// 01, The ad

The opening, on the record.

A Nivea YouTube spot opening with a direct product showcase. Watch the first three seconds, the can is on-screen immediately, which is the strongest thing this opening has going for it.

Nivea · YouTube · DE

Click to play the full ad.

15.1s · YouTube / brand channel
benchmark_only = true · scored 2026-05-18

// 02, The verdict

Composite 67, SHARPEN.

67
Verdict
SHARPEN

Three points below the SCALE floor. Four of five KPIs are Strong; Build Brand at 55 holds the verdict in SHARPEN. The fix is asset variety, not asset volume.

Verdict ladder: SCALE ≥ 70 and no KPI < 55 · SHARPEN 55–69 · REBUILD < 55 OR two+ KPIs < 45.

// 03, The five KPIs

Where the score comes from.

Frozen weights: 25 / 20 / 20 / 20 / 15. Bands: Strong ≥ 70 · Moderate 45–69 · Weak < 45.

// KPI breakdown, Nivea YouTube DE

Beat the Skip
25%
Beat the Skip, White background, NIVEA blue disc-logo in the top-right corner, brand_in_first_3s = yes, full stop.0.5s
75
Strong, brand familiarity carries the watch despite the static opening; Nivea is read instantly even in still frame.
Get Noticed
20%
Get Noticed, Continuing skin-focus shot with the NIVEA mark held, visual hierarchy is clean.2.0s
70
Strong, the blue palette stops the scroll without needing motion energy.
Brand Impact
20%
Brand Impact, The 'Schutz vor DEO-FLECKEN' problem caption, clothing close-up. Brand mark still anchored top-right.4.0s
70
Strong, brand_in_first_3s = yes; the can is the opening composition.
Sell Proposition
20%
Sell Proposition, Product family hero, Black & White Invisible 72h range across women's and Men's variants on a sky-blue stage.MID
70
Strong, benefit-led prop, specific claim about what the product does.
Build Brand
15%
Build Brand, Codes compounding, NIVEA blue, 72h proof point, Black & White range typography all aligned.75%
55
Moderate (rule's lower edge), asset variety is narrow; the same brand cue repeats rather than compounding with new ones.
// 04, Creative classification

What the pipeline tagged.

Hook type
Product-Showcase
Opening motion
Static
Brand in first 3s
Yes
Attention risk
None
Value prop type
Benefit-Led
CTA timing
End
Subcategory
Deodorant
Matrix label
goal
Note on matrix label. The pipeline tagged this ad with matrix_label = goal, which typically corresponds to the SCALE band. However the rule-derived verdict is SHARPEN, composite 67 is below the SCALE floor of 70. This is a minor inconsistency in the scoring runner where the matrix label is computed independently of the composite band rule. The rule-derived verdict (SHARPEN) is authoritative; we surface this as transparency rather than hide it.
// 05, What the model saw

A read of the opening pattern.

A product-showcase open, the can is the first thing the viewer sees. The model treats this as low-risk, low-ceiling: the viewer immediately understands what the ad is selling, but a static opening doesn't earn additional attention beyond what brand familiarity already buys.

Beat the Skip at 75 is unusual for a static opening; it works here because Nivea's brand familiarity is doing the heavy lifting. The model is reading the brand familiarity as if it were motion energy, which is fair for a household brand, but fragile for a less-known one. A challenger brand with this same opening would likely score 10–15 points lower on Beat the Skip.

Build Brand at 55 is the single soft KPI. The cause: the brand cue (logo, blue, can shape) appears once and repeats identically, rather than compounding. Build Brand rewards variety of distinctive asset moments, not volume of the same asset.

// 06, Recommendations

Three edits, ranked by lift.

Nivea ad closing frameEND
// Evidence, the close

End-card product hero, Black & White Invisible 72h with the 'NEU' badge. Full brand, claim and SKU lockup. The opening already does brand_in_first_3s right; the lift here is sharper hook-second visual stakes.

  1. Re-cut, don't re-shoot. Add a face mid-expression or a problem-state beat as frame 1; let the pack shot land at 1.5 seconds. Build Brand and Get Noticed both lift without changing the production.
  2. Introduce a second distinctive asset moment. A sonic ident, a recurring color flash, or a tagline overlay, anything the viewer encounters twice in the cut. Build Brand moves from 55 toward 70.
  3. Test a dynamic-opening variant. The static opening is leaving Beat the Skip potential on the table. A 2-second dynamic intro before the can reveal would lift Beat the Skip by 5–8 points without breaking the showcase strategy.
// 07, Benchmark context

Where this sits in the pool.

Against the SaliencyLab benchmark for YouTube · Beauty · DE, composite 67 sits in the upper-mid band, strong for a static product-showcase, but below the SCALE threshold. Static-opening creative tends to cap around this score even for high-familiarity brands.

Pool-wide, the top quintile by predicted score has shown 6.5× lift over the bottom quintile against public engagement outcomes (held-out OOS, YouTube view counts ρ +0.32, n=403, 5-fold CV as of 2026-05-05).

// 08, What this teaches

The general principle.

A high-familiarity brand can ride on its brand asset and still score Strong on four out of five KPIs. But Build Brand is the one KPI that demands work even from the most recognized logos, because Build Brand isn't about whether the viewer recognizes you, it's about whether the viewer is given multiple distinctive moments to associate with you.

The second teaching is about static openings. They are not broken, but they are capped. For a household brand the ceiling is around composite 67–70; for a challenger it drops 10 points. If you're a household brand looking to scale, a 2-second dynamic intro before the static showcase is almost always the right edit.

Deeper: Brand cue placement, the 1.5-second rule and Reading a RoastIQ verdict end to end.

Honesty footer. Scores from the RoastIQ pipeline 2026-05-18 against model_version gemini-2.5-flash / benchmark_pool_version scoring-runner-v1. Validated against public engagement signals only, held-out OOS Spearman ρ +0.30 to +0.32. Does not predict sales, ROAS, attributed conversion, or brand recall.