ask 'what demand is poorly served?' (demand-side)
You shipped a time-tracking SaaS in 90 days. Clean code, solid UX, fair Pricing at $12/month. Six months later you have 40 users and $480/month in Revenue. A competitor with worse software and a clunkier UI has 4,000 users paying $29/month. The difference is not engineering skill - they started by finding 4,000 Buyers who hated their current solution, and you started by building something you thought was cool.
Demand-Side thinking means you start every product, feature, and GTM decision by asking 'what Demand is poorly served?' instead of 'what can I build?' It is the single most common failure mode for technical founders - and the reason most software that works perfectly still loses money.
Demand-Side is an orientation, not a metric. It means your first question is always: where are real Buyers spending real money on something that leaves them unsatisfied?
You already know from the Demand lesson that Demand is external to your business, partially hidden, and sets the ceiling on your Revenue. You know from the Buyer lesson that a Buyer is a specific person with a nameable pain using an inferior alternative today.
Demand-Side thinking is the discipline of starting from those two facts before you write a line of code or spend a dollar of Marketing Spend. You are not inventing desire. You are locating existing desire that current solutions serve badly, and asking whether you can serve it better at a Cost Structure that yields Profit.
The opposite orientation is Supply-Side - starting from what you can build. Most engineers default to Supply-Side because building is what they are good at. This is exactly why most engineer-led products fail at Revenue, not at Execution.
Your P&L does not care how elegant your architecture is. It cares about Revenue minus costs. Revenue is bounded by Demand. So the first constraint on your P&L is whether enough poorly-served Demand exists for the thing you are building.
If you get the Demand-Side question wrong, every dollar you spend on Execution, every hire, every feature is an opportunity cost - capital and Labor allocated against a ceiling that is too low to ever produce Profit.
Conversely, if you find a pocket of Demand that is genuinely poorly served:
Demand-Side orientation does not guarantee Profit, but Supply-Side orientation in the absence of validated Demand almost guarantees losses.
Demand-Side analysis is a three-step process:
Step 1: Identify the poorly-served Buyer.
Use the Buyer framework you already know. Find a specific person, their specific pain, and the specific inferior alternative they use today. If you cannot name all three, stop. You do not have a Demand-Side thesis yet.
Step 2: Size the Demand at a Pricing level that supports your Unit Economics.
Demand is quantity at a price. You need to know not just how many Buyers exist, but how many exist at a Pricing level where your Cost Per Unit still leaves margin. A million people who would pay $1/month for something that costs you $3/month to deliver is not viable Demand - it is a trap.
Step 3: Validate that the pain is severe enough to drive switching.
Buyers have inertia. They are already using something. Your differentiation must be large enough that the Buyer's Expected Value of switching (benefit minus switching cost and risk) is clearly positive. This is where most 'better mousetrap' products die - the improvement is real but not large enough to overcome the switching cost.
Signals that Demand is poorly served:
Use Demand-Side analysis before every significant capital decision:
Demand-Side is less useful when:
The key decision rule: if you cannot articulate the poorly-served Demand in one sentence with a specific Buyer, a specific pain, and a specific inferior alternative, you are not ready to allocate capital.
You are a developer considering two product ideas. Both would take about 3 months and $15,000 in personal runway to build.
Idea A (Supply-Side): A beautifully designed invoicing tool for freelancers. You hate your own invoicing workflow and think you can build something better. You estimate 500,000 freelancers in the US use invoicing tools.
Idea B (Demand-Side): You noticed that construction subcontractors in your city complain constantly about getting paid late. They use paper invoices or generic tools that do not integrate with the lien-waiver process their general contractors require. There are ~12,000 subcontractors in your metro area.
Idea A Demand-Side check: Buyer = freelancer. Pain = 'invoicing is annoying.' Inferior alternative = FreshBooks, Wave, Stripe Invoicing, PayPal. Problem: the existing alternatives are cheap ($0-15/month), well-funded, and rated 4+ stars. The Demand exists but it is well-served. Your differentiation would need to be enormous to overcome switching costs. Realistic Market Share capture: maybe 200 users at $10/month = $2,000/month Revenue = $24,000/year. At $15,000 cost to build plus ongoing hosting and support, your break-even is marginal and your Lifetime Value per Buyer is low because Churn will be high (low switching cost means Buyers leave easily too).
