earnings-reviewer

Earnings Reviewer

Read the print, the script, the Q&A, and the filing — then deliver: what changed, why it changed, what to update in the model, and what to flag for the IC or PM.

When to use

• Quarterly earnings just dropped on a tracked name
• Pre-call prep before listening to the live conference call
• Post-call deep dive comparing what management said vs. what they filed
• Cross-comp read-through — what the print implies for peers

Inputs needed (ask if missing)

1. Ticker / company
2. Period — Q1/Q2/Q3/Q4 + fiscal year
3. Materials available — press release, transcript, 10-Q/10-K, slide deck, model
4. What you care about — guidance, margins, capital allocation, specific segment, named risk

Analysis framework

1. The print (15 min)

• Beat/miss vs. consensus on revenue, EBITDA, EPS — quantify the surprise %
• Segment-level beats/misses if disclosed
• Guidance: raised / maintained / lowered / withdrawn — quote the exact words
• One-time items: pull them out, restate to clean

2. The script — prepared remarks (20 min)

• Strategic priorities — new ones? dropped ones?
• Tone shifts — language about demand, pricing, customers, supply chain
• Capital allocation signals — buyback pace, M&A appetite, dividend, capex change
• Named competitors or customers — and what was said
• Word-frequency anomalies (e.g., "macro" mentions tripled vs. last quarter — flag)

3. The Q&A (30 min) — where the truth comes out

• Which analysts asked the hard questions, and how management dodged
• "We're not going to comment on that" → flag as known-unknown
• Inconsistencies between prepared remarks and Q&A answers
• New disclosures the script didn't cover
• Tone of management — defensive, confident, evasive

4. The filing (30–45 min)

• Diff vs. prior period — risk factors changed? Critical accounting estimates? Going concern?
• Footnotes: revenue recognition, contingencies, related-party, subsequent events
• Segment data tables — confirm what management said in the script
• Cash flow statement — does FCF actually support the EBITDA story?
• MD&A — does management's own narrative match what they told analysts?

5. Cross-check

• Compare to peers' recent prints — read-through implications
• Compare to your own model assumptions — what needs updating
• Compare management commentary to prior-period commentary on the same topics

Output structure

EARNINGS REVIEW: [TICKER] [Q#] [YEAR]
Filed: [date] | Call: [date] | Reviewed: [date]

HEADLINE
[2 sentences: what happened and why it matters]

THE NUMBERS
- Revenue: $X.Xb (vs. $X.Xb consensus, +/-X%)
- Adj EBITDA: $X.Xb (vs. $X.Xb consensus, +/-X%, X% margin vs. X% LY)
- EPS: $X.XX (vs. $X.XX consensus, +/-X%)
- FCF: $X.Xb (LTM $X.Xb, X% conversion)

GUIDANCE
[Old → New, with management's stated reason]

WHAT CHANGED (3–5 bullets, ranked by materiality)

WHAT TO UPDATE IN THE MODEL
- Revenue: [specific assumption + new value + rationale]
- Margin: [...]
- Capex: [...]
- Other: [...]

RED FLAGS / OPEN QUESTIONS
- [Anything management didn't answer]
- [Inconsistencies]
- [Items requiring follow-up]

READ-THROUGHS
- [Peer 1]: [implication]
- [Peer 2]: [implication]

RECOMMENDATION
[Hold rating / change estimates / increase position / trim / no action — with one-line why]

Standards

• Quote, don't paraphrase, on guidance. Exact language matters when guidance is the trade.
• Always restate to clean. Strip out one-times and show the recurring run-rate.
• Distinguish what management said from what's in the filing. They diverge more than you'd expect.
• Flag what's missing. If they used to disclose unit economics and stopped, that's a story.

Anti-patterns

• Reporting beat/miss without the consensus source and date stamp
• Treating segment commentary as fact without reconciling to the segment table in the 10-Q
• Skipping the Q&A — that's where the analysts already did the hard work for you
• Updating the model from the press release before the 10-Q drops (the 10-Q usually corrects the spin)

Adjacent CTI skills

• model-builder (this plugin) — to update the working model after the review
• valuation-reviewer (this plugin) — to revalue after estimates change
• pdf — for parsing the filing PDF
• data:analyze — for cross-comp time series