Ten lessons on evaluating opportunities, talking to customers, building a model, checking unit economics, choosing financing and a legal entity, reaching customers and pitching.
A college-level introduction for US-based founders. Jurisdiction and standards scope: US federal rules only (SBA programs, SEC securities exemptions, IRS and Internal Revenue Code). State law, non-US law, and tax years other than those named are out of scope. Nothing here is legal, tax or investment advice.
High-school algebra and percentages. No accounting background needed.
Course outline
From idea to opportunity: screening before building
Separate an idea from an opportunity, size a market bottom-up, and treat survival statistics as a base rate rather than a forecast.
Customer discovery: evidence over opinions
Design interview questions that produce evidence about past behavior and spending, and rank signals by the cost the customer paid to give them.
Business models: how value turns into revenue
Describe a business model as a chain from customer to value to revenue to cost, and match the revenue mechanism to the way customers receive value.
Unit economics: contribution, payback and lifetime value
Compute contribution margin, payback and lifetime value for one customer, and judge whether growth adds or destroys value.
Cost structure, break-even and runway
Separate fixed from variable cost, find break-even volume, and compute how many months of cash a plan really has.
Financing options: debt, equity and crowdfunding
Match a funding need to the right source, read SBA program terms accurately, and know what a loan guarantee does and does not do.
Equity financing, dilution and securities exemptions
Compute ownership after a priced round including an option pool, and choose between the main US federal exemptions for selling securities.
Legal entity choice and tax treatment
Compare sole proprietorship, LLC, S corporation and C corporation on liability, tax and ability to raise equity under US federal rules.
Go-to-market: channels, funnels and sales motion
Pick an ideal customer and channel, model the funnel, and compare channels by customer acquisition cost and payback.
Pitching: ask, evidence and milestones
Build a short pitch around a defensible ask, size the raise from burn and milestones, and avoid securities-law traps in public pitching.
Sources and curriculum note
Sources read October 4, 2026. Rules, program caps and tax figures change, so check the current source before a real decision. Toy numbers are labeled as such and are not observed data.
Read every lesson below. The interactive reader above contains the same explanations, with visual tools and quizzes.
1. From idea to opportunity: screening before building
Learning goal: Separate an idea from an opportunity, size a market bottom-up, and treat survival statistics as a base rate rather than a forecast.
An idea is a product someone can imagine. An opportunity is a problem that a specific group of customers has, feels strongly enough to pay to solve, and can be reached at a cost below what they will pay. Screening an idea means writing those parts down as claims you can test: who the customer is, what the pain is, what they do today, and how you would reach them. If any part is a guess, the idea is still a hypothesis.
Market size is easy to inflate and hard to defend. A top-down estimate takes a large industry figure and claims a small percentage of it, and the percentage is invented. A bottom-up estimate starts with a countable group, such as dog-owning households in one metro area, and multiplies by the share that has the pain, the share you can reach, the visits per year and the price. Each factor can be checked with a survey, a public count or a pilot. In this module, TAM means everyone who could use the product, SAM means the part your model and channel can serve, and SOM means what you can plausibly win in the next planning period.
Base rates keep optimism honest. The US Bureau of Labor Statistics Business Employment Dynamics table 7 tracks private-sector establishments by opening year. For establishments that opened in the year ended March 1994, 79.6 percent were still present one year later and 49.6 percent after five years. These are establishments, including new locations of existing firms, across all private industries, so the table does not give the failure probability of your venture. It does show why founders plan for a long stretch of uncertainty. Kerr, Nanda and Rhodes-Kropf argue in the Journal of Economic Perspectives that entrepreneurship is experimentation: success probabilities are low, highly skewed and unknowable until money is spent, which favors cheap, staged tests.
The SBA guidance on market research makes a practical point: gather demographic and competitive information early so the idea is confirmed or reshaped while changes are cheap. Scope note for the whole module: US federal rules only. Securities, tax and lending facts are cited to US federal sources. State law, non-US rules and tax years other than those named are out of scope, and nothing here is legal or tax advice.
