Market Research Playbook | Idea Score

Introduction

Great teams make better product bets by doing market research that is fast, focused, and falsifiable. At the Market Research stage, your goal is to size demand, find an entry wedge, understand incumbents, and identify where competition is weakest. You are not building yet. You are collecting evidence that lowers uncertainty and clarifies whether the problem and buyer are strong enough to support a real business.

This playbook gives a step-by-step process for market-research that avoids vanity metrics and gets you to a go or no-go decision. It covers the signals to collect now, practical ways to size demand without fantasy spreadsheets, and a scoring approach that protects you from overfitting thin data. If you prefer a structured workflow with automated competitor summaries and scorecards, Idea Score can operationalize the steps below so you spend your time on judgment, not manual scraping.

What needs to be true at this stage

Before you write a single line of code, validate these conditions with lightweight, disconfirming tests. Use hard thresholds so you can stop early when the answer is no.

Active demand exists - search queries, job posts, community threads, and reviews show recurring pain and real spend. Threshold: at least 3 independent demand signals pointing to the same problem within your target segment.
A clear buyer with budget - you can name the job title that signs or swipes a card, typical annual spend on adjacent tools, and one trigger that creates urgency. Threshold: at least 5 buyers describe a recent triggering event that would have justified purchase.
A wedge you can win - a narrow starting point where you deliver a 10x better experience or 2x faster workflow. Threshold: a single workflow with measurable time or error reduction that incumbents do poorly.
Accessible channel - you can predictably reach buyers through SEO, integrations, communities, or partnerships that incumbents underuse. Threshold: at least one channel with an acquisition cost hypothesis you can state and test.
Competitor gaps - something important is missing or poorly executed among incumbents. Threshold: 3+ negative review patterns or pricing tiers that punish exactly the use case you plan to serve.
Workable unit economics - basic back-of-napkin math indicates positive CAC to LTV potential for the wedge. Threshold: a plausible path to 3x or better LTV to CAC within 12 months at small scale.

Research inputs and evidence worth collecting now

Favor cheap evidence you can gather in days over bespoke research that drags on. Triangulate from multiple sources to avoid single-signal bias.

1) Demand signals you can collect in a week

Search and intent - benchmark monthly volume and CPC ranges for 5 to 10 buying-intent keywords. Healthy early markers: 2,000+ monthly queries across head and mid-tail terms, with CPC above 3 USD indicating perceived commercial value. Check click distributions and SERP features to gauge organic capture potential.
Job posts - scrape job boards for roles mentioning the workflow, tools, or mandated KPIs. Count frequency of named tools and responsibilities. If 50+ postings in your target region mention the pain or tool migration, it suggests pipeline for outreach and partner co-selling.
Community threads - mine Reddit, Stack Overflow, Slack communities, and vendor forums for recurring issues. Look for clusters like repeated error IDs, API quotas, or compliance blockers. Document thread links, dates, and severity language used by practitioners.
Review sites and app marketplaces - export recent G2, Capterra, and app marketplace reviews. Tag mentions of missing features, slow support, confusing pricing, or integration gaps. Three or more reviews citing the same gap within the last quarter is a live opportunity.

2) Competitor mapping that finds weak spots

Onboarding time - measure time-to-first-value. Create a stopwatch log for top competitors going from signup to the first real output. Anything above 30 minutes for SMB tools indicates friction you can exploit.
Pricing page traps - log paywalls, seat minimums, and overage fees. Note if critical integrations are paywalled to higher tiers. A pattern where SMBs are forced into enterprise tiers is a fertile wedge for a fairer offering.
Integrations matrix - build a yes or no table for the top 15 ecosystem integrations relevant to your wedge. Missing integrations with top platforms in your niche signal defensible differentiation with minimal build.
Policy and compliance constraints - catalog privacy, data residency, and audit trail requirements by segment. Incumbents that treat compliance as a bolt-on leave room for a product that bakes it in.

3) Buyer and segment clarity

Who buys and why - interview 5 to 10 buyers for 20 minutes each. Ask for the last time they solved the problem, the cost of inaction, and what they tried first. Collect the exact phrases they used in their search, the vendors they considered, and what blocked a purchase.
Segment economics - estimate typical revenue, headcount, and tool budgets for your intended buyers. You need enough budget headroom to support your planned pricing and expansion.
Switching signals - ask about migration history, data lock-in, and integration dependencies. If buyers cite brittle exports or specialist admin effort as blockers, your wedge should reduce switching cost up front.

