What Your Startup Should Actually Do in Its First 90 Days of Marketing

That sounds obvious. It is not. Because “doing marketing” feels productive. Posts go out. Emails get sent. A blog appears. Traffic moves. The founder sees activity and assumes progress. Six months later they have a content library nobody reads, a social presence nobody follows, an email list that does not convert, and a vague sense that marketing does not work for their company.

Marketing works. The sequence was wrong.

This plan has three phases. Each phase is 30 days. Each phase has specific deliverables. Each phase builds on the one before it. If you skip phase one and jump to phase two, phase two will not work. That is not a suggestion. That is a structural fact about how marketing systems operate.

The plan assumes you are a startup or SMB with fewer than 50 employees, a small or nonexistent marketing budget, and no dedicated marketing team. If you have all three of those things, you are the exact company this plan was written for.

Phase One: Days 1 to 30. Positioning, Website, and Tracking.

Nobody will see anything you produce during these 30 days. No posts. No campaigns. No launches. Nothing the board can point to. This is the phase that separates the companies that build engines from the companies that build content calendars and call it marketing.

Every founder wants to skip this phase. I know because I have watched them try. They want to start posting on LinkedIn by day three. They want to run ads by week two. They want to “build awareness” before they can answer the question “awareness of what, for whom?” And then they come back six months later wondering why the awareness produced nothing.

Deliverable 1: The buyer definition.

Write a document that describes your ideal buyer in specific, observable terms. Not “marketing leaders” or “small business owners.” Something like: “VP of Marketing at B2B SaaS companies with 20 to 100 employees, based in the US, who runs a team of 1 to 3 people and is responsible for pipeline generation but does not have a demand gen specialist.”

Include what this person worries about, where they look for solutions, what language they use to describe their problem, and what their buying process looks like. Talk to five existing customers to validate or correct your assumptions. If you do not have five customers, talk to five people who match the profile. These conversations will destroy at least two assumptions you were planning to build your entire marketing strategy on. That is the point.

If You Do Not Know Who Your Product Is For, It Will Not Sell covers the positioning mechanics in depth.

Deliverable 2: The one-sentence positioning statement.

Write one sentence that a stranger can read and understand what your company does, who it is for, and what outcome it produces. Not a tagline. Not a mission statement. A clear, specific, jargon-free explanation.

Bad: “We empower teams to do more with less through intelligent automation.” Good: “We help e-commerce companies reduce cart abandonment by 20 percent with automated checkout recovery emails.”

The second one names the buyer, names the outcome, and names the mechanism. A stranger reading it knows exactly what this company does. Test your positioning statement on three people outside the company. If any of them ask a clarifying question, the statement is not clear enough.

Deliverable 3: Fix the website.

The website needs to do three things in the first five seconds: tell the visitor what the company does, who it is for, and why to keep scrolling. Most startup websites fail all three because the hero section says something vague like “The Modern Platform for [Category]” and the visitor leaves without learning anything.

The Startup Website Problem Nobody Talks About walks through every element that needs to work. The short list: rewrite the hero section to match the positioning statement, replace the vague feature grid with specific outcomes, remove the logo bar if the logos are not recognizable, make the primary call to action visible and low-friction, and ensure the page loads in under 3 seconds.

If you need professional help with this, a fractional CMO can define the strategy and brief a designer in a fraction of the time it takes a founder to figure it out through trial and error.

Deliverable 4: Set up tracking.

Install Google Analytics 4 and set up at least one conversion event that tracks the action you want visitors to take. Form submission, demo request, contact page visit, or whatever the primary conversion is. Set up Google Search Console so you can see what search queries bring people to the site.

That is the entire tracking stack for phase one. Two free tools. Do not buy an analytics dashboard, a heatmap tool, or a reporting platform yet. You do not have enough data to justify any of them. More tools at this stage create more noise, not more insight.

What not to do in phase one: Do not start a blog. Do not create social media content. Do not run paid ads. Do not attend a conference. Do not hire a marketing person. I know the board wants to see activity. I know the investors are asking about go-to-market. I know the competitor just launched a podcast. None of that matters if the foundation is not set. A blog without positioning is content for nobody. Ads without a converting website are money set on fire. A marketing hire without a defined strategy becomes the wrong hire. The First Marketing Hire Is Almost Always the Wrong One explains why.

