Spreadsheets capture the “what,” but only event-based modelling gives you the “why” and the “when.”

‍

This is not another “Excel is dead” post.

Yet.

‍

Spreadsheets are the de facto granddaddy of financial modelling - the default language of finance and operations. Excel and Google Sheets have given us a way to run calculations, build formulas, and capture the what of business.

What was revenue last quarter?

What’s payroll this month?

What's our headcount?

Full points. Spreadsheets are great at producing those answers.

But they lack something critical in the age of AI…..

Context.

‍

Take spreadsheet cell AA22 with a value of 662.

If you’re lucky, the cell next to it has a label: Customers, Units.

Maybe even “Pickles per Jar.”

But what that single number doesn’t store is the metadata - the story that makes that number meaningful.

Who counted the pickles?

When were they counted?

Which factory made that jar?

And how did the pickle counter feel that day?

Spreadsheets are an illusion of context. They rarely tell you the why behind a number, and they don’t truly capture the when. Spreadsheets don’t model time, they mimic it. Dates live as numbers in other unrelated cells, but the system has no concept that a hire in April changes payroll in May and revenue in June.

Time in spreadsheets is a trick - something you force into place with formulas.

Having spent 25 years in the film, television and editing post-production industry, where the timeline is the dominant axis, this limitation jumps out. In that world, everything has a start frame and an end frame, and nothing happens unless those frames are active.

Business works the same way: hires begin on a start date, contracts expire on an end date, campaigns run for a fixed period. But spreadsheets don’t recognize this logic. They can store dates in cells, but they don’t understand time as a living sequence. They capture values, but not the narrative. At best, the reasoning shows up as a scribble in the margins - a comment that never becomes part of the calculation itself.

And in the era of AI, this gap matters more than ever. Businesses don’t just need numbers; they need the narrative that explains those numbers. AI can crunch data instantly, but if all it sees is a sea of cells, it can’t explain why things changed - or when those changes matter.

Cause and effect.

‍

The Limits of Cell-Based Modelling

‍

Spreadsheets are brilliant at showing numbers. But that brilliance comes at a cost: they reduce the entire business to numbers alone. Everything else - assumptions, intent, even the story behind those numbers - lives outside the model. That gap creates five big problems.

Misinterpretation

Numbers without story invite conflicting narratives. A revenue line of $120,000 in March might mean a seasonal bump to one person, a new client to another, and a one-off promotion to someone else. Without embedded context, everyone is telling their own story.

Oversimplification

Because complex spreadsheet models are fragile, teams often strip them down to stay manageable. But simplicity comes at a cost: critical nuance gets left out. If the Vancouver manufacturing location is underperforming, a P&L might only show the “what” (poor numbers) without capturing the “why” (a tax change, a supply chain issue, tariffs). Decisions made on stripped-down models are, by design, less informed.

Fragility and Error

Spreadsheets collapse under their own weight. One broken formula, one fat-finger entry, and the model can spill everywhere. There are countless real-world examples of billion-dollar mistakes caused by small spreadsheet errors.

Knowledge Silos

Spreadsheets lock knowledge to their creators. Hand a model to someone else and, without a guided tour, all they see are numbers and formulas. The assumptions and logic remain hidden.

AI Blind Spots

Machines can crunch numbers, but they can’t infer meaning without context. A spreadsheet can show “662 pickles,” but an AI can’t tell if that’s enough to fill the warehouse order unless it knows jar size, recipe, and supplier. Without cause-and-effect structure, AI insights remain shallow.

‍

Event-Based Modelling

‍

Event-based modelling treats your business like a story told on a timeline. Instead of burying assumptions in cells, you build with events: the moment you hire someone, re-order products, launch a campaign, raise prices, or open a new office.

What is an Event?

‍

In practice, an event is a business logic block. It’s a self-contained unit that encapsulates the formulas, calculations, and metadata that describe a real-world business action or asset. Some events take in input data, some transform that data, and some create conditional dependencies that trigger actions or calculate key KPIs.