Putting it to work: before you build anything, write a one-page screen. Name one customer, one costly pain, one current workaround, one channel and one number you could check this week. Then set a stopping rule, such as dropping the idea if fewer than three of ten prospects describe the pain unprompted. A written rule protects you from talking yourself into the idea after a few warm conversations.
Worked example
Estimate the bottom-up annual revenue for the mobile dog grooming case: 400,000 households, 25 percent own a dog, 12 percent of those would book, 6 visits a year at $70 per visit.
Dog-owning households: 400,000 x 0.25 = 100,000.
Households that would book: 100,000 x 0.12 = 12,000.
Visits per year: 12,000 x 6 = 72,000.
Revenue: 72,000 x $70 = $5,040,000 a year. This is a ceiling for the served group, not a forecast of what one new firm wins.
Practice problem and solution
A founder plans software for independent fitness studios in one metro. There are 2,400 gyms. 35 percent are independent studios. Of those, 25 percent still run on spreadsheets. The founder can reach 40 percent of the spreadsheet users in the first year. Price is $180 a month, billed for 12 months. What is the first-year revenue of the reachable group, in dollars? In one sentence, name the funnel assumption that most needs testing.
2,400 x 0.35 = 840; x 0.25 = 210; x 0.40 = 84 studios; 84 x $180 x 12 = $181,440.
Mental model: Opportunity = named customer + costly pain + payment evidence + reachable channel. Size it bottom-up from countable groups.
Common trap: Using a share of a giant industry as proof of demand.
2. Customer discovery: evidence over opinions
Learning goal: Design interview questions that produce evidence about past behavior and spending, and rank signals by the cost the customer paid to give them.
Customer discovery is the work of testing the claims in your opportunity screen with real people before building. Steve Blank, writing in the Harvard Business Review in 2013, describes the lean start-up as a methodology that is beginning to replace the traditional approach of writing a business plan, pitching it and selling as hard as you can. The SBA also describes lean business plans as a lighter alternative to a traditional plan that can run to dozens of pages. The shared idea is to treat the plan as a set of guesses to be checked.
Good interview questions ask about the past and the present, not the future. "Would you use this?" invites politeness. "Tell me about the last time this happened. What did you do next? What did that cost you?" produces facts you can compare across interviews. Avoid describing your solution early; a customer who hears the pitch will respond to the pitch instead of to their own problem. Listen for workarounds, because a workaround is evidence of a pain strong enough to cause effort.
Rank what you hear by what it cost the person. A compliment costs nothing. Agreeing to a follow-up call costs a little time. Sharing real data or introducing a colleague costs some reputation. A deposit or a paid pilot costs money, the strongest signal available before a product exists. Keep a tally of how many interviewees reach each rung instead of remembering the best conversation. Small samples are noisy, so state a range, set a threshold in advance and decide beforehand what result would make you change course.
Interviews also fail through selection. If you only talk to friendly contacts, or only to people who answered a cold email, the pool favors the curious. Record how each person was found and which rung of the ladder each one reached.
Putting it to work: keep a shared interview log with the date, how the person was found, their role, what they do today, what it costs them, and the highest rung they reached. After every five interviews, review the log for patterns and change one thing in the script. When two interviews contradict each other, look for a difference in segment rather than averaging the answers.
Worked example
Run the funnel for a discovery round: 40 interviews. 30 percent describe the problem as a top-three pain. Half of those already spend money on a workaround. Two-thirds of those agree to a paid pilot. What share of all interviewees agreed to a pilot?
Top-three pain: 40 x 0.30 = 12 people.
Spend on a workaround: 12 x 0.50 = 6 people.
Agree to a paid pilot: 6 x (2/3) = 4 people.
Share of all interviewees: 4 / 40 = 10 percent. The sample is small, so report the count and a range, not just the percentage.