Phase Two: Days 31 to 60. Pick One Channel and Prove It.

One channel. Not three. Not five. One.

This is where most founders lose discipline. The instinct is to be everywhere because being everywhere feels like strategy. It is not. It is a hedge. And hedging with limited resources produces the same result as doing nothing, just with more effort and less sleep.

Pick one primary channel. Commit to it for 30 days. Give it enough effort to produce a real signal about whether it works for your buyer. If it works, you have a foundation to build on. If it does not, you learned that in 30 days instead of 12 months.

How to choose: Go back to the buyer definition from phase one. Where does that person look for solutions to the problem you solve?

• If they search Google for answers, your channel is SEO and content. Start with Semrush or Ahrefs to identify the search terms your buyer uses. Write content that answers those specific queries. Publish on the company blog. This channel is slow. Expect three to six months before organic traffic produces meaningful pipeline. But once it compounds, it is one of the most durable channels in marketing.

• If they are active on LinkedIn and respond to thought leadership, your channel is organic LinkedIn. Post three times per week. Each post addresses a specific problem your buyer deals with. Engage with the buyer’s content. Comment on their posts with substance, not empty agreement. Send 10 to 15 personalized messages per week to people who match the buyer profile. Use the native LinkedIn platform, not an automation tool. When Selling Becomes Noise, the Market Builds Walls covers why automation on this channel is now producing negative returns.

• If your buyers respond to direct outreach via email, your channel is targeted cold email. Use io or Instantly for contact data and sequencing. Send 10 to 15 highly personalized emails per day. Not templates with merge fields. Messages that reference something specific about the recipient’s company, product, or market. Track replies and conversations, not opens.

• If your buyers rely on referrals and introductions, your channel is partnerships and network. Identify 10 people in your network who know your ideal buyers. Ask for specific introductions. Offer value in return. Build referral relationships with complementary service providers. This channel produces the highest quality leads and the lowest volume. It works when every lead matters.

The tracking system for phase two: Build a simple pipeline tracker. A spreadsheet works. Track the number of activities per week (posts, messages, emails), the number of conversations started, the number that moved to a qualified stage, and the number that entered the pipeline. Update it every Friday. After 30 days, you will have enough data to know whether the channel is producing results.

Benchmarks to watch: If LinkedIn outreach produces fewer than 3 qualified conversations per 60 messages over 30 days, the messaging is wrong or the targeting is off. If cold email produces fewer than 2 replies per 50 sends, the message is not specific enough. If content produces zero organic conversions in 30 days, that is normal for SEO; do not abandon it based on one month of data. If partnerships produce zero introductions in 30 days, the ask is not clear enough or the relationships are too thin.

What not to do in phase two: Do not add a second channel before the first one produces signal. Do not start paid ads to “supplement” organic efforts before organic efforts have proven the message works. Do not hire a social media manager because the LinkedIn posting feels like a lot of work. It is a lot of work. That is why most founders quit after two weeks and decide LinkedIn does not work for their business. LinkedIn works fine. They quit. How Smart Marketers Use AI Without Letting It Ruin the Work covers how to use AI to reduce the labor without reducing the quality.

Phase Three: Days 61 to 90. Measure, Systematize, and Decide.

This is where the plan becomes a system and the founder becomes dangerous.

By day 61, you have 30 days of channel data. Not gut feelings. Not anecdotes. Data. And data changes the conversation from “I think marketing is working” to “I know what is producing pipeline and I know what is not.”

Deliverable 1: The honest analysis.

Open the pipeline tracker. Answer five questions with numbers, not feelings.

How many qualified conversations did the channel produce? What was the conversion rate from outreach/content to conversation? What was the cost per conversation including your time valued at your hourly rate? Did the conversations match the buyer profile from phase one? How many entered the pipeline as real opportunities?