Think of our pickle manufacturing business. Events might include:

• Income (revenue from jars sold, either ongoing, one-off promotions, or wholesale contracts amortized over time)
• Expenses (cucumbers, vinegar, jars, operating costs)
• Employees (the new production line worker hired in June, or the seasoned sales rep who closes distribution deals)
• Loans (a financing arrangement for new factory equipment, complete with repayment schedules and interest baked in)
• Inventory and Re-Orders (restocking cucumbers or jars when supplies fall below threshold)
• Resources and Capacity (how many production lines are active, how many shifts can be staffed, and whether you can meet a big retail order)
• Project Bids or Deals (the probability-weighted chance of landing a new distribution contract with a grocery chain)

Each event carries its own:

• Start and end dates (when it’s active)
• Cadence (daily pickling runs, weekly shipments, quarterly sales contracts)
• Ledger connections (so it’s tracked to the right account, e.g., Income, Expense, Assets, or Resources)
• Advanced calculations (like depreciation of factory equipment, compounding interest on loans, or utilization rates for staff and production lines)
• Descriptions and metadata (so the context travels with the data, not hidden in a footnote)

Instead of hand-building these constructs in a spreadsheet with fragile formulas, events come with the business logic already baked in. They’re modular, reusable, and designed to snap together into a timeline of cause and effect.

This isn’t just about finance.

In capacity planning, you might line up Employee events against Distribution Deal events linked to Hubspot or Salesforce to see if your factory has enough shifts to produce the extra 100,000 jars if two contracts close at once. In supply chain, you might model Re-Order events that trigger when cucumber inventory falls too low, and immediately see how delayed shipments ripple through production and cash flow.

Each event knows when it happens, what it consumes or generates, and how it ripples across the rest of the business. Whether you’re running a pickle factory or a SaaS company, the logic is the same: events turn disconnected numbers into a living map of business cause-and-effect.

This makes event-based modelling more than financial; it’s operational and strategic too. It captures the chain of decisions and transformations that drive outcomes.

If spreadsheets are ledgers of numbers, event-based models are ledgers of decisions.

‍

1. Clear Cause-and-Effect

Every outcome is tied to a real-world event, not an abstract formula.

• Example: Add a “New Sales Hire” event on June 1. The model automatically accounts for salary, onboarding, ramp-up time, and revenue impact. No footnotes required - the “why” is baked in.

2. Timeline Awareness

Events unfold in sequence, just like real life.

• Example: A Product Launch event in September doesn’t just spike revenue. It pushes marketing spend into July and August. The ripple is visible across the timeline.

3. Context as Data

Assumptions aren’t scribbles - they’re structured.

• Example: A Supplier Contract Change shifts terms from Net 30 to Net 60. That change instantly ripples through cash flow, working capital, and even financing needs.

4. Resilience and Transparency

Events are modular building blocks, not tangled formulas. Updating one doesn’t break the rest of the model.

5. Built for AI

Because events carry type, timing, dependencies, and ledger assignment, AI isn’t just parsing numbers - it’s reading a storyline. That turns it from a pattern-spotter into a strategist.

‍

Why Context Matters for AI & Business Forecasting

‍

Click your heels together and repeat this three times...

Large Language

In a spreadsheet, “50,000” in cell D42 is meaningless on its own. Is it payroll? Marketing? A one-off supplier invoice? Even with labels, the relationships between numbers are buried in fragile formulas. For AI, that’s like being handed puzzle pieces without the picture on the box.

Yes, AI can extract patterns from raw spreadsheets or time-series data. But without explicit business events, it’s inferring context indirectly. Event-based modelling makes that context explicit - a far stronger foundation for reasoning and simulation.

Event-based modelling encodes causal structure directly into the data. Each event is an object with attributes - type, timing, dependencies, ledger assignment. When AI queries this data, it’s traversing a graph of decisions and outcomes. That makes it possible to answer richer questions:

• Spreadsheet AI: “Why did cash drop in July?”
• → Answer: “Expenses increased.” (True, but shallow.)
• Event-based AI: “Why did cash drop in July?”
• → Answer: “The Supplier Invoice event for $50,000 triggered in July, linked to COGS. It coincided with a Product Launch event that pulled marketing spend forward, resulting in a temporary cash dip.”

Context isn’t fluff - it’s infrastructure. It transforms AI from descriptive (“what happened”) to diagnostic and prescriptive (“why it happened” and “what to do about it”).

‍

Scenario Analysis and Simulation in AI Finance

‍

At its core, business is about asking what if?

What if we launch a product later?

Raise prices?

Delay a hire?

That's a great question.... Let me get back to you....

These are the questions leaders wrestle with daily - and spreadsheets make them painful to answer.

In spreadsheets, scenario analysis usually means copying entire tabs and hoping the formulas hold. It’s clunky, fragile, and hard to maintain. That’s why most teams only ever run a handful of scenarios, even though the real world throws dozens their way.

Event-based modelling flips this on its head. Because the model is a chain of cause-and-effect events, branching is natural. At any decision point, you can test alternate paths:

• Hire the salesperson in April vs. July.
• Launch the product in September vs. November.
• Negotiate supplier terms at Net 30 vs. Net 60.

These branches don’t destroy the base model - they sit side by side, letting you compare outcomes without breaking anything.

