…
)”. So we need to include those comments for each paragraph? Probably yes. We must not use placeholders. Must include the e-book promo paragraph at end with given link. We need to count words between 450-500 inclusive. Let’s aim for ~470 words. We need to include title line: “Title: From Numbers to Narrative: AI-Powered Board Deck Explanations for Revenue, COGS, and OpEx Variances”. Must include “AI” and “ai”. Title already includes AI. Need also lowercase “ai”? Could include both: “AI” and “ai”. Title can have both: “From Numbers to Narrative: AI & ai-Powered Board Deck Explanations”. But that seems odd. Maybe include “AI” in title and also somewhere in content we have “ai”. Requirement: “Title: SEO-friendly, include ‘AI’ and ‘ai'”. So title must contain both strings. Could be “From Numbers to Narrative: AI & ai-Powered Board Deck Explanations”. That includes “AI” and “ai”. Good. Now content: We’ll write several paragraphs with the WP block comments. We need to incorporate facts: AI-generated output, example prompts, layers, master prompt template, pro tip, OpEx classification checklist. We must not use placeholders. Write actionable content. Let’s draft about 470 words. We need to count words. Let’s draft then count. I’ll write content with paragraphs. Title line: “Title: From Numbers to Narrative: AI & ai-Powered Board Deck Explanations for Revenue, COGS, and OpEx Variances” Then blank line then HTML. We’ll need to ensure each paragraph is wrapped with…
. Headings maybe…
. Let’s craft. I’ll write:Why Variance Narratives Matter
Board decks live or die by the clarity of their financial explanations. Founders need to see not just that revenue missed budget, but why it happened and what it means for the next quarter. AI can turn raw variance numbers into a concise, three‑sentence story that satisfies both finance rigor and founder readability.
Continue. We need to embed the layers: Layer 1,2,3. Also master prompt template. Let’s write. We’ll need to count words. Let’s draft then count manually. I’ll write in a text editor mentally. Start after title line. Paragraph 1 heading and paragraph. I’ll write content:Why Variance Narratives Matter
Board decks live or die by the clarity of their financial explanations. Founders need to see not just that revenue missed budget, but why it happened and what it means for the next quarter. AI can turn raw variance numbers into a concise, three‑sentence story that satisfies both finance rigor and founder readability.
The Three‑Layer Framework
Chapter 4 of the e‑book introduces a simple stack: Layer 1 captures the raw variance (the numbers); Layer 2 identifies the root cause (internal or external driver); Layer 3 crafts the board‑ready narrative. Keeping each layer distinct prevents mixing data with interpretation and makes AI prompting straightforward.
Master Prompt Template
Use this master prompt for any P&L line:
You are a senior FP&A analyst. Given the following data, produce exactly three sentences: - Sentence 1: State what happened, putting the number in context (budget, prior month, prior year). - Sentence 2: Explain the cause, marking it as favorable or unfavorable and noting if it is a one‑time event or a trend. - Sentence 3: State the implication for the business and any recommended action, using plain language and no acronyms.
Prompt Examples
SaaS startup (revenue): “Last month MRR was $420K, 8% below the $456K budget and 5% below prior year. The shortfall stems from a delayed enterprise renewal that was expected to close in Q2, making it an unfavorable, one‑time event. Expect the renewal to close next month, which will bring MRR back on track and improve cash flow.”
Series A startup with marketing overspend (OpEx): “Marketing spend reached $180K, 22% over the $147K budget and 15% higher than the prior month. The excess came from a last‑minute push to acquire leads for a webinar that did not convert, an unfavorable, one‑time event caused by internal timing. Going forward, we will lock campaign budgets two weeks before launch and measure ROI in real time.”
OpEx Classification Checklist
Before prompting AI, run through this list for each OpEx line:
- Any known customer events (churns, expansions, new logos)?
- Compared to budget, prior month, or prior year?
- External (market, customer churn) or internal (hiring delay, pricing change)?
- Is the variance favorable or unfavorable?
- Is it a one‑time event or a trend?
