All chapters Part IX · Go to market
Chapter 48

Paid Acquisition

Paid acquisition is the only channel where you can turn money directly into users, which is exactly why it's dangerous. It amplifies whatever it's pointed at. Point it at a product that doesn't convert or retain, and you've built a machine that turns money into nothing, faster. Paid doesn't fix a broken funnel; it bills you to discover it's broken.


The concept

Paid acquisition is buying attention, ads on search, social, app stores, or elsewhere, that sends people to your product. Unlike content and social (Chapters 46-47), it's instant and it's rented: traffic flows while you pay and stops the moment you stop.

The entire discipline reduces to one inequality:

 LTV  >  CAC        (with margin, and paid back fast enough to survive)

 CAC  = what it costs to acquire one paying customer
 LTV  = what that customer is worth to you over their lifetime

If a customer is worth more to you than they cost to acquire, by a healthy margin, and soon enough that you don't go broke waiting, paid works and scales. If not, paid is a machine that loses money on every unit and loses it faster the more you spend. A common healthy target is LTV at least 3× CAC, with CAC paid back within months, not years.

Why paid comes after product-market fit, not before

This is the chapter's central warning: paid amplifies; it does not create. It multiplies whatever your funnel already does.

You cannot buy your way to product-market fit. Prove the funnel converts organically first (Parts VII, X), then use paid to scale what already works. Founders who start with paid usually just buy an expensive, precise measurement of how badly their product retains.

📐 Paid amplifies a funnel; it does not create one. Point it at a product that converts and retains, and it pours fuel on a working fire. Point it at one that doesn't, and you've built a machine that turns money into nothing, faster. Paid is a reward you earn with a proven funnel, not a shortcut to finding one.

The channels, briefly

ChannelIntentBest for
Search ads (Google)High, they're searching for itCapturing existing demand
Social ads (Meta, TikTok)Low, interrupting a scrollCreating demand, visual products
App store ads (ASA)High, searching the storeApps, bottom-funnel
Influencer / creatorVaries, borrowed trustNiche audiences, authenticity
Newsletter / niche sponsorshipsMedium, targeted, trustedReaching a specific audience

Search captures demand that exists; social creates demand that doesn't. Match the channel to whether people are already looking for what you sell.

Start small, measure, then scale

The correct shape of a paid experiment:

 1. small budget — enough to get signal, little enough that a wrong answer is cheap
 2. measure CAC and, crucially, whether those users RETAIN (Chapter 50)
 3. LTV > CAC with margin?  → scale gradually, watching the numbers hold
                            → NO: kill it, fix the funnel, or try another channel

The cardinal sin is scaling before you've proven the unit economics. A channel that looks fine at $10/day can bleed at $1,000/day, and CAC almost always rises as you scale (you exhaust the cheapest audience first).

Attribution is hard and getting harder

Knowing which ad produced which paying, retaining customer is genuinely difficult, privacy changes (iOS ATT), cookie deprecation, and multi-touch journeys all blur it. Don't over-trust any single platform's self-reported numbers (every ad platform grades its own homework and claims credit generously). Watch your blended CAC, total spend divided by total new customers, as the honest backstop.

The bootstrapper's stance on paid

For a bootstrapped founder with limited cash, paid is usually not the first channel, for three reasons: it requires a proven-converting funnel you may not have yet; it requires money you may not have; and it teaches you less about your customer than doing organic distribution yourself. The small budget you do have is often better spent building the owned, compounding channels (content, social, community, Chapters 46-47) than renting traffic that stops the day you stop paying. Reach for paid to scale a proven funnel, not to find one.


📐 Best practice

Prove the funnel converts organically before spending on paid.

Know your LTV and CAC, and require LTV > CAC with healthy margin (aim ~3×+).

Start with a small budget sized so a wrong answer is cheap.

Measure retention, not just installs, acquired users that churn are a loss.

Match the channel to intent, search for existing demand, social to create it.

Scale gradually, watching CAC rise and unit economics hold.

Watch blended CAC as the honest backstop; distrust platform self-reporting.

Kill losers fast, a channel that doesn't hit the math is a leak.

Prefer owned channels early; use paid to scale, not to discover, fit.

Keep creative fresh, ad fatigue raises CAC over time.


💀 Common mistakes

⭐ Starting with paid before the funnel converts. Paying to fill a leaky bucket; buying an expensive measurement of how badly you retain.

Not knowing CAC or LTV, spending blind, with no definition of success.

Measuring installs/clicks, not paying-and-retaining customers.

Scaling before the unit economics are proven, fine at $10/day, bleeding at $1,000.

Over-trusting platform-reported attribution, every platform over-claims credit.

Ignoring that CAC rises as you scale, the cheap audience runs out.

Renting traffic instead of building owned channels with a limited budget.

Set-and-forget, no monitoring, no fresh creative, rising CAC unnoticed.

Chasing vanity reach, impressions and clicks that never convert.

Confusing a channel problem with a product problem, no ad fixes a product people don't want.