This is also where AI supercharges the workflow. With structured events, AI can evaluate every decision point: “Scenario A dips cash in Q3 but recovers faster; Scenario B preserves cash short-term but slows growth.” Instead of guessing which what-if to test, you can explore them all.

Simulating.

Not just forecasting.

In short, event-based modelling doesn’t just make scenario analysis easier - it makes it strategic.

‍

Why Event-Based Modelling is the Future of Financial Modelling Software

‍

Spreadsheets were built for an era of static reports and backward-looking analysis. They’re powerful, but they were never designed to handle the complexity, velocity, and interconnectedness of modern business. Event-based modelling is built for the future.

Scalability Without Collapse

‍

As businesses grow, spreadsheets buckle. More customers, more products, more geographies - it all means more tabs, more formulas, more risk of error. Event-based models scale cleanly. Adding a new customer type, a new product line, or even a whole new region isn’t a restructuring exercise; it’s just another event in the chain. The model expands naturally, like a map, without breaking the core logic.

Integrations and Automations

The future isn’t static data entry - it’s live connections. Event-based systems are API-native, designed to plug directly into the tools you already use. A “New Deal Won” event can sync straight from HubSpot. A “Supplier Invoice Received” can flow directly from QuickBooks. Instead of manually wrangling CSVs, you’re working with real-time data. Better yet, events can trigger automations - if this happens, then simulate that outcome. Finance stops being a monthly retrospective and becomes a daily, living system.

Clarity for Stakeholders

One of the quiet revolutions of event-based modelling is how it changes communication. Try explaining a complex Excel model to a non-finance stakeholder—the conversation usually collapses into “trust me, the formula works.” With event-based models, the story is visual and intuitive. “We hire three reps in April, revenue grows in June. We delay the launch, marketing costs shift.” Decision-making becomes collaborative because the model mirrors the way people actually think.

Auditability and Drill-Down

Another future-proof feature: every event is explicit, timestamped, and auditable. You know who made the change, when, and why. Need to drill into cash flow in July? You can trace it back directly to the events that drove it - whether it was a supplier contract, a product launch, or a hiring surge. Spreadsheets bury that logic in formulas; event-based models make it transparent.

Ready for AI and Agents

Most importantly, event-based modelling is the foundation for AI. Structured events give AI the hooks it needs to test scenarios, simulate outcomes, and even act as an assistant that can propose alternatives on the fly. Instead of giving AI a maze of cells, you give it a structured map of decisions and dependencies. That makes AI not just descriptive, but prescriptive.

Overcoming the Comfort Zone of Spreadsheets

‍

I get it.

I’ve built some beautiful spreadsheets.

Finance pros are protective of spreadsheets. And rightly so - they’ve banked their careers mastering them. They’re trusted, auditable, and familiar. When the board asks a question, spreadsheets deliver in the language everyone knows.

But that comfort comes with a cost. The business landscape is shifting under our feet. AI, real-time data, and the demand for rapid scenario analysis aren’t slowing down. Spreadsheets excel at backward-looking reports, but they stumble at forward-looking strategy.

That doesn’t mean abandoning what you know - it means evolving. Just as businesses once moved from handwritten ledgers to digital spreadsheets, the next leap is from spreadsheets to event-based frameworks.

The truth is, sticking with spreadsheets feels safe - but it’s also risky. Your competitors won’t stay loyal to a fragile tool if they can simulate ten scenarios in real time while you’re still updating formulas.

Spreadsheets got us here. Event-based models will take us forward.

‍

Setting Up for the AI Era

‍

Spreadsheets are excellent at capturing the what, but they leave out the why and the when. That gap leads to misinterpretation for humans, and shallow insights for AI.

Event-based modelling changes the equation. By treating every decision as an event on a timeline, it turns financial models into living cause-and-effect maps. It makes scenario analysis flexible, collaboration safer, and AI insights actionable.

Most importantly, it prepares businesses for the future. Instead of handing machines a pile of disconnected cells, you hand them structured narratives: ledgers of decisions, not just ledgers of numbers.

Numbers without context are just jars without labels. Event-based modelling puts the labels on the jars, lines them up on the shelf, and lets AI help decide which recipes to scale.

At whatifi, we believe the next generation of financial and strategic modelling won’t be cell-based at all. It will be event-based: timelines, scenarios, and decisions linked together in living cause-and-effect maps. We built whatifi to turn that philosophy into reality, so leaders can run dozens of what-ifs in real time and give AI the context it needs to be more than a pattern-spotter.

The future of decision-making is about context, not just calculations. Event-based modelling is the bridge. And the time to cross it is now.