Picked by QualifierEvidence by Numbers

The buyer

TED Talks

brand · X @tednews (verified) · from the TED Talks × Hacks thesis →

Why this side

TED Talks is the right side to approach for audience data because it is a verified brand and studio whose audience already strongly likes Qloo's data. Hacks is kept as evidence only because it is a television show with an unverified public contact route, though its audience also shows strong affinity.

never contactedHacks (TV show) is evidence only.

0.90
Hacks fans → TED Talks
0.82
TED Talks fans → Hacks
Pair score0.61WATCH

Who the audience is

TED Talks audience by age

■ over-indexes (right) · ■ under-indexes (left), relative to Qloo's baseline

Under 25-0.06
25–29-0.11
30–34-0.69
35–44+0.06
45–54+0.19
55++0.03

Where

Cities where both audiences are strong

◎ Los Angeles

TED Talks's strongest US cities

San JoseWashington DCNew YorkLos AngelesBuffaloSan DiegoOaklandSeattle

Taste the two audiences share

indieindie rockelectronicalternative rockrock

Three more partners Qloo found for TED Talks

The pitch sells Qloo's breadth, not one deal. Affinity out = TED Talks fans → partner; back = partner fans → TED Talks.

One Strange RockTV show
out 0.94back 0.91
HorizonTV show
out 0.95back 0.90
Avishai Cohenartist
out 0.91back 0.85

Outreach

9 found, 5 verified. Only verified contacts get a message. See the plan →

Qloo taste data (hackathon API). Aggregate audience affinity; not causal. Picked by the Qualifier and assembled by the Numbers agent with no human review.