anomalous engagement) | Mixed ages and incomes; includes a bachelor’s-level older male and a lower-income younger female cosmetologist | A minority report being 'pulled in' or highly affected in a way that suggests emotional engagement without commitment to continue viewing-this profile may indicate strong momentary attention but low tolerance for sustained exposure to distressing historical imagery. | Mario Bockus, Brianna Chapman |
Shared Mindsets
| Trait | Signal | Agents |
|---|---|---|
| Emotional distress about violence involving women/children | Nearly all respondents highlighted mothers, children, or the baby carriage as the primary upsetting elements driving negative reactions and decisions not to continue. | Sam Norstrom, Peace Evangelista, Sandra Falcinelli, Kaila Smith, Precious Rai, Brent Guevara, Gregory Cumbo, Brianna Chapman, Maribel Miller, John Grimm |
| Low intent to continue watching | Emotional aversion and, secondarily, format fatigue (silent, black-and-white, intertitles) are the dominant reasons participants would stop viewing. | Peace Evangelista, Sam Norstrom, Daniel Sassaman, Sandra Falcinelli, Precious Rai, Kaila Smith, Brent Guevara, Brianna Chapman, Maribel Miller, John Grimm, Gregory Cumbo |
| Recognition of formal/visual power | Many participants can identify and sometimes admire composition, staging, and dramatic force-even while reporting emotional repulsion that reduces appetite for continued viewing. | Sandra Falcinelli, Daniel Sassaman, Mario Bockus, Brianna Chapman, Maribel Miller, John Grimm |
| Perception of stylization/theatricality lowering credibility | Repeated references to silent-film acting and intertitles lead several respondents to interpret the excerpt as staged, which makes it feel less like a credible representation and more like an artistic dramatization. | Sandra Falcinelli, Sam Norstrom, Daniel Sassaman, Brianna Chapman, Brent Guevara, Maribel Miller, Gregory Cumbo, Mario Bockus |
Divergences
| Segment | Contrast | Agents |
|---|---|---|
| Craft-aware admiration vs. viewing avoidance | Higher-education respondents articulate and admire formal qualities but still report low willingness to continue, separating aesthetic appreciation from viewing tolerance. | Sandra Falcinelli, Daniel Sassaman, John Grimm, Mario Bockus |
| Perceived stylization vs. perceived realism | A majority treat the piece as stylized/theatrical and less credible, while a minority (notably one older graduate-educated respondent) found the underlying situation believable and credible. | Sandra Falcinelli, Sam Norstrom, Daniel Sassaman, Brianna Chapman, Brent Guevara, Maribel Miller, Gregory Cumbo, Mario Bockus, John Grimm |
| Emotional engagement without sustained viewing | Some respondents report being emotionally 'pulled in' (strong momentary engagement) yet still indicate they would stop watching-an ambivalent signal separating immediate attention from continued consumption. | Mario Bockus, Brianna Chapman |
Overview
Quick Wins (next 2–4 weeks)
| # | Action | Why | Owner | Effort | Impact |
|---|---|---|---|---|---|
| 1 | Add neutral content advisory and context slate | Reduces surprise and sets expectations (historical dramatization, depictions of civilian harm) to mitigate immediate aversion without politicizing. | Editorial/Programming | Low | High |
| 2 | Translate/overlay intertitles + readable captions | Removes comprehension friction that viewers called out as making the clip feel like work; boosts perceived clarity and accessibility. | Localization/Post-Production | Low | Med |
| 3 | Reposition under Film History/Education rails | Aligns viewer mindset with purpose (learn/understand vs. relax/entertain), improving tolerance for distressing imagery. | Content Operations/Programming | Low | Med |
| 4 | Retitle/thumbnail to emphasize context over shock | Shift from sensational imagery (e.g., baby carriage) to historical framing to lower immediate aversion and credibility concerns. | Marketing/Creative | Low | Med |
| 5 | De-target family/general comfort cohorts | Panel signals broad avoidance among rural adults and family-minded segments; reduce unintended exposure that drives churn. | Growth/Personalization | Low | High |
| 6 | Rapid A/B of framing variants (advisory vs. none; history vs. entertainment) | Quantify uplift on continue rate and watch-more intent within two weeks to lock the highest-ROI packaging. | Research & Analytics | Low | High |
Initiatives (30–90 days)
| # | Initiative | Description | Owner | Timeline | Dependencies |
|---|---|---|---|---|---|