Practice problem and solution
In a discovery round, 80 prospects are interviewed. 25 percent have the problem in their top three. 40 percent of those pay for a workaround today. 75 percent of those accept a paid pilot. The firm can reach 1,600 similar prospects and assumes the same rate as the sample. How many pilots would that imply? (A hypothesis, not a forecast.) In one sentence, say why the result is a planning target and not a forecast.
80 x 0.25 = 20; x 0.40 = 8; x 0.75 = 6 pilots. The pilot rate is 6/80 = 7.5 percent. 1,600 x 0.075 = 120. Only 6 people drove that rate, so treat 120 as a rough planning target with a wide range.
Mental model: Ask about past behavior and spending. Rank signals by what they cost the customer and keep counts.
Common trap: Counting enthusiasm as demand.
3. Business models: how value turns into revenue
Learning goal: Describe a business model as a chain from customer to value to revenue to cost, and match the revenue mechanism to the way customers receive value.
A business model is a short answer to four questions. Who is the customer? What do they get that they value? How does money reach the business, and when? What does it cost to deliver, and which of those costs rise with each customer? Two firms can serve the same customers and still have different models if one charges per use and the other charges per month, or if one sells to consumers and the other sells to the companies that employ them.
Revenue mechanisms fit different kinds of value. A subscription fits value that arrives continuously, such as software that is used every week. A one-time sale fits a durable good. A marketplace charges a take rate on the value of transactions it enables, so its revenue depends on getting both buyers and sellers to show up. A razor-and-blade model sells a low-margin base product and earns on repeat purchases of consumables. The mismatch to avoid is charging for something on a schedule that does not match when the customer feels the benefit.
Check the model by following one dollar. In a marketplace, a customer pays the platform, the platform pays the provider and keeps a share, and a payment processor takes a fee from the total. Net revenue is the take rate minus those pass-through costs, not the headline take rate. Founders who quote revenue as the full transaction value overstate their business. Some marketplaces record only the fee as revenue; this module uses the simpler convention of net revenue after processing, stated in each example.
Models are hypotheses. Pricing, channel and customer segment are the parts founders most often change after their first pilots. Write the model as one sentence for each part, then list the single assumption that would hurt most if wrong.
Putting it to work: draft three versions of your model on one page each, for example subscription, per-use and a hybrid, and write the single riskiest assumption under each. Take the assumption with the biggest consequence to your next five customer conversations. Revenue mechanism and price are two of the cheapest things to test before you build, because customers can react to a quote.
Worked example
A services marketplace processes $2,500,000 of bookings a month. It keeps a 12 percent take rate. Payment processing costs 3 percent of all bookings and the platform bears that cost. What is monthly net revenue after processing?
Gross take: $2,500,000 x 0.12 = $300,000.
Processing cost: $2,500,000 x 0.03 = $75,000.
Net revenue: $300,000 - $75,000 = $225,000. Headline revenue overstates the business by a third.
Practice problem and solution
A two-sided marketplace for home repair has many homeowners who browse but few skilled repair providers. Providers are scarce and can easily work off-platform. Which side should the founder first focus on winning, and why? Answer with the side, then explain in one or two sentences.
Win the scarce side first (supply). Homeowners will not return if they cannot book a provider, and providers are the side that can leave. A strong answer says supply is the constraint and names a reason providers stay, such as steady demand or protected payments, instead of subsidizing homeowners who already come.
Mental model: A model is customer, value, revenue mechanism and cost. Follow one dollar to net revenue.
Common trap: Quoting transaction volume as revenue.
4. Unit economics: contribution, payback and lifetime value
Learning goal: Compute contribution margin, payback and lifetime value for one customer, and judge whether growth adds or destroys value.
Unit economics asks what happens financially for one unit, usually one customer. Contribution margin is price minus the costs that rise with each unit, such as materials, payment fees and support. Fixed costs like rent and salaries are not included because they do not change with one extra customer. The contribution margin is what each customer contributes toward fixed costs and profit.
Customer acquisition cost, CAC, is the total sales and marketing spend in a period divided by the new customers it produced. Payback period is CAC divided by monthly contribution margin: the number of months until one customer has repaid what it cost to win them. Payback tells you how much cash the company must carry while it grows. Fast payback means growth funds itself; slow payback means each new customer needs outside money.