- Write exactly three sentences, no acronyms, assuming a non‑finance founder.
- Include sales funnel metrics if available.
Pro Tip: Three Queries per Client
Before the board meeting, run three FP&A Genius queries per client: one for revenue, one for COGS, and one for OpEx. This habit guarantees that every material line gets a polished narrative, reduces last‑minute scrambling, and builds a repeatable workflow you can scale across multiple startups.
For a comprehensive guide with detailed workflows, templates, and additional strategies, see my e-book: AI for Solo Fractional CFOs Serving Startups: How to Automate Board Deck Financial Section Drafting and Variance Narrative Generation.
Now we need to count words. Let’s count manually. I’ll copy the text and count. Title line: “Title: From Numbers to Narrative: AI & ai-Powered Board Deck Explanations for Revenue, COGS, and OpEx Variances” Count words in title line? Usually title line may not count? The requirement: article must be between 450-500 words. Likely includes everything after “Title:”? Safer to count everything after the title line? But they said “Start with the title as a plain heading (e.g., “Title: …\n\n”), then HTML content.” So the title line is part of the article. We’ll count it. Let’s count words. I’ll write the entire article as a string and count. I’ll do manually. Title line words: Title: (1) From(2) Numbers(3) to(4) Narrative:(5) AI(6) &(7) ai-Powered(8) Board(9) Deck(10) Explanations(11) for(12) Revenue,(13) COGS,(14) and(15) OpEx(16) Variances(17) Wait need to recount: Let’s split: “Title:” =1 “From”=2 “Numbers”=3 “to”=4 “Narrative:”=5 “AI”=6 “&”=7 “ai-Powered”=8 “Board”=9 “Deck”=10 “Explanations”=11 “for”=12 “Revenue,”=13 “COGS,”=14 “and”=15 “OpEx”=16 “Variances”=17 So 17 words. Now we need to count the rest. I’ll go paragraph by paragraph. Paragraph after title: blank line then heading.Why Variance Narratives Matter
This heading line maybe not counted as words? It’s inside HTML but we should count visible words: “Why Variance Narratives Matter”. That’s 4 words. But the HTML tags and comments likely not counted. Usually word count refers to visible text. We’ll count visible words only. Safer to count only the content withinetc and headings text. We’ll count visible words. Thus we need to count visible words in headings and paragraphs. Let’s list visible text: Heading: “Why Variance Narratives Matter” => words: Why(1) Variance(2) Narratives(3) Matter(4) =4. Paragraph 1: “Board decks live or die by the clarity of their financial explanations. Founders need to see not just that revenue missed budget, but why it happened and what it means for the next quarter. AI can turn raw variance numbers into a concise, three‑sentence story that satisfies both finance rigor and founder readability.” Let’s count. Sentence1: Board(1) decks2 live3 or4 die5 by6 the7 clarity8 of9 their10 financial11 explanations12. =>12 Sentence2: Founders13 need14 to15 see16 not17 just18 that19 revenue20 missed21 budget,22 but23 why24 it25 happened26 and27 what28 it29 means30 for31 the32 next33 quarter34. =>22 more? Let’s count from Founders: Founders(13) need14 to15 see16 not17 just18 that19 revenue20 missed21 budget,22 but23 why24 it25 happened26 and27 what28 it29 means30 for31 the32 next33 quarter34. So adds 22 (13-34). Total so far 34. Sentence3: AI35 can36 turn37 raw38 variance39 numbers40 into41 a42 concise,43 three‑sentence44 story45 that46 satisfies47 both48 finance49 rigor50 and51 founder52 readability53. => adds 19 (35-53). Total =53. So paragraph1 visible words =53. Now heading2: “The Three‑Layer Framework” => The1 Three‑Layer2 Framework3 =>3 words. Paragraph2: “Chapter 4 of the e‑book introduces a simple stack: Layer 1 captures the raw variance (the numbers); Layer 2 identifies the root cause (internal or external driver); Layer 3 crafts the board‑ready narrative. Keeping each layer distinct prevents mixing data with