| 1 | Context-first packaging for archival violence | Create a repeatable template: 10–15s on-screen context slate, optional 15–30s host VO from a credible film historian, scene-level chaptering (incl. baby-carriage marker), and a brief "Why it matters" end-card. Preserve original cut while offering a contextualized version. | Programming + Creative Production | 6–8 weeks for V1 across priority titles | Talent booking or VO license, Standards & Practices review, CMS support for dual versions and chapter markers, Legal approval of historian claims |
| 2 | Accessibility and modernization pass | Standardize English intertitle overlays, improved captions, and an optional period-appropriate score/narration toggle to reduce fatigue without altering the source. Maintain a clearly labeled Original track. | Post-Production/Localization + Player Engineering | 8–10 weeks for tooling + first batch | Audio mixing resources, Public-domain/commissioned music rights, Player UI toggle support, QA for timing/sync |
| 3 | Audience segmentation and distribution rules | Define cinephile/education segments; suppress broad-family placement. Prioritize recommendations within Film/Media Studies, Documentary, and History affinities; exclude relaxation/comfort rows. | Data Science/Personalization | 6–8 weeks for models and policy rollout | Viewer taxonomy updates, Experimentation platform, Programming policy documentation, Compliance review |
| 4 | Distress-to-Engagement measurement framework | Operationalize a "Distress vs. Engagement Index" using panel scores + on-platform signals (early exits, skip to chapter, complaint rate, survey Top-2-Box). Set greenlight thresholds for packaging choices. | Research & Analytics | 4–6 weeks to stand up dashboards and thresholds | Expanded sample beyond 12-agent panel, Post-view micro-survey integration, Data pipeline instrumentation |
| 5 | Safety and sensitivity editorial guidelines | Codify neutral advisory language, chaptering standards for distressing scenes, and review checkpoints for content featuring women/children in harm. Train editors and QC teams. | Standards & Practices + Editorial | 3–4 weeks for policy + training | Legal review, DEI advisory input, Training materials |
| 6 | Educator and partner pilot | Bundle contextualized clips with short explainers and discussion guides for film/history programs and streaming education channels to reach high-intent audiences. | Partnerships/BD | 8–12 weeks to secure 2–3 pilots | Partner outreach, Supplemental material rights, Co-marketing approvals |
KPIs to Track
| # | KPI | Definition | Target | Frequency |
|---|---|---|---|---|
| 1 | Continue Rate to Minute 3 (CR3) | Percent of starts that reach 3:00. Baseline from raw clip; measure uplift after packaging changes. | +20% vs. baseline within 60 days | Weekly |
| 2 | Watch-More Intent (WMI) | Post-view survey Top-2-Box to 'How likely would you be to watch more?' on this title set. | +15 percentage points vs. control framing | Biweekly |
| 3 | Perceived Credibility Lift | Top-2-Box agreement that the host/source felt credible (or that historical context was trustworthy). | +10 percentage points after adding context slate/host VO | Monthly |
| 4 | Early Exit/Complaint Rate | Share of viewers exiting before 60s or filing a content complaint/report. | -30% vs. baseline after repositioning and advisories | Weekly |
| 5 | Targeting Precision | Percent of impressions delivered to defined cinephile/education/history segments. | ≥80% of impressions within target cohorts | Monthly |
| 6 | Context Element Engagement | Percent who view context slate/host intro without skip; completion rate of context segment. | ≥60% viewed, ≥50% completed; ≤20% skip rate | Weekly |
Risks & Mitigations
| # | Risk | Mitigation | Owner |
|---|---|---|---|
| 1 | Perception of 'sanitizing' or altering a classic with added context/score. | Offer both versions with clear labels (Original and Contextualized); default by audience segment. | Editorial/Standards & Practices |
| 2 | Small-panel bias (12-agent) misguides broader decisions. | Triangulate with on-platform A/B data and expand sample before hard policy changes. | Research & Analytics |
| 3 | Context intros increase early drop-off. | Cap intros at 10–15s, front-load value, and A/B test placement (pre-roll vs. mid-roll chapter card). | Product/UX |
| 4 | Rights or cost constraints for music/narration tracks. | Use public-domain or commissioned assets with clear rights; phase rollout by ROI. | Legal + Post-Production |
| 5 | Over-suppression reduces discovery and limits learning. | Time-boxed tests with guardrails; promote to high-intent rails and partners to balance reach. | Programming/Growth |
| 6 | Backlash to content advisories as political signaling. | Use neutral, descriptive language; apply consistently across archival titles. | Standards & Practices |
Timeline
- Weeks 0–2: Quick wins live (advisory, translations, repositioning, targeting); launch initial A/B framing tests.