Lifetime value, LTV, estimates the total contribution a customer will bring. In this module, with constant monthly churn c and monthly contribution m, expected lifetime is 1/c months and LTV is m divided by c. A 5 percent monthly churn gives a 20-month life. This is a toy model. Real churn is not constant, older customers often differ from new ones, and a young company has little data. Treat LTV as a planning range. A common screen is LTV greater than a multiple of CAC, but the multiple is a convention, not a law.
Do not mix averages across segments. A channel that brings low-churn customers and one that brings high-churn customers produce different LTV even at the same price. Compute unit economics per channel and per plan, and use contribution margin, not revenue, so growth is not mistaken for profit.
Putting it to work: build a small spreadsheet with one row per channel and plan and compute contribution, CAC, payback and LTV from the same inputs. Add a downside column where churn is higher and prices are lower, and look at which input moves the result most. If a small change in churn flips LTV minus CAC from positive to negative, collect retention data before spending on growth.
Worked example
A software plan costs $60 a month. Variable cost is $18 a month. Monthly churn is 5 percent and CAC is $300. Find contribution, payback in months, and LTV.
LTV minus CAC: $840 - $300 = $540. This is value per customer before fixed costs.
Practice problem and solution
A plan costs $90 a month. Variable cost is $27 and support costs $9 a month per customer. Monthly churn is 4 percent. CAC is $420. Using LTV = monthly contribution / churn, what is LTV minus CAC per customer, in dollars? In one sentence, say why support cost belongs in the contribution.
Mental model: Unit economics = contribution per customer vs. cost to win. Check payback and the sensitivity to churn.
Common trap: Calling revenue growth healthy when each customer loses money.
5. Cost structure, break-even and runway
Learning goal: Separate fixed from variable cost, find break-even volume, and compute how many months of cash a plan really has.
Fixed costs stay the same over the volume range you are considering, such as rent, salaries and software contracts. Variable costs rise with each unit. Many costs are in between: a support team can be fixed for a while and then needs another hire. Write down the range of volume over which each cost estimate is valid.
Break-even volume is fixed cost divided by contribution margin per unit. If fixed costs are $18,000 a month and each unit contributes $30, break-even is 600 units. If price changes, contribution changes, and so does break-even. Raising price by a little can cut break-even sharply, but it also may cut demand, so test it with real customers. Always check break-even against capacity. A plan that needs 1,200 units a month with capacity for 900 is not a plan.
Cash is not profit. Net burn is cash going out minus cash coming in during a month, and runway is cash on hand divided by net burn. If customers pay 30 days after an invoice, revenue exists on paper while the bank balance stays low, and costs still have to be paid. Annual prepayment does the opposite and funds growth. Seasonal businesses should measure runway on the cash path, not on an average month.
Plan with ranges. Compute runway under the base case and a downside case with lower sales and slower collections, then decide in advance what you will cut at six months of runway left. Raising money takes months, so runway below that is an emergency.
Putting it to work: keep a 13-week cash forecast that lists expected receipts by the week they will arrive, not the week they are billed. Update it every Friday and compare it with the bank balance. Decide in advance the runway level at which you will cut costs or start raising money, because decisions made in a cash crunch tend to be worse.
Worked example
Fixed costs are $18,000 a month. Each unit sells for $45 with $15 of variable cost. Find break-even units and then break-even revenue.
Contribution per unit: $45 - $15 = $30.
Break-even units: $18,000 / $30 = 600.
Break-even revenue: 600 x $45 = $27,000 a month.
Check against capacity and the cost of reaching 600 buyers each month.
Practice problem and solution
A shop has $165,000 in cash. Fixed costs are $27,500 a month. It sells 200 units a month, each at $120 with $44 of variable cost. Assuming steady sales and no financing, how many months of runway does the shop have? Round to one decimal. In one sentence, say what the runway figure assumes about sales.
