Market research checklist before you spend on cold traffic ads

Cold traffic does not care that your team is smart. It cares whether the first second feels recognizable, the next seconds feel credible, and the landing feels continuous.

This checklist is written for the night before you slide budget from “learning” to “we need revenue now.”

Last reviewed: April 2026. Platform policies and attribution settings change—confirm current Meta and TikTok guidance for your account type.

The checklist (copy into Notion or print)

A. Who you will lose money learning from

• ICP v0 written in half a page (situation, trigger, constraint, proof bar)
• Non-customers named explicitly (“we will not buy clicks from X segment yet”)
• One primary metric for this sprint (not twelve)

B. What you are allowed to say

• Claims policy attached (numbers, superlatives, health/finance rules if any)
• Must-not-say list from legal or risk
• Proof artifacts chosen (demo clip, stat, customer clip, policy badge)

C. What competitors already own in the mind

• Five to eight substitutes captured with Ads Library links
• Three saturated angles identified (“everyone says this”)
• One white-space hypothesis tied to VoC

D. Creative + landing continuity

• Hook → body → LP tells one story (read aloud test)
• Mobile pass on LP in the same thumb posture as the ad
• Load and clarity checked—vanity animations cost conversions

E. Operations (the boring hero)

• Naming + folder convention for assets
• Kill rules written before spend (CPA, frequency, creative fatigue)
• Friday retro scheduled—calendar invite, not a hope

One-line “go / no-go”

If you cannot check every box in A + B + D, do not call it research—call it gambling.

Where Pinnacle AdForge helps

Run market awareness and synthesis so the checklist lives next to hooks and assets—signup when you want the chain in one workspace.

Bounded cold tests (money psychology)

Write the maximum you are willing to spend to learn, not to “win,” on the first cold sprint. Learning budgets have stop dates; ego budgets have invoices. Pair spend caps with the kill rules in section E so nobody negotiates with themselves at midnight in Ads Manager.

Competitive saturation scan (thirty minutes)

List the three loudest claims in your category’s ads this month. If your planned hook is word-for-word in that list, you are not testing positioning—you are testing who has the bigger wallet. Rewrite toward buyer language from VoC, not vendor language from homepages.

E-E-A-T: cite platform standards before you cite ambition

Meta and TikTok publish advertising standards and category restrictions. If your product touches health outcomes, financial results, politics, or kids, your pre-spend checklist should include a policy pass link saved next to creative—same weight as the landing QA link.

The “Monday morning” doc (one page, no deck)

Media buyers deserve a single page: ICP v0, three hooks with proof, top objections, must-not-say list, LP URL, and kill rules. If your research output is only a deck, you externalized the cost of understanding to the person holding the spend keys—rude, inefficient, and oddly common.

When to delay cold (without shame)

Broken payments, misleading pricing on LP, known fulfillment SLA misses—fix plumbing first. Cold traffic is a magnifier; it will magnify leaks as confidently as it magnifies wins.

Creative-LP parity check (read aloud, once)

Read the first line of the ad and the hero headline back-to-back. If you laugh—not in the good way—you found a continuity bug cheaper to fix in Docs than in spend.

Benchmarks without self-deception

Write down one external benchmark you will trust (industry report, peer cohort, or last year’s same season). If you have none, your “success” definition is mood-based. Mood-based optimization is how grown adults refresh dashboards like slot machines.

Stakeholder map (who gets a veto)

List who can stop a test: legal, brand, finance, founder, platform policy. If someone is not named, they will appear surprisingly on Wednesday anyway—invite them early with a one-pager, not a crisis thread.

Post-mortem template (even if the test “won”)

• What did we believe?
• What did we observe?
• What will we stop saying?
• What will we try next?

Winning tests without post-mortems teach ego; losing tests with post-mortems teach math.

Seasonality note (write it before July pretends it is January)

If your product is seasonal, your pre-spend checklist needs a calendar column: what was true last peak may be false this peak (inventory, shipping, competitor promos). Cold traffic has memory; your LPs should too.

One honest metric hierarchy

Pick one north-star for the first cold sprint (e.g. qualified demo, purchase, or start trial)—then list two guardrails (frequency, CPM creep). Three numbers beat thirty that nobody acts on.

Attribution humility (last click lies sometimes)

Write one sentence: what we will not pretend to know after two weeks (incrementality, full funnel lift). Humility prevents hero narratives that age into embarrassing decks.

Payment and refund path QA (unsexy, costly if skipped)

Click the same path a cold user would: pay, fail a card on purpose once, request support. If that journey feels hostile, ads will import hostility at scale.

Key takeaways

• Kill rules before spend—emotions spike after money, not before.
• Continuity beats clever—ad and LP are one product experience.
• Non-customers named saves more money than a clever audience hack.

People also ask

What market research should I do before running cold ads?

Clarify who is not a customer, what promise you can prove fast, what competitors already claim, and what landing truth matches the hook. Then define success metrics and kill rules before spend.

How much should I spend on cold traffic tests?

Enough to reach statistical or operational clarity for your primary metric—often a bounded weekly test budget with pre-written kill rules rather than open-ended spend.

What is the difference between warm and cold traffic research?

Warm audiences already know you; cold audiences owe you nothing—your research must explain first-second recognition and immediate credibility.

Do I need a landing page before cold ads?

You need a destination that completes the promise in the ad—often a dedicated landing path for cold even if you also have a homepage.

What documents should research produce for media buyers?

A one-page ICP v0, three angles with proof, objection list, and must-not-say list—plus links to sources.

FAQ

How long should pre-launch market research take?

Days to two weeks for focused sprints beats month-long perfectionism that misses the quarter.

What is the biggest pre-launch research mistake?

Beautiful decks with no testable customer language attached to hooks and landing pages.

How does Pinnacle AdForge support pre-spend research?

AdForge chains market awareness and synthesis into messaging so research becomes assets instead of a PDF. Start with market awareness research.

How do I try Pinnacle AdForge before scaling cold spend?

Signup, run one bounded research-to-hook sprint in a project, then scale only what survives kill rules.

---

Checklists age—revisit this after major product, policy, or positioning changes.