← All articles 30-DAY PLAYBOOK · #6

The complete 30-day playbook

Every tool, every script, every decision. The full retrospective on the 30-day zero-to-revenue playbook, what worked, what didn't, what I'd do differently, and the consolidated stack of skills the series has linked to so you have one page to bookmark.

Over 30 days you can move from no business to first paying customer with a small set of clear decisions: pick a real problem in week one, validate it with 5 to 10 conversations and a pre-sell in week two, build the operational stack (CRM, landing page, payment flow, email sequences) in week three, and start the inbound engines (programmatic SEO, AI Overview citations, short video) in week four. The total infrastructure cost is around $13 a month. The decisions that matter most are not technical, they're about who you choose to serve and when you ask them for money.

The four weeks at a glance

This series has walked through five core articles, each covering a phase of the 30-day playbook. The shape is roughly:

Week 1 Set up + listen

Getting set up and starting to listen. Cursor and Claude Code installed, the first folder created, the first prompt written. Then the first conversations with potential buyers: not selling yet, just understanding the problem.

And the why post that anchors the whole series sits underneath everything: this isn't a hustle culture pitch. It's the playbook I've actually used across five businesses, sharpened by AI tools that didn't exist three years ago.

What worked: the load-bearing moves

A few things that, looking back, were genuinely load-bearing.

Starting with conversations, not code. The businesses I've built that worked all had a discovery week before the build week. Five real conversations with people who had the problem produced a sharper brief than any amount of solo thinking. The build, when it came, was much faster because I wasn't guessing about what to make.

Pre-selling before building. The first paid dollar is the single highest-signal moment in any business. It separates 'people are being polite about my idea' from 'people will move money for this'. Every business of mine that took off had a paying customer before the product was fully built. Every business that stalled had a polished product and no customers.

One folder per business. Keeping each business as a self-contained folder on my laptop made it easy to context-switch, easy to back up, and easy to walk away from when an idea didn't work. The discipline is also philosophical: the business is the folder, not a tangle of SaaS subscriptions held together by Zapier. You own the whole thing.

Letting the AI write the boring parts. The first few times you ask Claude Code to write a CRM or an email sequence, it feels like cheating. After a month it just feels normal. The cumulative time saved (probably 15 hours a week vs how I worked two years ago) goes back into customer conversations and shipping the next version. That's the actual unlock.

What didn't work: the mistakes I'd skip next time

And a few things that didn't, that I'll do differently next time.

Trying to ship too much in week three. The temptation, once the validation came back positive, was to build everything at once. The CRM, the landing page, the lead pipeline, the email sequences, the proposal flow, the booking system, all in one week. The result was that nothing was quite finished and the first customer experience was rougher than it needed to be. The lesson: ship the minimum needed to serve the first customer well, then iterate.

Skipping the devil's advocate. The one time I didn't pressure-test the idea against the strongest skeptical case, I ended up six weeks into a build for a market segment that turned out to be too small. The exercise feels uncomfortable. It's also the cheapest insurance against the most expensive kind of mistake.

Not picking a single inbound channel to start. I tried to launch the SEO engine, the video clips, and the social presence all in the same week. The work spread too thinly across all three to make any of them work. Starting with one channel (programmatic SEO for B2B, short video for B2C), getting it producing, then layering the next, would have moved faster.

The three judgment calls that decide everything

Underneath the playbook, there are really three judgment calls that decide whether the 30 days work. The rest is execution.

Who do you choose to serve? The single most important decision. Pick a buyer whose problem you understand, who has money to spend, and who you'd be happy to talk to weekly for the next year. Get this wrong and no amount of execution polish makes up for it.

What's the smallest version of the offer that proves the problem? Not the smallest version that makes you proud. The smallest version that gets a yes or a no from a real buyer. Almost always smaller than your first instinct.

When do you ask for money? Sooner than feels comfortable. The fear of the ask is the most common point of failure. The first ask is the hardest. By the fifth it stops feeling like a big deal.

Everything else (the tools, the systems, the AI workflows) is downstream of those three calls. Get the calls right and the playbook works. Get them wrong and the best tooling in the world won't save it.

The consolidated stack of skills

Every AI skill the series has linked to, on one page, organised by the phase they show up in. The last three (LinkedIn Carousel Design, Product Brief Writer, Stakeholder Simulator) are the cross-cutting ones that didn't fit cleanly into a single week but show up often enough across the 30 days that they're worth having ready.

Skill What it does
JTBD Extractor Turn raw customer input into Jobs-to-be-Done statements
User Interview Analyzer Pull themes from call transcripts
Research Synthesis Engine Structure findings into a build-or-pivot doc
Opportunity Solution Tree Map outcomes to solutions and find weak links
Persona Generator Build working buyer profiles from real evidence
Devil's Advocate Stress-test the case against your idea
Experiment Designer Write a one-page test plan with clear pass/fail
Signal-Based LinkedIn Outreach Find buyers who just gave off a relevant signal
AI Website Cloning Spin up a clean landing page in an afternoon
SaaS Stack Scanner Audit what you're paying for and what to cut
Lead-Gen Apify Scrape leads from LinkedIn, Google Maps and more
Lead Generation End-to-end pipeline: scrape, enrich, score, queue
PRD Generator Turn an idea into a proper product spec the AI can build from
Technical Spec Writer Write the architecture brief for the harder work
Roadmap Builder Sequence the work so it actually ships
Prioritization Engine Score candidate features by impact and effort
Programmatic SEO Generate hundreds of pages from one template
Answer Engine Optimization Win citations in Google's AI Overviews
Local Clip Engine Turn one long video into five ready-to-post clips
Feedback Decoder Turn raw customer feedback into a content brief
LinkedIn Carousel Design Ship branded carousels without Adobe
Product Brief Writer Frame a feature so collaborators ship the right thing
Stakeholder Simulator Pressure-test a pitch against your toughest stakeholder