If the numbers are positive, the channel works. Move to systematizing it. If the numbers are negative, diagnose before you abandon. Low conversation rates might mean the messaging is wrong, not the channel. Low quality leads might mean the targeting is off. No conversions from the website might mean the site is not doing its job, which sends you back to phase one deliverable 3 to fix the conversion path.

The Problem Is Not Traffic. It Is Friction, Confusion, and Bad Customer Journeys covers the diagnostic framework for identifying where the conversion path breaks.

Deliverable 2: Document the system.

Write down the marketing motion as a repeatable process. What gets created, how often, on which channel, targeting which buyer, with what message, tracked by which metrics. This is not a content calendar. It is an operating manual. It describes the exact steps that produce pipeline so the process does not depend on one person’s memory.

This document is the marketing system. It is what allows the founder to step back or hand the work to someone else without losing quality. It is what a fractional CMO or first marketing hire will use to ramp up. Without it, every transition restarts the learning process from zero.

Deliverable 3: Set CRM up properly.

If you started with HubSpot Free CRM in phase one, now is the time to configure it to reflect the system you built. Create deal stages that match your actual sales process. Set up a pipeline view that shows where every opportunity sits. Build a simple dashboard that tracks the three to five metrics that connect to revenue: pipeline generated, conversion rates by stage, average deal cycle, and customer acquisition cost.

Do not buy a separate reporting tool. HubSpot’s free reporting covers what a company at this stage needs. Adding Databox or Looker Studio is premature unless you have data from multiple channels that need centralization, which you should not have yet because you are proving one channel at a time.

Deliverable 4: The next 90 day plan.

Based on what you learned, set specific targets for the next cycle. Not “grow awareness.” Numbers. Generate 15 qualified conversations. Produce 8 pipeline opportunities. Convert 2 to revenue. Increase outreach reply rate from 4 percent to 7 percent. Reduce website bounce rate from 70 percent to 55 percent.

These numbers give the next 90 days a scoreboard. Without a scoreboard, the work drifts back toward activity for its own sake, which is how stacks bloat, content calendars fill with filler, and vanity metrics replace commercial outcomes.

What This Plan Does Not Include (and Why)

No paid advertising. Paid ads are a multiplier. A multiplier applied to an unproven message, an unconverting website, or an undefined buyer wastes money. Prove the system organically first. Once you know which message, which channel, and which conversion path produce pipeline, paid ads can accelerate it. Not before.

No brand campaign. Brand is built through consistent, specific, valuable interactions with the market over time. It is a byproduct of the system, not a separate initiative. If the system works, the brand takes care of itself. If you spend money on a brand campaign before the system works, you have a strong brand and an empty pipeline.

No marketing team. One person can run this plan. The founder plus a fractional CMO or strategic advisor is the right configuration at this stage. Building a team before the system is proven creates the same problem as building a stack before the strategy is defined: you are staffing a function that does not yet know what it produces.

No elaborate tech stack. Phase one requires Google Analytics and Google Search Console. Phase two adds one channel tool (Semrush, Apollo, or nothing if the channel is partnerships). Phase three configures the CRM. Total monthly spend through all 90 days: $0 to $129 depending on channel choice. Compare that to the $2,000+ most startups spend on tools before they can explain what marketing is supposed to accomplish.

The Plan Works Because the Sequence Works

Every failed marketing effort I have seen in 20 years of doing this follows the same pattern. The company skipped the foundation, spread too thin across channels, measured the wrong things, and compensated with more tools, more content, and more activity instead of fixing the underlying problem. They mistook motion for progress and spent months running in place.

This plan avoids that pattern by respecting the sequence. Positioning before content. One channel before five. Measurement before optimization. System before scale. Each phase produces something specific that the next phase builds on, and by day 90, the company has something most startups never build: a repeatable marketing system that produces pipeline on a budget that does not require a board meeting to approve.

I am not going to tell you this plan is easy. Phase one requires the kind of honest self assessment most founders avoid. Phase two requires the discipline to ignore every channel except one. Phase three requires the willingness to let data overrule instinct. None of that is comfortable.

But comfortable is what produced six months of activity and an empty pipeline. This plan produces an engine. And an engine is worth the discomfort.

For more on building durable marketing and growth systems, visit the Transmyt blog.