4) Bottom-up demand sizing

Skip top-down TAM slideware. Use a quick bottom-up estimate tied to reachable customers:

Define the Ideal Customer Profile by industry, size, and stack. For example, US-based marketing agencies on QuickBooks Online with 10 to 50 employees.
Count reachable accounts via public filters or marketplaces. Example: 35,000 agencies in NA that use QuickBooks Online and HubSpot.
Estimate conversion using similar tools' adoption rates or your channel math. Example: 2 percent annual adoption in niche communities yields 700 accounts.
Multiply by initial ARPA. Example: 700 accounts at 150 USD per month equals 1.26 million USD annual revenue potential for the wedge.

Document ranges, not single point estimates. State what would need to be true for the numbers to hold, and what would falsify them.

5) Quick stage landing experiments

At the Market Research stage, it is fine to run a stage landing page that tests demand and messaging fit. Do not oversell or collect preorders. Instead, validate search language and target segments with:

A one-page pitch focused on the wedge and the workflow you accelerate
One clear CTA for a research interview or waitlist signup
3 to 5 ad groups that mirror your keyword clusters, targeting the buyer title
Benchmarks: a clickthrough rate above 2 percent and an interview signup rate above 5 percent indicate resonance in early tests

For niche inspiration and validation patterns, see Micro SaaS Ideas: How to Validate and Score the Best Opportunities | Idea Score.

How to score ideas without overfitting early data

Early market-research data is sparse and noisy. Treat scores as directional, not definitive. Anchor your scoring to evidence grades so low-confidence inputs do not dominate decisions.

Scorecard dimensions and thresholds

Pain intensity - 1 to 5. Use language and behavior evidence. A 5 means frequent phrases like "blocked" or "breach risk", plus clear actions such as scripts, hacks, or manual spreadsheets.
Budget and willingness to pay - 1 to 5. A 5 requires proof of spend on adjacent tools or line items, and job posts that mention the mandate. Tie to observed pricing of incumbents.
Urgency and triggers - 1 to 5. A 5 requires time-bound triggers like audits, renewals, seasonal surges, or vendor deprecations that force decisions.
Reachable demand size - 1 to 5. A 5 requires bottom-up reach in thousands of accounts or hundreds with high ARPA, validated by directories, partner lists, or marketplaces.
Competitive gap - 1 to 5. A 5 requires multiple negative review clusters and verifiable missing integrations or unfavorable pricing structures for your segment.
Access to buyer - 1 to 5. A 5 requires at least one low-cost channel with clear targeting and lead access such as an integration marketplace or high intent SEO.

Evidence grading and confidence weighting

Assign each dimension a confidence multiplier based on evidence quality:

A-grade - buyer interviews with transcripts, paid intent data, or hands-on competitor tests. Confidence multiplier 1.0.
B-grade - public benchmarks, review analysis, aggregated search data. Confidence multiplier 0.8.
C-grade - anecdotal threads, friend feedback, assumptions. Confidence multiplier 0.5.

Final dimension score equals the raw score times the confidence multiplier. For example, a competitive gap score of 4 with B-grade evidence yields 3.2. Sum across dimensions for a 30-point max, then convert to a 0 to 100 scale for easier comparison.

Automated reports and scorecards inside Idea Score apply similar weighting, which helps teams prevent a single rosy data point from inflating the whole idea.

Avoiding overfit and false precision

Keep categories coarse - 5-point scales and broad bins are enough at this stage.
Use ranges and sensitivity checks - optimistic, base, and conservative scenarios for demand and pricing.
Track contradictions - when interview and search signals conflict, log both and note which you would need to overturn.
Timebox research cycles - one to two week sprints with a fixed number of interviews and data sources per cycle.

