The Stack You Actually Need (Day One)
I spent two years building marketing from the ground up at a fintech startup. We had zero budget, one marketer (me), and a founder who expected results. This forced me to get scrappy. Here's what I learned: a tight, interlinked stack beats a sprawling platform every time.
You need four things. Not ten. Four.
The Tools That Actually Work at Pre-PMF Scale
Email + Automation: Brevo or Mailchimp
Forget the "email marketing vs. automation platform" debate. At your stage, they're the same thing. Brevo (formerly Sendinblue) and Mailchimp both let you send campaigns and set up basic workflows. Brevo edges ahead because it's more automation-friendly and integrates better with landing pages.
Cost: Brevo is free up to 300 contacts and unlimited emails. Mailchimp is free up to 500 contacts. Both are genuinely free, not freemium bait-and-switch.
CRM: Notion or Airtable (Not a "Proper" CRM Yet)
You don't need Salesforce. Seriously. Build your CRM in Notion or Airtable. I know this sounds weird, but here's why: at pre-PMF, your sales process isn't defined yet. You're still figuring out what stage matters. A flexible database beats rigid software every time.
I tracked leads in a Notion table with simple fields: name, email, company, source, stage (interested / trial / waiting for feature), last contact, notes. That's it. That's 90% of what enterprise CRMs do for startups, except you control the workflow.
Switch to Salesforce or HubSpot later, when you actually have a defined sales process and want automation at scale.
Analytics: Google Analytics 4 + Hotjar
GA4 is free and tells you where people come from, what pages they hit, what they do, and where they drop off. That's what matters. Hotjar's free tier gives you session recordings and heatmaps — see exactly how people use your product.
Don't over-instrument early. You don't need Segment or a data warehouse. You need to know: traffic source, top pages, conversion funnel, where people abandon. GA4 and Hotjar cover this.
Landing Pages: Webflow or Carrd
Don't build landing pages in your main product. Webflow is overkill design-wise but solid if you know what you're doing. Carrd is simpler — dirt cheap, quick to set up, good for lead magnets and waitlists. Use it.
What to Automate First (and When)
This is where most startups mess up. They automate too early, too much, and send robotic email sequences to people who've never heard of them.
Automate These (Right Now)
Welcome email: Someone signs up for your email list or requests a demo. Send them a welcome message within 5 minutes. Warm, personal, sets expectations. This is standard and it works.
Lead scoring (simple version): If someone clicks a link in your email 3+ times, they're engaged. Move them to a "hot leads" list. That's not complex AI — it's common sense.
Abandoned signup follow-up: Someone started signing up but didn't finish. Send them one follow-up email 24 hours later. Don't spam. One shot.
Weekly reporting: Automate a report that lands in your Slack channel every Monday: new leads, trial signups, activation rate, churn. One email, set it and forget it.
Don't Automate These (Yet)
Complex nurture sequences: "If person in tech AND company > 50 employees AND opened email AND clicked link, then send X else send Y." This is premature. At pre-PMF, you don't have enough data. You'll just send irrelevant emails.
Sales outreach: Cold email sequences, personalized drip campaigns. Don't automate this. Sales needs to be human. If you're pre-PMF, you should be manually reaching out anyway to understand what resonates.
Content distribution: Automating your blog posts to 50 different channels feels productive. It's not. Write good content, share it manually to the 2-3 channels where your audience actually is.
The Real Mistake: Automating Before You Know Your Funnel
You can't automate what you don't understand. Before you set up sequences, map your funnel: How do people discover you? What's the first email they see? When does someone become a lead? When are they ready to talk to sales? What causes churn?
Get these answers manually first. Talk to 10 customers. Then automate.
When to Upgrade From This Stack
Keep this setup until one of these is true:
- You have consistent product-market fit (CAC < 3x LTV, churn is predictable, net revenue is growing month-over-month).
- You have 500+ contacts and your Notion CRM is creaking.
- You're running enough volume that manual lead routing is eating 20+ hours a week.
- You need integrations that Brevo/Mailchimp don't support natively (e.g. your custom app, a specific payment processor).
Then move to HubSpot Professional or Klaviyo (for ecommerce). By then you'll know what you actually need.
AI-Assisted Automation (The New Frontier)
LLMs are starting to make automation smarter without complexity. Email subject line generation, identifying engaged leads from email opens, personalised follow-up suggestions — this is early but real.
Platforms like Brevo are integrating AI-driven copy suggestions. Don't rely on them for everything, but for a startup marketer flying solo, having a suggestion engine beats staring at a blank screen.
The rule: let AI suggest, you decide. Never auto-send without human review, especially early on.
My Bias: Build in Public
Some of your best automation comes from just being honest about your build. "Sign up to watch how we launch this feature." "Let me know what you'd pay for this." These are higher-intent than any email sequence because people chose to be part of the journey.
The Real Test
If your marketing stack takes more than 4 hours a week to maintain, it's too complex. If you're spending more time configuring automation than talking to customers, you've gone wrong.
The best stack is the one that disappears. It works, you don't think about it, you get the data you need to make decisions. At your stage, simple and manual beats complex and automated.
Get there first. Scale later.
Need help setting up your marketing engine?
I can audit your current stack and show you exactly what to keep, what to cut, and what to automate first. Let's build something lean and actually useful.
Book a Call