What I think happens next

Predictions are mostly wrong, so take this with the usual salt. A few things I think are likely over the next 12 months for anyone running this playbook:

The cost of the build keeps falling. The same AI tools that make a CRM a weekend project today will probably make a working e-commerce store, a working analytics platform, and a working booking marketplace weekend projects within the year. The skills the series covers won't go stale. The complexity ceiling of what one person can ship will rise.

Distribution will get harder, not easier. As more solo founders ship, the noise floor on every channel rises. The advantage shifts from 'who can build it' to 'who can position it'. Taste and clear writing get more valuable, not less. The discipline of the validation loop (knowing exactly who your buyer is and what problem you solve) becomes the difference between getting found and disappearing in the noise.

Owned audiences will matter more than rented ones. Algorithm changes have eaten more than a few businesses I've watched up close. Email lists, customer databases, and direct relationships keep their value through every shift. Build the inbound surfaces, but capture the email or the contact every time, so when the algorithm changes you still have a way to reach the people you've earned.

The part the mechanics don't teach

This series has been deliberately practical. The mechanics, the tools, the order to do things in. But the part I want to leave you with is the part that none of the mechanics teach.

Every business I've started has been a small act of faith. I didn't have a plan. I had a hunch, a real problem in front of me, and a willingness to take the first step. The plan caught up later. The systems caught up later. The skills caught up later. The first step was always smaller than I thought it needed to be, and the first dollar always came from somewhere I hadn't predicted.

If you've read this far, you probably have a hunch sitting in the back of your mind. A problem you've seen up close. A buyer you understand. A small offer you could write tonight. The 30-day playbook is just a structured way to give that hunch a chance to either prove itself or get out of your head. Either outcome is useful. Both make the next idea easier.

The hardest part isn't the playbook. It's deciding that the hunch deserves the 30 days. Once that decision is made, the rest is just execution, and execution is the easy part now.

Common questions

Two things. First, I'd spend the first three days on customer research instead of jumping into building. Every business of mine that struggled was one where I built first and validated second. The discipline of talking to five people before opening Cursor would have saved me months. Second, I'd take the pre-sell ask sooner. The fear of asking for money before the product is ready is the most common procrastination disguised as professionalism. The first dollar is the only validation that doesn't lie.
Building before they've talked to anyone. The reason this mistake is so common is that building feels productive and talking to strangers feels uncomfortable. The order matters more than the work. A bad product with the right buyer earns money. A great product with the wrong buyer earns nothing. Spend the first week on conversations even if it feels slower. The product timeline catches up fast once you know what you're actually building.
Around $230 if you're starting from scratch, almost all of which is the Claude Code subscription. The infrastructure is $13: $3 for Neon Postgres, $10 for Resend email. Cursor is $20 if you go straight to Pro. Claude Code Max is $200. There's no SaaS stack on top of that, no agency fees, no developer contracts. Compared to the typical 'launch a business' starter cost (a website builder, a CRM, an email tool, plus a few hundred for a logo and a designer), the cost has roughly halved. Compared to the all-in cost of a venture-backed startup doing the same thing, you're in another universe.
Two structural moves. First, write down the three things you'll ship next month before this month ends. Decide while the energy is high, not when you're tired in week six. Second, automate the parts of your week that don't need your judgment: the lead scraping, the email follow-ups, the customer onboarding, the invoice sending. Every hour you free up from repetitive work goes into the higher-value work of talking to customers and shipping the next thing. Momentum is mostly about removing the friction that makes the next move feel hard.
That's a real possibility, and it's not the end of anything. The 30-day window is a forcing function for honesty, not a contract. If money hasn't moved, you're getting information: the problem you picked isn't painful enough yet, or you're talking to the wrong people, or the offer is shaped wrong. Take the lesson. Keep the systems you built (the CRM, the landing page, the email setup, the lead pipeline are all reusable). Point them at the next idea. The second 30-day loop is faster than the first, because the infrastructure is done and you've already lived through the awkwardness of asking people for money.
A paid call. Almost any business can be tested with 'I'm offering 20-minute strategy calls on X for $50, first three slots'. It doesn't need a website. It doesn't need a brand. It needs a one-paragraph description, a Stripe link, and three messages sent to people who have the problem. If anyone pays for the call, you've proven the problem is real to at least one buyer. From that one call you get language, objections, and the next person to talk to. The first dollar is much smaller than people think.
Set a hard rule: every Sunday, ask 'what did I ship that a customer saw or paid for this week?' If the answer is nothing, you spent the week on internal tooling and you owe yourself a course correction. Internal tools matter, but they should never crowd out customer-facing work for more than a week. The healthiest founders I know alternate weeks: one week heads-down building or fixing internal stuff, the next week entirely customer-facing. The rhythm keeps both sides moving.

Want to run the 30 days with a room of other founders?

The SoloStack workshop is a 1.5-day live build. You walk in with an idea and walk out with the CRM, the landing page, the payment flow, and the boilerplate, plus a small private community of people running the same playbook alongside you.

See the workshop →