Idea B Demand-Side check: Buyer = construction subcontractor (owner-operator, $200K-$2M annual Revenue). Pain = getting paid 60-90 days late because lien waivers are manual and general contractors use the paperwork delay as float. Inferior alternative = paper forms, generic PDF templates, sometimes nothing. The pain is acute - late payment is a Cash Flow crisis for small operators. CSAT on current solutions is essentially zero because there is no real solution. Your differentiation is clear: automate the lien-waiver-to-payment workflow.
Idea B Unit Economics check: If 5% of 12,000 subcontractors adopt = 600 Buyers. Pricing at $49/month (justified because the pain is Cash Flow, not convenience - you are accelerating tens of thousands in receivables). Revenue = 600 x $49 = $29,400/month = $352,800/year. Cost Per Unit with cloud hosting and support: ~$8/month. Profit per Buyer: $41/month. Payback Period on $15,000 investment: under 2 months after reaching 600 users.
Insight: Idea A has 40x more potential Buyers but the Demand is well-served. Idea B has far fewer Buyers but the Demand is poorly served, the pain is severe (Cash Flow, not convenience), and the Pricing power is 5x higher because no adequate alternative exists. Demand-Side thinking would reject Idea A and pursue Idea B despite the smaller total addressable Demand.
You run a $40,000/month ARR project management tool for marketing agencies. You have 200 paying Buyers at $200/month. Your Churn Rate is 6% monthly. Your engineering team can build one major feature this quarter. Two candidates:
Feature X: AI-generated status reports (your team is excited about it)
Feature Y: Client-facing approval workflows (you keep hearing about this in cancellation surveys)
Demand-Side signal check for Feature X: Source of idea = internal team excitement (Supply-Side signal). Number of Buyers who requested it = 3 out of 200. Current workaround Buyers use = copy-paste into ChatGPT (free, 30 seconds). Severity of pain = low. This is a Supply-Side feature - it starts from 'what can we build?' not 'what Demand is poorly served?'
Demand-Side signal check for Feature Y: Source of idea = cancellation surveys (direct Churn signal). Number of Buyers who cited this = 38 out of the last 60 churned Buyers (63%). Current workaround = email chains with clients, screenshots, lost approvals. Severity of pain = high (agencies lose billable hours to approval chaos, and some lose clients over missed deliverables).
P&L impact calculation: Current Churn = 6%/month = 12 Buyers/month lost = $2,400/month in lost Revenue. If Feature Y reduces Churn by half (conservative, since 63% of churned Buyers cited it), you save 6 Buyers/month = $1,200/month = $14,400/year in retained Revenue. Over 2 years with Compounding retention, the cumulative Revenue saved exceeds $35,000. Feature X has no measurable Demand signal to project against.
Insight: Cancellation surveys are one of the purest Demand-Side signals available. When Buyers tell you why they left, they are telling you exactly what Demand you served poorly. Feature Y addresses demonstrated, poorly-served Demand. Feature X addresses your engineering team's enthusiasm - a Supply-Side signal that tells you nothing about Buyer pain.
Demand-Side means starting every decision from 'what Demand is poorly served?' - not 'what can I build?' The distinction is the single biggest predictor of whether a technical founder's product reaches sustainable Revenue.
Poorly-served Demand has observable signals: high Churn on incumbents, low CSAT with high Market Share, Buyers building workarounds, and Buyers paying for bundles when they only need one piece. If you cannot find these signals, the Demand may be well-served already.
The size of the Demand matters less than how poorly it is served. 600 Buyers with acute pain and no alternative will generate more Revenue and Profit than 50,000 Buyers with mild inconvenience and ten competitors.
Confusing your own frustration with poorly-served Demand. You are one person. Your annoyance with a tool is a Supply-Side signal (it tells you what you can imagine building), not a Demand-Side signal (it tells you nothing about how many Buyers share the pain or whether existing alternatives are actually inadequate for them). Validate with Buyers who are not you.