- Weeks 3–6: Finalize sensitivity guidelines; stand up Distress-to-Engagement dashboards; iterate on packaging from test results.
- Weeks 6–8: Ship Context-first packaging V1 (context slate, chaptering); begin accessibility/modernization pass on priority titles.
- Weeks 8–10: Player toggle for optional score/narration; expand segmentation policies platform-wide.
- Weeks 10–12: Educator/partner pilots launch; finalize KPI targets and roll into greenlight gates for archival content.
Objective and context
A+E Global screened an excerpt of Battleship Potemkin (1925, public domain) to a standardized 12-agent calibration panel to capture immediate reactions to the visual and audio experience. The analysis focuses on cross-question themes (credibility, emotion, intent to continue), persona-level patterns, and implications for packaging and distribution of archival/historical content.
What we heard: immediate responses
- Dominant driver of aversion: harm to women/children. Across respondents, the baby carriage and scenes involving mothers/children were repeatedly cited as most upsetting, directly suppressing willingness to continue (reported by rural adult women, older rural men, younger/professional urban women, and a Spanish-speaking lower-income male).
- Format fatigue reduced tolerance. The silent-film form (black-and-white, intertitles) was described as “work” by several, contributing to low intent to continue among older rural men and others.
- Craft recognized but not converting. Higher-education/craft-aware viewers articulated admiration for composition, staging, and dramatic force, yet this did not translate into viewing persistence.
- Credibility split on stylization. A plurality framed the scene as theatrical/staged (silent-era acting, intertitles), modestly lowering perceived realism; a minority still found the underlying situation believable.
- Emotional engagement without continuation. A small subset reported being “pulled in” despite discomfort-signaling strong momentary attention that did not map to continued viewing intent.
Persona patterns and divergences
- Rural adult women (29–61): Upset by women/child imagery; aesthetic recognition did not overcome aversion, producing low willingness to continue.
- Older rural men (55–62): Described the clip as grim/tiring; silent format read as homework; stylization reduced credibility and interest.
- Higher-education/craft-aware: Praised formal qualities but maintained low continuation; separated aesthetic appreciation from tolerance.
- Lower-income, Spanish-speaking male (52): Strong affect centered on the baby carriage/caregiving; low willingness to continue; emotional impact eclipsed formal considerations.
- Younger/professional urban women (29–31): Distressed and unlikely to continue, yet more likely to deem the situation credible-indicating realism can coexist with avoidance.
- Anomalous engagement: Minority “pulled in” despite distress, evidencing attention spikes without sustained consumption.
Implications and recommendations
For archival/historical titles depicting civilian harm, context-first packaging, targeted distribution, and accessibility upgrades are required to convert momentary attention into sustained viewing while preserving integrity.
- Quick wins (low effort/high impact): Neutral content advisory and context slate; translate/overlay intertitles and improve captions; reposition under Film History/Education rails; retitle/thumbnail for historical framing vs. shock; de-target family/comfort cohorts.
