How to Analyze a Business: A Practical SaaS Framework

Business analysis is the process of explaining how a company creates value, captures value, and converts that value into cash. It is broader than reading a profit-and-loss statement and more disciplined than writing a product description.

A useful analysis answers five questions:

1. What problem does the company solve, and for whom?
2. How does it acquire and retain customers?
3. How does pricing turn usage into revenue?
4. What costs are required to deliver and grow the product?
5. Which facts are reported, and which numbers are only estimates?

Linear is a useful case study because it publishes meaningful operating information, but not the complete financial detail that an investor would receive from a public company. The method below keeps those two kinds of information separate.

Start With the Company’s Status

The first question is not “What is the revenue?” It is “What disclosure regime applies to this company?”

A domestic public company in the United States files annual Form 10-K reports. The SEC describes Form 10-K as a comprehensive overview of the business and financial condition that includes audited financial statements. A private company does not normally provide the public with the same standardized package.

That distinction changes the research method:

The following checklist is useful when deciding whether a piece of business information normally belongs in a public company’s filing:

Linear should therefore not be analyzed as if it had published a public-company annual report. Its official pages disclose selected metrics, while detailed revenue composition, COGS, operating expenses, and product-level profitability remain undisclosed in the sources reviewed here.

Build an Evidence Ledger

Before calculating anything, classify every important statement. This prevents an estimate from quietly becoming a “fact” during editing.

Use language that matches the evidence. “Linear reported” is appropriate for the company’s growth announcement. “Sacra estimates” is appropriate for third-party research. “This model assumes” is appropriate for a margin calculation.

Analyze the Product and Industry

Describe the product in terms of the job it performs, not only its feature list. For Linear, the core workflow is product development: teams organize issues, projects, initiatives, cycles, code changes, requests, and agent activity in one system.

Then define the market narrowly enough to identify real alternatives. Linear sits at the intersection of:

• engineering issue tracking;
• product and portfolio planning;
• work intake and customer feedback;
• AI-assisted execution and agent workflows.

Competition should be analyzed by the buyer’s alternative, not by a random list of similar applications. Jira represents an established work-management incumbent with free, Standard, Premium, and Enterprise tiers, automation, AI features, and enterprise controls. Plane represents a different alternative: it combines project management, documentation, AI agents, and cloud, self-hosted, and air-gapped deployment options.

The useful question is not “Which product has more features?” It is “Why would a customer switch, and what would make switching painful?” Relevant dimensions include workflow quality, integrations, data control, security, implementation effort, ecosystem, and the amount of organizational context stored in the product.

Explain the Business Model

Linear’s visible model is product-led subscription SaaS:

• the Free plan lowers the cost of trial and supports bottom-up adoption;
• Basic is listed at $10 per user per month;
• Business is listed at $16 per user per month and adds advanced workflow, analytics, integration, and AI-related features;
• Enterprise uses custom annual pricing and adds capabilities such as SAML, SCIM, granular administration, security controls, onboarding, priority support, and account management.

This structure suggests several growth loops:

1. A team adopts the product for one workflow.
2. More seats and teams are added as the workflow becomes valuable.
3. The customer upgrades for privacy, administration, security, support, or AI features.
4. More organizational context makes the product harder to replace.

These are business-model mechanisms, not proof of a particular financial result. The pricing page does not disclose plan mix, discounts, contract values, or the percentage of revenue attributable to AI.

Separate ARR From Revenue

ARR is an annualized measure of recurring subscription run rate. It is not necessarily the same as GAAP revenue for the period. It may exclude one-time services and it does not, by itself, describe cash collections, deferred revenue, COGS, or net income.

Linear reported that it had passed $100M ARR, had more than 40,000 paying companies, and had reached 177% net revenue retention. A simple division gives:

$100M ARR / 40,000 paying companies = $2,500 ARR per company on average

This is an arithmetic average, not actual ARPA. A small number of enterprise accounts may contribute much more than small teams, and the customer-count definition and cohort details are not fully published.

NRR is also a cohort metric, not a profit metric. A high NRR can indicate expansion, upgrades, and seat growth among existing customers, but it does not reveal acquisition cost, gross margin, or cash conversion.