Sizing Demand without a Pricing constraint. Saying 'there are 10 million potential users' is meaningless without asking 'at what Pricing level, and does that Pricing level cover my Cost Per Unit?' A large Demand at a price below your Cost Structure is worse than a small Demand at a price with healthy margin - the large number just tricks you into spending more before you discover the Unit Economics do not work.
You are evaluating a B2B SaaS idea: a compliance-tracking tool for small medical practices (1-5 doctors). You estimate 120,000 such practices in the US. List the three Demand-Side questions you would need to answer before writing any code, and describe what a positive signal would look like for each.
Hint: Use the Buyer framework (who, what pain, what inferior alternative) and then add the Pricing and switching-cost checks from this lesson.
Q1: Who is the Buyer and what is their specific pain? Positive signal: The office manager (not the doctor) spends 8+ hours/week on compliance paperwork, has been fined or warned by a regulator in the past 2 years, and considers it the worst part of their job.
Q2: What inferior alternative do they use today? Positive signal: They use spreadsheets and paper checklists, or they pay a consultant $500+/month who is unreliable. If they already use a well-rated software tool at a reasonable price, the Demand is not poorly served.
Q3: At what Pricing can I serve this Demand and still have positive Unit Economics? Positive signal: Practices currently spend $500+/month on consultants or eat $2,000+ in fines annually, so a $99-199/month tool has clear value. Your Cost Per Unit (hosting, support, compliance database updates) is under $20/month, leaving $79-179 in margin.
Your existing product has 1,000 Buyers at $50/month ($50,000 ARR). Your Churn Rate is 4%/month. You survey 40 recently churned Buyers and find: 22 say 'too expensive,' 11 say 'missing integration with Salesforce,' and 7 say 'switched to competitor X.' Rank these three Churn reasons by Demand-Side signal strength and explain your ranking.
Hint: Think about which reasons point to poorly-served Demand versus Pricing mismatch versus competitive loss. Which ones tell you something actionable about what the Buyer actually needs?
Rank 1: 'Missing Salesforce integration' (11 Buyers). This is the strongest Demand-Side signal. These Buyers have a specific, nameable pain (manual data transfer between your tool and Salesforce), an inferior alternative (doing it by hand or using Zapier), and they told you exactly what they need. You can estimate the Revenue impact: if building the integration retains 11 of every 40 churned Buyers, that is a 27.5% reduction in Churn, saving ~11 Buyers/month x $50 = $550/month = $6,600/year.
Rank 2: 'Switched to competitor X' (7 Buyers). Moderate signal. This tells you a competitor is serving Demand better, but you need to dig deeper - what specifically does competitor X do that you do not? The answer to that question is the Demand-Side insight.
Rank 3: 'Too expensive' (22 Buyers). Weakest Demand-Side signal despite being the most common response. 'Too expensive' usually means 'I do not perceive enough Value Creation to justify the Pricing' - it is a differentiation problem, not a Pricing problem. Cutting price destroys margin without fixing the underlying poorly-served Demand. The real question is: what would make these Buyers feel $50/month is a bargain? That answer is the Demand-Side insight.
Demand-Side thinking builds directly on your understanding of Demand (the external ceiling on Revenue) and Buyer (the specific person with nameable pain). Where Demand told you what exists and Buyer told you who specifically, Demand-Side tells you how to orient your decisions - always starting from poorly-served Demand rather than from your own capabilities. This connects forward to Supply-Side, which is the complementary orientation (what can you build or deliver efficiently?). Strong Operators use both: Demand-Side to pick what to build, Supply-Side to optimize how to deliver it. It also feeds directly into Pricing (you can only price well if you understand what pain the Buyer is escaping), differentiation (your edge comes from serving Demand that others serve poorly), customer segmentation (which Buyer groups have the most poorly-served Demand), and Unit Economics (whether the Demand you found can be served at a Cost Structure that produces Profit).
Disclaimer: This content is for educational and informational purposes only and does not constitute financial, investment, tax, or legal advice. It is not a recommendation to buy, sell, or hold any security or financial product. You should consult a qualified financial advisor, tax professional, or attorney before making financial decisions. Past performance is not indicative of future results. The author is not a registered investment advisor, broker-dealer, or financial planner.