Contribution per unit: 120 - 44 = $76. Monthly contribution: 200 x 76 = $15,200. Net burn: 27,500 - 15,200 = $12,300. Runway: 165,000 / 12,300 = 13.41 months, or 13.4.
Mental model: Break-even = fixed cost / contribution. Runway = cash / net burn. Check capacity and cash timing.
Common trap: Treating profit as cash.
6. Financing options: debt, equity and crowdfunding
Learning goal: Match a funding need to the right source, read SBA program terms accurately, and know what a loan guarantee does and does not do.
Every financing choice trades something. Debt keeps ownership but creates repayment that must be paid from cash flow. Equity gives up ownership and control but has no required monthly payment. Bootstrapping, using customer revenue and personal savings, keeps both but limits speed. The match depends on what the money is for: equipment with a long life suits a loan, a risky prototype with uncertain results suits equity or savings.
The US Small Business Administration describes its 7(a) program as providing loan guaranties to lenders. Borrowers apply through a participating lender, not SBA. The SBA page lists a maximum 7(a) loan of $5 million and says SBA guarantees 85 percent of loans of $150,000 or less and 75 percent above that, and caps interest rates at a margin over the base rate that falls as the loan gets larger: base rate plus 6.5 percent for loans up to $50,000, plus 6.0 percent from $50,001 to $250,000, plus 4.5 percent from $250,001 to $350,000, and plus 3.0 percent above $350,000. These figures were read from SBA pages in October 2026; check the current page before relying on them. The guaranty is a feature for the lender. Your obligations to the lender are set by your loan documents, which you should read.
Other SBA programs have different purposes. Microloans are up to $50,000 through nonprofit intermediaries, with an average of about $13,000, and cannot be used to pay existing debt or buy real estate. The 504 program finances major fixed assets through Certified Development Companies and cannot be used for working capital or inventory. The SBA page for 504 loans shows $5 million in its summary line and $5.5 million in its body, so confirm the cap with a lender or CDC.
Equity crowdfunding under SEC Regulation Crowdfunding lets an eligible company raise up to $5 million in 12 months through an SEC-registered broker-dealer or funding portal, with limits on what non-accredited investors can put in, and securities generally cannot be resold for one year. The next lesson covers equity in more detail. Whatever the source, ask what happens if the plan runs six months behind.
Putting it to work: list every financing need for the next 24 months with its purpose, amount and life of the asset, then match each to a source. Talk to a lender or a nonprofit intermediary before assuming you are eligible; each lender and intermediary sets its own requirements within program rules. Read the loan documents for collateral, personal guarantees and covenants.
Worked example
A 7(a) loan of $40,000 is quoted at the maximum allowed for its size. Assume a hypothetical base rate of 4.0 percent. What is the highest allowed rate?
Loan size $40,000 is under $50,000, so the tier is base rate + 6.5 percent.
Add the margin: 4.0 + 6.5 = 10.5 percent.
Interest for one year on a $40,000 balance would be at most about $4,200. Real loans amortize, so interest falls as principal is repaid.
Practice problem and solution
A $240,000 SBA 7(a) loan has a guaranty of 75 percent because it is above $150,000. A borrower defaults when $180,000 of principal is unpaid. Ignoring interest and fees, assume the guaranty covers 75 percent of that unpaid principal loss. How much of the loss, in dollars, remains with the lender? In one sentence, state the toy-model assumption behind the guaranty share.
The guaranteed part is 0.75 x $180,000 = $135,000. The lender keeps $180,000 - $135,000 = $45,000. This toy model ignores how SBA calculates actual guaranty payments; the sources say only that the guaranty goes to the lender.
Mental model: Match the money to the need. Read program rules for allowed use and who the guaranty protects.
Common trap: Assuming a government guaranty means a borrower owes nothing.
7. Equity financing, dilution and securities exemptions
Learning goal: Compute ownership after a priced round including an option pool, and choose between the main US federal exemptions for selling securities.
