The single most lucrative slide in the activist canon. McDonald's $46bn real estate. Darden's $67–86 SoP vs. $48 today. Phillips 66's $183 vs. $103. The "big number" the deck is engineered to deliver.
See
patterns/sum-of-parts.mdfor the rhetorical logic. This file covers the physical slide build.
The 2-slide structure
The canonical SoP reveal splits across two slides:
Slide 1 · The build-up table
Multiple EBITDA EV
Segment Comp set (median) (LTM, $M) ($M)
─────────────────────────────────────────────────────────────────
Real estate Mall REITs (4) 14.5x $1,820 $26,390
Royalty stream Franchise QSR (3) 23.0x $890 $20,470
McOpCo Restaurant ops (5) 7.5x $1,210 $9,075
─────────────────────────────────────────────────────────────────
Sum $55,935
Less: net debt at parent ($8,200)
Less: separation friction (5%) ($2,800)
─────────────────────────────────────────────────────────────────
Sum-of-parts equity value $44,935
÷ shares outstanding (1.05B)
PER SHARE $42.80
━━━━━━━
Current share price $33.00
Implied upside +30%
Critical features:
- Net debt and friction explicitly subtracted — pre-empts "you ignored stranded costs"
- Multiple median, not max — show the range in a footnote, but use median for the headline number
- One row per segment, max 4–5 segments — more becomes a spreadsheet
- Source line for each comp set (10-K segment data + peer 10-Ks)
Slide 2 · The "big number" reveal
The visual punchline. One slide, full page.
┌─────────────────────────────────────────────────────────┐
│ │
│ │
│ │
│ │
│ $42.80 │
│ per share │
│ │
│ vs. $33.00 today │
│ │
│ │
│ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ │
│ │
│ The market values McDonald's as a restaurant operator. │
│ It owns $46bn of real estate. │
│ │
└─────────────────────────────────────────────────────────┘
Specifications:
- Number is enormous (160–200pt serif) — the only thing on the slide
- Comparison to current in 24pt below (smaller, neutral colour)
- One-line explanation in 18pt italic serif — the thesis distilled
- No chart, no source line (sources are on the build-up slide)
Variations
A · The bar chart variant (current vs. SoP, side by side)
$50 ┤ ╔═════════════════════╗
│ ║ $42.80 ║
│ ║ Sum-of-parts ║
│ ║ ║
│ ╠═════════════════════╣
$40 ┤ ╔═══════════════╗ ║ Real estate $25 ║
│ ║ $33.00 ║ ║ ║
│ ║ Current ║ ║ ║
│ ╠═══════════════╣ ╠═════════════════════╣
$30 ┤ ║ Restaurant ║ ║ Royalty stream $14 ║
│ ║ operator ║ ║ ║
│ ║ multiple ║ ╠═════════════════════╣
│ ║ ║ ║ McOpCo $4 ║
│ ╚═══════════════╝ ╚═════════════════════╝
│
│ Today Sum-of-parts
└───────────────────────────────────────────────────
Used by: Pershing Square / McDonald's (2005), Trian / DuPont (2015), Elliott / Phillips 66 (2025). Stacked-bar segmentation makes the "hidden value" visually concrete.
B · The scenario table
When the SoP has multiple plausible outcomes (depending on multiple assumptions), use a 3-column table: Bear / Base / Bull.
Scenario Bear Base Bull
─────────────────────────────────────────
Real estate (×) 12.0x 14.5x 17.0x
Royalty (×) 18.0x 23.0x 28.0x
McOpCo (×) 6.0x 7.5x 9.0x
─────────────────────────────────────────
Per share $36 $42.80 $50.20
─────────────────────────────────────────
Upside vs. $33 +9% +30% +52%
Adds defensibility — sceptics see you've considered the downside.
C · The conglomerate-discount overlay
Add a horizontal reference line on the bar chart marking the average conglomerate discount (~15–25% historically). Frames the SoP gap as the discount the market is currently pricing in.
Common mistakes
- Optimistic multiples without justification. If you assigned peer-MAX multiples instead of median, your number is academic. Use median + sensitivity.
- Ignoring stranded costs. Spinning costs $20–60M annually in stranded overhead for a typical mid-cap. Subtract it.
- Hiding the build-up table. Some decks put SoP only as the big number — sceptics need the build to verify. Show both slides.
- Per-share number rounded too aggressively. "$42.80" is more credible than "$45" even if your underlying math has identical precision. The decimal signals work was done.
- Forgetting net debt. Recurring error. Always start from EV, end at equity, divide by share count.
Exemplars
- Pershing Square · McDonald's (Nov 2005) — the masterclass. Two-slide structure with build-up table + reveal.
- Pershing Square · Howard Hughes Sohn (May 2017) — single-slide reveal, $200+ per share thesis.
- Trian · DuPont (Feb 2015) — 6-segment SoP, scenario table variant.
- Trian · PepsiCo (Jul 2013) — snacks-vs-beverages reveal with precedent multiples cited.
- Elliott · BHP Billiton (Apr 2017) — DLC + petroleum SoP, single reveal.
- Elliott · Phillips 66 (2025) — modern variant, $183 per share big-number with stacked-bar segmentation.
See also
patterns/sum-of-parts.md— when to use this patternvaluation/sum-of-parts.md— the analytical frameworkslides/peer-gap-chart-recipe.md— the sibling visual (used to defend each segment's chosen multiple)slides/visual-craft.md— the "big number" principle