Mistakes that create false confidence at this stage

TAM theater - using top-down market size to justify any idea. If you cannot name the first 200 accounts you can actually reach, you do not have a real SAM.
Friend and founder bias - coding urgency from peers, not buyers. Early praise without budget signals does not count.
Feature mirroring - copying competitor features without verifying the job and context that made those features necessary.
Misreading search demand - head terms look big but convert poorly. Favor long-tail intent and compare advertiser density to gauge commercial value.
Ignoring switching costs - underestimating data migration, governance, or custom workflows that keep buyers stuck even when they dislike a tool.
Channel blindness - assuming you can acquire users through the same saturated channels incumbents dominate, instead of looking for thin-edge channels or integration ecosystems.
Premature solutioning - writing a feature spec before you can articulate the buyer's trigger, success metric, and the alternative they use today.

What a strong decision memo looks like before moving on

The memo is your stage landing artifact. It summarizes evidence, scorecard results, and a clear yes or no with next tests. Keep it concise, specific, and falsifiable.

Sections to include

Problem and wedge - one sentence problem plus your thin-edge workflow. Example: automate invoice chasing for QuickBooks Online marketing agencies with under 50 staff, reducing DSO by 7 days.
Buyer and trigger - buyer title, budget owner, and two triggers. Example: agency operations manager, cash flow crunch after tax season, and lost revenue due to late payments.
Demand evidence - 3 to 5 strongest signals. Example: 2,400 monthly searches across long-tail terms, 65 recent reviews complaining about payment reminders, 120 job posts mentioning DSO targets.
Competitor gaps - 3 patterns. Example: slow onboarding, paywalling automated dunning to enterprise tier, no native Slack reminders.
Bottom-up size - reachable accounts and ARPA range. Example: 30,000 target accounts, 1 to 3 percent adoption yields 300 to 900 customers at 100 to 200 USD per month equals 360,000 to 2.16 million USD ARR potential for the wedge.
Scorecard summary - dimension scores with confidence. Example: Pain 4A, Budget 3B, Urgency 4B, Reachable demand 3B, Competitor gap 4A, Access 3B. Weighted total 72 out of 100.
Risks and unknowns - top 3. Example: low response rates from agencies on cold outreach, potential QuickBooks API limits, seasonality affecting signup rates.
Next tests and kill criteria - 2 week plan. Example: 10 buyer interviews, 2 onboarding tear-downs, and 1 landing test. Kill if interview conversion falls below 3 percent and payment reminder automation is considered non-critical by 70 percent of buyers.

If you operate as a team, align on the memo format so you can compare ideas side by side. For collaborative workflows and shared scorecards, see Idea Score for Startup Teams | Validate Product Ideas Faster.

Conclusion

Market research is not a pro forma step. It is where you size demand, find the wedge that gets you in the door, and discover competitor gaps that translate into a pricing and positioning advantage. Keep the work fast and practical: gather multiple independent signals, quantify the reachable market from the bottom up, and weight scores by evidence quality. The outcome is a crisp decision memo that says what must be true and how you will falsify it next.

The right tools accelerate the grind without replacing your judgment. Automated reports, competitor summaries, and evidence-weighted scoring in Idea Score help founders and teams focus on the decisions that matter, not on wrangling spreadsheets. If your idea sits in workflow automation or micro SaaS, explore patterns and validation tactics in Workflow Automation Ideas: How to Validate and Score the Best Opportunities | Idea Score.

FAQ

How much research is enough before I decide to proceed?

Timebox one to two weeks per idea. Aim for 5 to 10 buyer interviews, a clear bottom-up size, and a scored view of demand, budget, urgency, competition, and access. If you cannot find 3 independent signals for demand and at least one competitive gap, pause or pivot the wedge.

What if search volume is low but the problem seems painful?

Plenty of B2B niches are low search but high value. Look for alternate signals: job posts with the mandate, active community threads, and high CPC on narrow queries. Map where buyers hang out and validate outreach conversion via communities or partner channels. Low search is not a veto if budgets and urgency are strong.

How do I avoid bias in expert or friend interviews?

Recruit outside your network through communities or job boards. Use a fixed script, avoid leading questions, and ask for the last time they solved the problem, what they tried first, and what they would stop paying for to fund your solution. Record and transcribe so you can code themes rather than remember selectively.

What is a realistic first-year revenue target from a wedge?

Work backward from reach and conversion. Example: a niche with 5,000 reachable accounts, 1.5 percent conversion, and 120 USD monthly ARPA equals roughly 108,000 USD ARR. Increase assumptions only when you have channel data and proof of budget. Expansion can come from adjacent workflows once the wedge sticks.