Selling equity means selling securities. In the US, offers and sales of securities must be registered with the SEC unless an exemption applies. The SEC small-business pages list the main exemptions, including Regulation Crowdfunding and Regulation D. This lesson covers US federal securities rules only and does not cover state securities law, which can add requirements, or the laws of other countries. It is educational and not legal advice; a securities lawyer should review any real offering.
The arithmetic is simple. Post-money valuation equals pre-money valuation plus new money. An investor who puts in $3 million on a $9 million pre-money valuation owns 3 / 12 = 25 percent after the round. If the investors require an option pool, say 10 percent of the post-round shares reserved for future hires, and the term sheet creates the pool before the round, the pool dilutes only the existing holders. Founders then own less than the headline suggests. Price per share is pre-money divided by the pre-money shares including the new pool.
Rule 506 of Regulation D has two paths. Under Rule 506(b), there may be no more than 35 purchasers who are not accredited in a 90-day period, those purchasers must be sophisticated alone or with a representative, and the issuer may not use general solicitation or general advertising, which Rule 502(c) prohibits. Under Rule 506(c), the issuer may generally solicit, but every purchaser must be an accredited investor and the issuer must take reasonable steps to verify that status. A natural person is accredited under Rule 501 with net worth above $1,000,000, excluding the primary residence, or income above $200,000 in each of the two most recent years ($300,000 jointly with a spouse or spousal equivalent) with a reasonable expectation of the same this year; other categories exist. Regulation Crowdfunding allows up to $5 million in 12 months through a registered intermediary, with limits on non-accredited investors.
The practical rule: choose the exemption before you speak publicly about the offering. A public post describing terms can be general solicitation. Capture each investor's status in writing and ask counsel before any sale.
Putting it to work: build a simple cap table in a spreadsheet and model two or three rounds, including the option pool, before negotiating terms. Compare headline valuation with founder ownership after the pool. Before you tell anyone outside your inner circle that you are raising, decide which exemption you plan to use and ask a securities lawyer to confirm it.
Worked example
Founders and advisors hold 8,000,000 shares. A new investor puts in $3,000,000 at a $9,000,000 pre-money valuation. A new option pool equal to 10 percent of the post-round shares is created before the round and counts in the pre-money. Find the investor percentage and the price per share.
Existing holders own 5,000,000 shares. A new investor invests $2,000,000 at a $6,000,000 pre-money valuation. A new option pool equal to 20 percent of the post-round shares is created and counted in the pre-money. What is the price per share, in dollars, to two decimals? In one sentence, say why the pool is counted in the pre-money.
Mental model: Post-money = pre-money + new money. Option pools in the pre-money dilute existing holders. Choose the exemption before you advertise.
Common trap: Quoting dilution without the option pool.
8. Legal entity choice and tax treatment
Learning goal: Compare sole proprietorship, LLC, S corporation and C corporation on liability, tax and ability to raise equity under US federal rules.
This lesson covers US federal tax and the general SBA description of entity types. State law governs how entities are formed, so check the state where you register. Nothing here is legal or tax advice.
A sole proprietorship is the default when you do business without registering anything. The SBA notes the business assets and liabilities are not separate from the owner's, and that raising money is hard because you cannot sell stock. A limited liability company is created under state statute. The IRS says a domestic LLC with at least two members is classified as a partnership by default, a one-member LLC is treated as part of its owner's return, and an LLC can elect a different classification on Form 8832. The SBA says LLC members are considered self-employed and pay self-employment tax. The IRS states the self-employment tax rate is 15.3 percent: 12.4 percent for Social Security and 2.9 percent for Medicare.
A C corporation is a separate taxpayer. The SBA says corporate profits can be taxed twice, once at the company and again as dividends to shareholders, and that corporations can raise funds by selling stock. An S corporation is a corporation that elects pass-through treatment using Form 2553. The IRS lists the requirements: a domestic corporation, only allowable shareholders (individuals, certain trusts and estates, and not partnerships, corporations or non-resident aliens), no more than 100 shareholders, and one class of stock. The one-class rule and the limit on shareholders make S status awkward for venture-style equity rounds.
