startup-financial-modeling

Overview of startup-financial-modeling skill

What startup-financial-modeling does

The startup-financial-modeling skill helps an agent build a 3-5 year startup finance model for planning, fundraising, and operating decisions. It is aimed at early-stage companies that need more structure than a generic “make me a forecast” prompt, especially when revenue depends on customer acquisition, retention, pricing, hiring, and cash burn over time.

Who should use this skill

This startup-financial-modeling skill is best for founders, finance leads, operators, startup advisors, and product-minded analysts who need investor-ready logic without starting from a blank sheet. It is especially useful for seed and Series A style questions such as:

• How long is runway under current hiring plans?
• What revenue path is implied by acquisition and retention assumptions?
• When do we need to raise again?
• How do best/base/worst cases change burn and cash-out date?

Real job-to-be-done

Most users do not just want “a model.” They want a forecast they can defend. The value of startup-financial-modeling is that it pushes the model toward explicit drivers: cohort growth, ARPU, retention, cost buckets, burn, runway, and scenario analysis. That is more decision-useful than a flat top-line CAGR forecast.

What makes it different from a generic finance prompt

The main differentiator is structure. The skill centers startup-specific modeling patterns such as:

• cohort-based revenue logic
• detailed operating expense categories
• cash flow and runway analysis
• fundraising scenario planning

That makes it a better fit for SaaS and recurring-revenue startups than a one-shot prompt that jumps straight to totals without showing assumptions.

Important limits before you install

This skill is a document-driven guide, not a packaged spreadsheet, code library, or rules engine. There are no bundled scripts, templates, or support files in the skill folder. You should install startup-financial-modeling if you want a stronger prompting framework for an AI agent, not if you need audited financial modeling standards, accounting compliance, or a ready-made Excel model.

How to Use startup-financial-modeling skill

startup-financial-modeling install context

The upstream skill does not include its own install command in SKILL.md, but directory users typically add skills from the parent repository. A common pattern is:

npx skills add https://github.com/wshobson/agents --skill startup-financial-modeling

After install, the main source to read is:

• plugins/startup-business-analyst/skills/startup-financial-modeling/SKILL.md

Because there are no extra references or helper files here, nearly all usable guidance is in that single file.

Read this file first

Start with SKILL.md and focus on these sections in order:

1. Overview
2. Core Components
3. Revenue Model
4. Cost Structure
5. cash flow, runway, and scenario sections lower in the document

That path gives you the actual modeling logic faster than skimming the whole repository.

What inputs the skill needs

startup-financial-modeling usage is much better when you provide explicit drivers instead of asking for “a projection.” At minimum, give the agent:

• business model: SaaS, marketplace, fintech, usage-based, services-hybrid
• pricing: plans, ARPU, contract length, expansion assumptions
• acquisition: monthly new customers or leads-to-close funnel
• retention: logo churn, revenue churn, or cohort retention curve
• COGS: hosting, support, payment fees, third-party tools
• operating costs: headcount plan, salaries, marketing spend, G&A
• starting cash and existing burn
• fundraising assumptions: round size, timing, dilution target, runway goal
• planning horizon: usually 36, 48, or 60 months

If you leave these blank, the model becomes generic quickly.

Strong prompt shape for startup-financial-modeling

A good prompt for the startup-financial-modeling guide should ask for both assumptions and outputs. Use a structure like:

• company stage and business model
• current metrics
• target horizon
• required scenarios
• output format
• specific finance questions to answer

Example:

“Use the startup-financial-modeling skill to build a 36-month model for a B2B SaaS startup. We have 120 customers, $28k MRR, 2.5% monthly logo churn, $240 ARPU, and add 18 new customers per month today. Assume CAC starts at $900 and improves 10% over 12 months. Team is 8 people today and grows to 14 over 18 months. Starting cash is $1.4M. Show base, upside, and downside cases with monthly revenue, COGS, opex, burn, runway, and suggested raise timing.”

That is far more actionable than “make a startup forecast.”

Translate a rough goal into model-ready inputs

If you only know the goal, turn it into driver questions first. For example:

Rough goal:

• “I need investor financials for a seed raise.”

Better prompt:

• “Use startup-financial-modeling for Finance planning. Build a 48-month monthly model for a seed-stage B2B SaaS company. Ask me for any missing assumptions before modeling. Include customer growth by cohort, retention, pricing, COGS, hiring plan, burn, runway, and a financing case with a $3M raise in month 6.”

This works because it gives the agent permission to collect missing inputs before guessing.

Best workflow in practice

A practical workflow for startup-financial-modeling usage is:

1. Define the business model and time horizon.
2. Give current baseline metrics.
3. Ask the agent to list missing assumptions.
4. Confirm core drivers before projection.
5. Generate monthly model outputs.
6. Add best/base/worst scenarios.
7. Stress test cash-out date and raise timing.
8. Convert the model into an investor or board narrative.

The key step is the assumptions check before projection. That is where most bad outputs can be prevented.

What the skill appears best at

Based on the source, startup-financial-modeling is strongest when the company has recurring revenue and customer cohorts matter. It is well suited to:

• B2B SaaS
• subscription products
• retention-driven businesses
• early-stage fundraising planning
• runway and burn analysis

It is less naturally suited to one-off project businesses unless you adapt the revenue logic.