The professional workflow

 1. PREREQUISITE: funnel converts and retains ORGANICALLY (Parts VII, X)

 2. KNOW YOUR NUMBERS — target CAC, estimated LTV, payback period

 3. PICK ONE CHANNEL matched to intent

 4. SMALL TEST BUDGET — sized so a wrong answer is cheap

 5. MEASURE — CAC and, crucially, whether acquired users RETAIN

 6. DECISION
    LTV > CAC with margin? → scale gradually, watch the math hold
    NO → kill it; fix the funnel or try another channel

 7. WATCH BLENDED CAC; distrust single-platform attribution

 8. REFRESH CREATIVE to fight ad fatigue

 9. ADD CHANNELS only after one is proven and profitable

Tools, websites & costs

NeedToolCost
Search adsGoogle AdsYour budget
Social adsMeta Ads, TikTok AdsYour budget
App store adsApple Search AdsYour budget
Creator/influencerDirect outreach, marketplacesVaries
Newsletter sponsorshipsSwapstack, directVaries
Attribution (mobile)AppsFlyer, Adjust$$-$$$ (later)
Conversion trackingPlatform pixels + your analytics (Chapter 38)$0
CreativeCanva, CapCut$0
Landing pagesYour site, Unbounce$0-$$

The tools are free; the spend is the cost, and it's real money you can lose. Unlike every other channel in this Part, paid's downside is measured in dollars, not just time. Treat the first budget as tuition and cap it.


Alternatives & trade-offs

Paid vs organic. Paid is instant, scalable, and rented, it stops when you stop and requires a proven funnel. Organic (content, social, community) is slow, compounding, and owned. Bootstrappers usually start organic and add paid to scale a proven funnel; funded companies can start paid earlier because they can afford to learn expensively.

Search vs social ads. Search captures high-intent existing demand (people looking for a solution) and usually converts better; social creates demand and reaches people not yet looking, at lower intent but larger scale. Start with whichever matches your demand type.

Paid vs the same money into product/content. A small budget spent on ads buys a burst of rented traffic that ends; the same money (or time) spent building an owned channel or improving conversion compounds. Early, the compounding investment usually wins, reserve paid for when you have a funnel worth pouring fuel on.

DIY vs agency. DIY is cheaper and teaches you the channel; an agency brings expertise and markup and is worth it only once you have budget and proven economics to scale. Never hand an agency an unproven funnel, you'll pay them to spend your money into a leaky bucket.

Broad vs narrow targeting. Broad targeting reaches scale and wastes spend on the wrong people; narrow targeting converts better and caps your ceiling. Start narrow (high intent, cheap to validate), widen only once the economics prove out.


Checklist


📓 Case Study: the paid channel that was correctly never used

Project: SOLIS. Unusually, the most defensible decision here was a decision not to do the thing the chapter is about.

⭐ Paid acquisition was deliberately deprioritised in favour of owned channels, the right call. With a limited budget and an unlaunched, never-organically-validated product, the plan explicitly favoured building the content engine (Chapters 46-47) over renting traffic through ads. That's exactly the bootstrapper's stance this chapter argues for: with a small budget and an unproven funnel, the money and effort belong in owned, compounding channels, not in paid traffic that stops the day you stop paying. Spending on ads would have violated the chapter's central rule, paid amplifies a proven funnel; it can't create one, because the SOLIS funnel had never been proven to convert or retain anyone.

Why using paid would have been a mistake here, a clean illustration of the core warning. SOLIS had, at the end:

Every single input the paid inequality (LTV > CAC) requires was missing. Running ads into that would have been the textbook error: paying to send strangers into an unmeasured, unvalidated, leaky funnel, buying an expensive measurement of how badly the product retained, when the product hadn't even launched. Not spending was correct.

⚠️ But the reason paid was right to skip is itself the deeper problem. Paid was correctly deprioritised because the funnel was never validated, and the funnel was never validated because the product never launched (Chapter 45). So the "right call" on paid sits on top of the project's central failure. Skipping paid was correct; the reason it was correct (nothing was ever proven) is the thing that should have been fixed. A validated, launched product might well have earned the right to test paid, SOLIS never got there.

🚩 Entirely untested, by design. No ad was ever run, no CAC ever measured, no channel ever tested. There is no data, which, uniquely for this Part, is the correct outcome: with nothing proven, there was nothing to amplify. The lesson is the discipline of not spending, not any result from spending.

What generalises:

  1. Not spending on paid is often the right early call. With a small budget and an unproven funnel, owned channels beat rented traffic.
  2. Every input paid requires, conversion, retention, LTV, CAC measurement, must exist first. If any is missing, paid amplifies a void.
  3. But "we correctly skipped paid" can mask "we never validated anything." Skipping paid because you're pre-fit is right; staying pre-fit forever is the failure. Paid is a reward for a proven funnel, the goal is to earn the right to use it, not to avoid it permanently.

Lessons

  1. ⭐ Paid amplifies; it doesn't create. Prove the funnel converts and retains organically before spending, you cannot buy product-market fit.
  2. The whole discipline is one inequality: LTV > CAC, with margin and fast payback. Know both numbers or don't spend.
  3. Measure retention, not installs. Acquired users who churn are a loss you paid for.
  4. Start with a small budget sized so a wrong answer is cheap; scale only proven economics.
  5. CAC rises as you scale, fine at $10/day can bleed at $1,000/day.
  6. Distrust single-platform attribution; watch blended CAC.
  7. For bootstrappers, owned channels usually beat paid early, the limited budget compounds better in content than in rented traffic.
  8. Not spending on paid is often the right call, but make sure the reason isn't that you never validated anything.
  9. Paid is a reward for a proven funnel. Earn the right to use it; don't avoid it forever or reach for it too soon.

Next: Chapter 49: Metrics That Matter →

This completes Part IX, Go to market.

Useful? Share this chapter