Section 1202 of the Internal Revenue Code lets a taxpayer exclude part of the gain on qualified small business stock. The Cornell LII text of 26 U.S.C. 1202 shows an applicable percentage of 50 percent at 3 years, 75 percent at 4 years and 100 percent at 5 years or more for stock acquired after the date of the 2025 amendment, a per-issuer cap equal to the greater of $15 million or 10 times basis, and $75 million of aggregate gross assets. The stock must be issued by a C corporation, acquired at original issue, and the company must be in an active qualified trade or business, with excluded fields that include health, law, engineering, consulting and finance services. This is a reason founders who expect to raise venture capital often start as C corporations, but eligibility is technical.
Putting it to work: write down your founders' plans for equity, tax and exit in plain words, then take that list to a lawyer and an accountant in your state. Ask what changing entity later would cost in tax and paperwork. The SBA notes that conversion is possible but may have tax consequences and restrictions, so the cheapest time to choose well is before you take money or issue shares.
Worked example
A founder holds C corporation stock that qualifies as QSBS and was acquired after the 2025 amendment date. She sells after holding it for 4 years with a $6,000,000 gain. Assuming every other requirement is met, how much gain is excluded?
Hold period of 4 years gives an applicable percentage of 75 percent.
Excluded gain: 0.75 x $6,000,000 = $4,500,000.
The cap is the greater of $15,000,000 or 10 times basis, so the cap is not binding. $1,500,000 remains taxable under the ordinary rules.
Practice problem and solution
Two founders plan a software company. They expect to issue preferred stock to venture investors within a year and want any later QSBS planning to be possible. Which entity should they consider forming, S corporation, LLC taxed as a partnership, or C corporation? Give the entity and a brief reason tied to the IRS S corporation rules and Section 1202.
C corporation. An S corporation may have only one class of stock and cannot have corporate or partnership shareholders, which conflicts with preferred stock and many venture funds. Section 1202 applies to stock of a C corporation. An LLC taxed as a partnership does not issue stock for QSBS purposes.
Mental model: Choose the entity for liability, tax and the equity you plan to issue. S and QSBS rules are technical, so confirm with a professional.
Common trap: Choosing an entity for a low-cost form fee and ignoring the investors.
9. Go-to-market: channels, funnels and sales motion
Learning goal: Pick an ideal customer and channel, model the funnel, and compare channels by customer acquisition cost and payback.
Go-to-market is the plan for reaching the right customers repeatedly at a cost below their value. Start with an ideal customer profile: the narrow group that has the pain, can pay and is easy to reach. A broad launch to everyone produces thin feedback and high cost. Pick one beachhead segment, win it, and expand from there.
Channels have different economics. Self-serve online channels can scale, but they need a product that explains itself and a price low enough to buy without a conversation. Outbound sales fits higher prices and complex decisions, with a longer sales cycle and higher cost per customer. Partners, such as an accounting firm that recommends your tool to clients, can have low CAC because trust is borrowed, but they depend on someone else's priorities. Match the price to the motion: a $20 plan cannot support an hour of sales time.
Model the funnel as stages and conversion rates: reach, lead, trial, paid. Multiply to get customers, and divide spend by customers for CAC. Always test the stage with the biggest drop first. A channel with cheap leads and poor conversion can lose to one with fewer, better leads. Measure each channel separately, because a blended average hides the best and worst.
Early on, go-to-market is learning. Run a few small experiments at a time with a stated threshold, such as at least 20 qualified conversations a week, then drop the channels that miss it. Sales cycle length matters for cash: a six-month enterprise sale requires runway to match.
Putting it to work: choose one channel for the next four weeks, set a spend limit and a weekly target for qualified conversations, and record the conversion at each stage. If a channel misses the threshold twice in a row, change the message or the audience before changing the channel. Keep the number of channels small enough that you can tell which one produced a customer.
Worked example
Channel A: $6,000 spend, 300 leads, 20 percent start a trial, 25 percent of trials pay. Channel B: $9,000 spend, 90 leads, 50 percent trial, 60 percent of trials pay. Find each CAC.
