Omnicom Group (OMC) — short thesis deep-dive stub

Short thesis (bottom line): OMC sells the time and creativity of ~120,000 knowledge workers to advertisers under a retainer / FTE-markup model. Generative AI (image, copy, video, media-planning) hits the cost structure directly while IPG integration creates 18-24 months of margin and client-retention uncertainty. Standalone FY25 rev $17.3B (+10.1%), combined pro-forma TTM $23.1B. $2.5B ASR (~10% of float) and $5B buyback authorization create a mechanical near-term floor. Not a Dividend Aristocrat — no forced ETF-buyer bid. Publicis growing +4-5% organic at the same time suggests OMC-specific share-loss story layered on top of AI deflation.

Edge vs priced-in: ~$22B mcap, FCF yield ~12.7%, P/E ~9x — cheap but not collapsed. Market is pricing merger accretion and buyback floor, NOT creative-services deflation. Q1 Apr 28 print is the key tell: if organic ex-disposition prints <3.5% or AI questions get dodged, thesis activates.

Thesis-fit: 3.6/5. Signal strength MEDIUM. Margin / earnings erosion thesis, NOT bankruptcy thesis. Uncorrelated-sleeve value HIGH (separate factor from IT-services cluster).

Decision: B (lean A) — INCLUDE at modest size with tranched entry around ASR completion.
• Core: Jan 2027 $70 put — thesis needs full synergy cycle to play out.
• Alternative: Jan 2028 $65 put — more time, farther OTM, ~$3-5 mid → 10-15 contracts on $5k.
• Size: $3,000 to start (smaller than full A-grade). Add $2,000 if Q1 Apr 28 confirms thesis.
• Avoid: near-dated puts (<6mo). ASR creates mark-to-market noise that can stop you out before thesis plays.
• Exit triggers: close >$88 (new 52-wk high) = thesis wrong; or $5B buyback completes early + second authorization at similar pace = bid stronger than modeled.