You raised your pre-seed, you have $5,000 to spend on marketing this quarter, and you have no idea where to put it. Welcome to the club. Founders waste an enormous amount of capital on this exact decision — either spreading it thin across ten channels they read about on Twitter, or going all-in on the trendy channel of the month.

This is the framework we use with every startup client. It works because it forces concentration, builds in learning budget, and protects you from the most expensive mistake in marketing: quitting a channel two weeks before it would have started working.

The 70/20/10 rule

Borrowed from Google and refined for startups, the 70/20/10 split forces concentration on what works while keeping a small budget for exploration.

70% on validated channels — channels proven to work for your category. Meta Ads for DTC. Google Ads for high-intent B2B. LinkedIn for enterprise SaaS. Don’t guess — look at competitors and ask 5 customers where they discovered similar products.

20% on emerging channels — channels likely to work but unverified for your specific business. Maybe TikTok organic, maybe Reddit Ads, maybe a niche newsletter sponsorship. Enough budget to give a fair test, not enough to bet the company.

10% on experimental channels — long shots with asymmetric upside. Affiliate partnerships, podcast sponsorships, community building. Most experiments fail; the ones that work redefine your stack.

A $5K example allocation, line by line

$3,500 — Validated channel (Meta or Google Ads)

Pick whichever matches your customer’s intent. Selling something they actively search for (legal services, plumbers, B2B tools)? Google. Selling something new they don’t know exists yet (DTC product, novel SaaS, course)? Meta.

Allocate:

• $2,500 — actual ad spend (delivered to Meta or Google)
• $1,000 — creative production (3-5 hooks × 3 variations + UGC if Meta; RSA copywriting + landing page polish if Google)

$1,000 — SEO foundation

Audit + on-page optimization + 2 cornerstone blog posts targeting your buyer’s top search terms. This won’t move the needle this quarter — SEO traffic shows up around month 4-6. But you’re building an asset that compounds long after this $1,000 is gone.

$500 — Experimentation

Pick ONE emerging channel for this quarter. Not three. One. Options:

A different lens: time-to-revenue by channel

This is why most startups should start with paid channels (fast feedback) and layer SEO + content underneath (slow compounding). Pick ONE paid channel that matches your customer intent, get it profitable, THEN expand.

The full first-90-days flow

Common $5K marketing mistakes

• Spreading too thin. $500 spread across 10 channels learns nothing on any of them. You need enough budget per channel to hit statistical significance — typically 50+ conversions before judging.
• Cutting too soon. Most channels need 14-21 days to exit Meta’s/Google’s learning phase. Killing on day 5 means you paid for the learning and got none of the optimization.
• No tracking infrastructure. If your Pixel/CAPI isn’t firing, your conversion tracking is broken, or you don’t have UTM hygiene, you’re flying blind. Track this in week 1 before you spend a dollar.
• Optimizing the wrong metric. Clicks aren’t leads. Leads aren’t customers. Customers without LTV math aren’t a business. Always track to revenue, not engagement.
• Skipping creative budget. Most startup ads fail because the creative is bad, not because the algorithm is broken. Budget at least 25-30% of spend for production.
• Trusting an agency without weekly reports. If you can’t see exactly what was done, what worked, and what changed each week, you’re paying for activity not outcomes.

When to expand from 1 channel to 2

Don’t expand until your first channel is profitable — meaning CAC is below LTV with a healthy buffer. Specifically:

What about budgets above $5K?

Same percentages, bigger numbers. $25K/quarter? 70% on validated ($17.5K), 20% emerging ($5K), 10% experimental ($2.5K). $250K/quarter? Same math. The discipline is the constraint, not the dollars. Founders who lose $25K because they didn’t learn from $5K will lose $250K too.

Concentration wins the early game. Pour your first $5K into one channel hard enough that you actually learn something, then scale or kill based on data.