A: 300 x 0.20 = 60 trials; 60 x 0.25 = 15 customers; CAC = $6,000 / 15 = $400.
B: 90 x 0.50 = 45 trials; 45 x 0.60 = 27 customers; CAC = $9,000 / 27 = $333.33.
Channel B costs more in total but less per customer. The better channel is the one with the lower CAC for equal customer value.
Practice problem and solution
Channel A: $7,500 spend, 250 leads, 30 percent start a trial, 20 percent of trials pay. Channel B: $4,800 spend, 80 leads, 50 percent start a trial, 50 percent of trials pay. What is the CAC of A minus the CAC of B, in dollars, rounded to two decimals? In one sentence, say why Channel A's cheaper leads do not make it the cheaper channel.
Mental model: Channel economics are customers per dollar. Test funnel stages and match price to sales motion.
Common trap: Optimizing cost per lead.
10. Pitching: ask, evidence and milestones
Learning goal: Build a short pitch around a defensible ask, size the raise from burn and milestones, and avoid securities-law traps in public pitching.
A pitch is a claim that a specific amount of money will move the company to a specific, checkable point. It is not a summary of the business. The structure is short: the customer problem, the evidence that it is real, how the model earns money, the team, the amount you are asking for, and what that money achieves by when. The SBA describes a traditional business plan and a lean startup plan that is typically one page; a short summary format suits a pitch, and a fuller plan can sit behind it for investors who ask.
Investors look for different things at different stages. A survey of 885 institutional venture capitalists at 681 firms by Gompers, Gornall, Kaplan and Strebulaev, reported by the National Bureau of Economic Research, finds that VCs see the management team as more important than business-related characteristics such as product or technology when selecting investments. This describes the surveyed VCs' stated views, not what every investor does and not a guarantee about outcomes. Use it as a reason to show why this team can execute: relevant experience, evidence of speed in learning and early customer commitments.
Size the ask from a plan, not a round number. Raise enough to reach a milestone that makes the next raise easier, plus a buffer for delays. If monthly net burn is $90,000, an 18-month plan needs $1,620,000, and a 15 percent buffer gives $1,863,000. Say what each part of the money buys: hires, product and customer acquisition. Do not hide the dilution: state the valuation or the instrument terms you propose and be ready to discuss them.
Be careful about where you pitch. A public livestream or demo day can be general solicitation if you are offering securities and plan to rely on Rule 506(b), which Rule 502(c) bars. Coordinate with counsel before posting terms publicly, and consider 506(c) or Regulation Crowdfunding if the offering will be public. Do not make claims you cannot support: cite the source for market numbers and mark estimates as estimates.
Putting it to work: rehearse the pitch for three listeners who have not seen it and ask each to repeat the ask and the proof point from memory. If they cannot, shorten it. Prepare a one-page appendix with the sources for every market number, and a list of the five hardest questions with short, evidence-based answers. Keep a record of who you spoke to and what they were told.
Worked example
A company has $90,000 monthly net burn and wants an 18-month plan with a 15 percent buffer on the total. Find the amount to ask for.
Plan need: $90,000 x 18 = $1,620,000.
Buffer: $1,620,000 x 0.15 = $243,000.
Ask: $1,620,000 + $243,000 = $1,863,000. Round to a figure that is easy to say, such as $1.9 million, and state it.
Practice problem and solution
A founder plans to stream a live public pitch with the offering terms and then accept money from three friends who are not accredited under Rule 506(b). Name the main securities-law problem in one or two words, then explain.
General solicitation. Rule 502(c) bars general solicitation or general advertising in a Rule 506(b) offering, and a public livestream with terms can be one. The founder would need a different route, such as Rule 506(c) with only verified accredited purchasers or Regulation Crowdfunding, or must keep the offering private.
Mental model: Pitch = claim + evidence + ask + milestone. Size the raise from burn. Choose the exemption before you speak publicly.
Common trap: Posting terms publicly before choosing the exemption.