- Initiatives: Context-first packaging template (10–15s on-screen context; optional 15–30s historian VO; chaptering including the baby-carriage scene; “Why it matters” end-card). Accessibility/modernization pass (standardized English intertitle overlays, improved captions, optional period-appropriate score/narration toggle with an Original track). Audience segmentation rules (prioritize cinephile/education/history affinities; suppress relaxation/comfort rows). Distress-to-Engagement measurement framework (blend panel scores with exits, skips, complaints, survey Top-2-Box). Safety/sensitivity editorial guidelines for depictions of women/children in harm.
Risks and guardrails
- Perception of “sanitizing” a classic: Offer Original and Contextualized versions; default by audience segment.
- Small-panel bias (n=12): Validate with A/B tests and larger samples before hard policy shifts.
- Context intros increasing drop-off: Cap at 10–15s; A/B test pre-roll vs. mid-roll chapter card.
- Rights/costs for music/narration: Use public-domain/commissioned assets; phase by ROI.
- Over-suppression limiting discovery: Time-boxed tests with guardrails; balance via high-intent rails and partners.
Next steps and measurement
- Weeks 0–2: Ship advisories, intertitle translations/captions, repositioning, and targeting; launch framing A/Bs.
- Weeks 3–6: Finalize sensitivity guidelines; build Distress-to-Engagement dashboards; iterate packaging from tests.
- Weeks 6–8: Release Context-first packaging V1 and scene chaptering; begin accessibility/modernization pass.
- Weeks 8–10: Enable player toggle for optional score/narration; expand segmentation policies.
- Weeks 10–12: Launch educator/partner pilots; codify KPI gates for archival content.
- KPIs: Continue Rate to Minute 3 (target +20% vs. baseline in 60 days); Watch-More Intent Top-2-Box (+15 pts vs. control); Perceived Credibility Lift (+10 pts post-context); Early Exit/Complaint Rate (−30% post-repositioning/advisories); Targeting Precision (≥80% impressions to cinephile/education/history segments).
-
Please rate your agreement with each statement about this clip: (a) I could follow what was happening without additional context. (b) Reading the intertitles/on-screen text felt like effort. (c) The black-and-white visuals reduced my engagement. (d) The camera work was compelling. (e) The pacing felt too slow for me. (f) The violent content felt excessive for me.matrix Identifies specific friction points to decide on context slates, edits, or pacing tweaks.
-
Which emotions did you feel while watching this clip? Select all that apply.multi select Maps emotional profile to inform targeting, advisories, and positioning tone.
-
How did the music score affect your engagement with the clip?semantic differential Assesses audio’s contribution to guide soundtrack selection, mix, or removal.
-
Please prioritize which presentation additions would make you more open to viewing similar clips: brief historical context slate; neutral content advisory; modern captions translating intertitles; optional voice-over narration; less intense music score; none of these.maxdiff Ranks mitigation features to prioritize in the next cut.
-
Under which framing would you be most open to viewing similar content: film history/education, classic cinema curation, social history/propaganda analysis, general entertainment, or none?single select Guides content rail placement and audience targeting.
-
What is the maximum clip length you would be comfortable watching for this type of content in one sitting? Please answer in minutes.numeric Sets acceptable duration guidelines for packaging and programming.
Main insights: Willingness to continue was low across segments; a minority reported being “pulled in” despite distress, and craft‑aware viewers admired the visual composition yet still opted out. Credibility split: many labeled it stylized/theatrical with no true host to trust, while a smaller group found the scenario believable enough to feel credible.
Takeaways:
- Add a neutral content advisory and brief context slate, and overlay/translate intertitles to reduce surprise and comprehension friction
- Reposition under Film History/Education rails and de-target family/comfort cohorts; aim distribution at cinephile/education/history segments
- Rapid A/B framing tests; track CR3, Watch‑More Intent, and early‑exit to validate packaging before scaling
| Participant | Response | Actions |
|---|