Output formats to request

Do not leave output format vague. Ask the agent for one or more of these:

• monthly table for 24-60 months
• assumption summary table
• scenario comparison table
• fundraising timeline
• break-even month estimate
• board-ready explanation of key sensitivities

If you want to move the work into Sheets or Excel, ask for plain tables with formulas described in words.

Common adoption blockers

Before installing startup-financial-modeling, most blockers are not technical. They are input quality problems:

• no retention assumptions
• no clear pricing logic
• no headcount plan
• no split between COGS and opex
• no starting cash or debt context
• asking for annual outputs when runway needs monthly detail

The skill helps with structure, but it cannot invent reliable operating assumptions for you.

How to get better scenario planning

The repository clearly emphasizes scenario analysis, so use that deliberately. A useful scenario setup is to vary only a few drivers per case:

• acquisition volume
• retention/churn
• ARPU or expansion revenue
• hiring pace
• fundraising timing

If every line changes in every scenario, the result becomes hard to explain to investors or operators.

startup-financial-modeling skill FAQ

Is startup-financial-modeling worth installing?

Yes, if you want an agent to use startup-specific finance logic instead of generic forecasting language. The startup-financial-modeling skill gives a clearer modeling frame than ordinary prompts, even though it does not ship with spreadsheet files or automation.

Is startup-financial-modeling good for beginners?

Yes, with one caveat: beginners still need to supply basic business assumptions. The skill can organize the model, but it does not remove the need to understand terms like ARPU, churn, COGS, burn, and runway.

How is it different from asking ChatGPT for a forecast?

Ordinary prompts often skip driver logic and jump to summary numbers. startup-financial-modeling is more useful when you need the path from assumptions to outputs, especially for cohort revenue, cost categories, and cash planning.

Can I use startup-financial-modeling for non-SaaS companies?

Sometimes. It fits best when revenue can be modeled through repeated customer behavior and recurring economics. For transactional or services-heavy businesses, you may need to rewrite the revenue section around bookings, utilization, project margins, or take rate.

Does the skill generate a spreadsheet?

Not by itself. The repository evidence shows only SKILL.md in this skill folder, with no templates or scripts. Expect guidance for an agent, not a downloadable financial model.

When should I not use startup-financial-modeling?

Skip it if you need:

• audited financial statements
• tax or GAAP advice
• lender-grade financial packages
• cap table legal modeling
• a plug-and-play Excel workbook

It is also a weak fit if your business has no meaningful cohort, retention, or recurring revenue pattern.

How to Improve startup-financial-modeling skill

Give driver-level inputs, not outcome targets

The fastest way to improve startup-financial-modeling output quality is to stop prompting with targets like “get to $10M ARR.” Instead provide the drivers that could produce that result:

• customer adds by month
• retention by cohort
• pricing by segment
• upsell timing
• channel-level CAC
• hiring ramp
• infra cost per customer

This makes the model explainable, not aspirational.

Ask the agent to separate assumptions from calculations

A common failure mode is hidden assumptions mixed into the forecast. Improve results by explicitly asking for:

1. assumption table
2. formula logic
3. monthly outputs
4. scenario deltas
5. key sensitivities

That makes bad assumptions easier to spot before you rely on the numbers.

Force monthly detail for runway questions

If your goal is cash planning, monthly detail matters. Annual views hide cash-out risk. For better startup-financial-modeling for Finance results, ask for monthly projections at least until the next expected raise or break-even point.

Tighten retention and expansion assumptions

Weak retention inputs are the biggest quality risk in recurring-revenue models. Instead of saying “churn is low,” specify something like:

• 3% monthly logo churn in months 1-12
• net revenue retention of 105% for enterprise accounts
• expansion starts after month 4 for 20% of retained customers

Even rough numbers like these are better than vague optimism.

Improve the cost model with hiring timing

Many startup models understate burn because headcount is too generic. Provide:

• role
• start month
• fully loaded salary
• commission if relevant
• one-time recruiting or equipment costs

This materially improves burn and runway outputs.

Use scenario discipline

Do not ask for ten scenarios. Ask for three clear cases and define what changes in each one. Example:

• Base: current conversion and churn hold
• Upside: churn improves 20%, CAC improves 15%
• Downside: sales hiring slips 3 months, churn worsens 25%

This keeps the scenario logic decision-useful.

Request a sanity-check section

A good way to improve the startup-financial-modeling guide output is to ask the agent to flag unrealistic assumptions, such as:

• ARR growth far above hiring capacity
• gross margin that conflicts with infrastructure costs
• CAC payback worse than cash runway
• raise timing that occurs after cash-out

That catches model issues a polished table can hide.

Iterate after the first draft

The first draft should not be your final model. Improve it by asking:

• Which 3 assumptions drive most variance in runway?
• What would make this model investor-unconvincing?
• Which metrics need real data instead of estimates?
• What changes if hiring is delayed by one quarter?

This turns the skill from a one-time output into a planning tool.

Adapt the revenue logic if your business is atypical

If you install startup-financial-modeling for a marketplace, fintech, or hybrid services company, tell the agent to replace default SaaS assumptions where needed. Examples:

• marketplace: GMV, take rate, buyer/seller cohorts
• fintech: transaction volume, interchange, loss assumptions
• services hybrid: billable headcount, utilization, project margin

Without this adaptation, the output may look polished but fit the business poorly.