Model Revenue Composition Carefully

When a private company does not publish segment revenue, begin with mechanisms rather than invented percentages.

Do not infer revenue mix directly from price. A $10 plan and a $16 plan do not tell us how many customers use each plan, what discounts were granted, or how Enterprise contracts are priced.

Map the Cost Structure

For a SaaS business, distinguish delivery costs from growth costs.

COGS and gross profit

COGS may include cloud infrastructure, databases, storage, bandwidth, backups, observability, third-party services, delivery-related support, and AI inference. Gross profit is:

Gross profit = Revenue - COGS
Gross margin = Gross profit / Revenue

AI makes the model more usage-sensitive. Inference cost depends on model choice, request volume, context size, caching, and whether usage is included or metered. The product page can establish that AI features exist; it cannot establish Linear’s AI COGS.

Operating expenses

Operating expenses generally include:

• research and development, product, design, and security;
• sales, marketing, customer success, and account management;
• legal, finance, recruiting, compliance, and administration;
• offices, travel, insurance, and other corporate expenses.

Linear’s growth announcement says it had more than 30 open roles across engineering, product, design, sales, support, and other functions. That indicates continued hiring, but it does not disclose total headcount, compensation, stock-based compensation, or operating expenses.

Use an Explicit Profit Scenario

If the income statement is unavailable, a scenario can still make the economics concrete. It must be labeled as a model, and every assumption must be visible.

The following example treats the reported $100M ARR as a revenue proxy. It is not a reported Linear income statement.

The implied margins are 82% gross margin and 24% operating margin. They are outputs of the assumptions, not facts about Linear. A sensitivity table makes the uncertainty more honest:

Do not call $18M “Linear’s COGS” or $24M “Linear’s net profit.” The model does not include taxes, interest, stock-based compensation, depreciation, one-time items, or cash-flow timing. It estimates operating profit, not net income.

Treat Financing as Context, Not Revenue

Sacra compiles the following primary financing history for Linear:

Linear separately announced a $99M tender offer at a $2.5B valuation in August 2026. The company described the tender as liquidity for current and former employees. A tender offer is a secondary transaction when existing shareholders sell shares; it is not automatically a new primary capital raise and should not be added to revenue.

Financing helps explain how a company reached its current product and team scale. It does not prove product-market fit, profitability, or the value of the latest valuation.

What the Analysis Can Conclude

A disciplined conclusion is narrower than a confident spreadsheet. For Linear, the available evidence supports these statements:

• the company sells a recurring subscription product with Free, per-seat, Business, and custom Enterprise packaging;
• the company reported more than $100M ARR, more than 40,000 paying companies, 177% NRR, and positive cash flow in its August 2026 growth update;
• the product competes with broad work-management platforms, developer tools, and AI-oriented alternatives;
• Sacra reports approximately $134.2M of primary funding;
• detailed revenue mix, COGS, operating expenses, team size, net income, and product-level profit are not established by the reviewed public sources.

The right final sentence is therefore not “Linear earns $24M in operating profit.” It is: “Under an explicitly stated 18% COGS and 58% operating-expense scenario, a $100M ARR revenue proxy would produce $24M of illustrative operating profit. Linear’s actual income statement is not publicly disclosed in the sources reviewed.”

That distinction is the core skill in business analysis: calculate when calculation is useful, but preserve the boundary between evidence and inference.

References

• Investor.gov: Form 10-K — explains the public-company annual report, audited financial statements, and ongoing filing obligations.
• Linear Pricing — published plan prices, packaging, Enterprise controls, AI features, and customer-count statement.
• Linear About — Linear’s founding year, product positioning, customer scale, and distributed-team description.
• Linear: Sharing Linear’s growth — company-reported ARR, paying companies, NRR, cash-flow position, tender offer, valuation, agent adoption, and open roles.
• Sacra: Linear — third-party ARR, revenue, and financing compilation.
• Atlassian Jira Pricing — Jira plans, automation, AI features, enterprise controls, and pricing context.
• Plane — Plane’s project-management, AI, migration, self-hosting, and air-gapped product positioning.