Setting Up Analytics From Day One: What to Track and Why

Stop Flying Blind

You built something. People are signing up. But do you know what they actually do after they land on your site? Do you know where they come from, what feature they try first, or where they give up and leave?

Most founders put off analytics. They tell themselves they'll "add tracking later" once things get going. Then six months in, they have thousands of users and zero insight into what's working and what isn't. Every product decision becomes a guess.

The opposite problem is just as common. Some founders install every analytics tool they can find, tag every button click, and end up with dashboards full of data they never look at. More data does not mean better decisions. It usually means more noise.

The right approach is somewhere in between. A small, focused set of metrics that you check regularly and actually use to make decisions.

The Only 5 Metrics That Matter Early On

When you have fewer than 1,000 users, you don't need 50 metrics. You need five. These are the numbers that tell you whether your startup is healthy or dying:

1.Signups (or account creations). How many new people are trying your product each day or week? This is your top of funnel. If this number is flat or declining, nothing else matters.

2.Activation rate. Of the people who sign up, what percentage actually do the core thing your product exists for? For a project management tool, that might be creating their first project. For a design tool, it's completing their first design. If people sign up but never activate, your onboarding is broken.

3.Retention (Day 7 and Day 30). Are people coming back? A startup with 1,000 signups and 2% Day 30 retention has a leaky bucket. Fix retention before you pour more users in the top.

4.Revenue (or conversion rate if you have a free tier). How many users are paying, and how much? For pre-revenue startups, track how many users hit the point where they'd realistically pay.

5.Referral source. Where are your best users coming from? Not just the most users, but the ones who activate and stick around. This tells you where to double down.

That's it. Resist the urge to track everything else until these five metrics are solid and you're checking them weekly.

Choosing Your Analytics Tool

You have four realistic options, and each fits a different situation:

Google Analytics (GA4). Free, widely used, and good enough for basic traffic and source tracking. The interface is confusing, and event tracking requires some setup, but you can't beat the price. Best for: founders who just need basic traffic data and aren't ready to invest in a dedicated product analytics tool.

PostHog. Open source, self hostable, and has a generous free tier. Combines product analytics, session recordings, feature flags, and A/B testing in one tool. Best for: technical founders who want product analytics and don't mind a bit of setup. PostHog is what we use at PostYourStartup.co.

Plausible. Lightweight, privacy focused, and incredibly simple. No cookie banners required because it doesn't use cookies. Best for: founders who want clean traffic analytics without the complexity of GA4, or those targeting privacy conscious users.

Mixpanel. Powerful event based analytics with strong funnel and retention analysis. The free tier covers up to 20 million events per month. Best for: SaaS products where you need detailed user journey analysis and cohort breakdowns.

Pick one and set it up today. The tool matters far less than actually having data flowing in. You can always switch later.

Setting Up Event Tracking That Actually Helps

Page views alone tell you almost nothing useful. You need event tracking to understand what users do inside your product.

Here's the minimum set of events every startup should track:

Signed up. The moment a user creates an account. Include the referral source as a property so you can see which channels produce real users.

Activated. When a user completes the core action that defines your product's value. Define this carefully. It should be the "aha moment" where a user gets what your product does.

Key feature used. Track usage of your 2 to 3 most important features. Not every button click. Just the actions that correlate with users who stick around.

Upgrade started / Payment completed. If you have a paid tier, track both the intent to upgrade and the actual completion. The gap between these two numbers tells you where your checkout flow is losing people.

Churned. When a user cancels, downgrades, or goes inactive (no login in 14 or 30 days, depending on your product's natural usage frequency).

When naming events, use a consistent format. Something like `user_signed_up`, `project_created`, `payment_completed`. Pick a convention and stick with it. Inconsistent naming makes your data messy fast, and cleaning it up later is painful.

Vanity Metrics Will Mislead You

Pageviews, total registered users, and social media followers feel good. They go up and to the right, and you can screenshot them for Twitter. But they tell you almost nothing about whether your business is working.

Here's how to tell the difference:

Vanity metric: "We have 5,000 registered users." Actionable metric: "Of 5,000 registered users, 800 activated and 200 are active weekly."

Vanity metric: "We got 10,000 pageviews this month." Actionable metric: "Our blog post on X drove 400 signups at a 4% conversion rate."

Vanity metric: "Our Twitter has 2,000 followers." Actionable metric: "Twitter drove 50 signups last month, 30 of whom activated."

The difference is context and action. A good metric tells you what to do next. A vanity metric just tells you a number went up.

When you catch yourself celebrating a vanity metric, ask: "What decision would I make differently based on this number?" If the answer is "none," stop tracking it.

Cohort Analysis: Your Secret Weapon

Cohort analysis means grouping users by when they signed up and comparing their behavior over time. It's the single most useful analytical technique for early stage startups, and most founders don't use it.

Here's why it matters. Say your overall Day 30 retention is 15%. That sounds bad. But when you break it down by cohort, you might discover that users who signed up in January had 8% retention, February had 12%, and March had 22%. Your product is getting better, and the overall average is being dragged down by older cohorts.

Without cohort analysis, you'd look at that 15% and panic. With it, you can see the trend is positive and keep doing what's working.

Most analytics tools support cohort analysis out of the box. PostHog, Mixpanel, and even GA4 have cohort reports. Set up a monthly cohort view based on signup date, and track activation and retention for each group.

Check this report monthly. It's the best indicator of whether your product changes are actually improving the user experience.

Funnel Tracking: Where Users Drop Off

A funnel is a series of steps users take toward a goal. For most startups, the primary funnel looks something like:

1.Lands on website

2.Signs up

3.Completes onboarding

4.Uses core feature

5.Returns within 7 days

Tracking each step and the conversion rate between them shows you exactly where users are falling off. If 1,000 people visit your site and 100 sign up, that's a 10% conversion rate at step one. If only 30 of those 100 complete onboarding, you know onboarding is your biggest problem.

This sounds basic, but it's remarkable how many founders optimize their landing page when the real problem is their onboarding flow, or vice versa. Funnel data removes the guesswork.

Set up your primary funnel in your analytics tool on day one. Review it weekly. Focus your improvement efforts on the step with the biggest drop off, because that's where you'll get the most impact for the least effort.

Build a Dashboard You Actually Check

The best analytics setup in the world is useless if you don't look at it. And you won't look at a dashboard that takes 10 minutes to load or requires you to click through five screens.

Build one simple dashboard with your 5 key metrics. Keep it to a single screen. Here's a layout that works:

Top row: Signups this week, activation rate this week, weekly active users

Middle row: Signup to activation funnel, retention curve (Day 1, Day 7, Day 30)

Bottom row: Top referral sources, revenue or conversion rate

Make this your browser homepage. Or set a weekly calendar reminder to review it every Monday morning. The cadence matters more than the tool. A Google Sheet you check every week beats a beautiful Mixpanel dashboard you ignore.

Some founders tape their key metrics to the wall next to their monitor. Whatever it takes to keep the numbers in front of you.

Attribution: Know Where Your Best Users Come From

Not all traffic is equal. A hundred visitors from a targeted Reddit post might produce 20 signups and 10 activated users. A thousand visitors from a viral tweet might produce 50 signups and 3 activated users.

If you only track volume, you'll think Twitter is 2.5x better than Reddit. But if you track activation, Reddit users are 6x more valuable.

Here's how to set up basic attribution:

UTM parameters. Add UTM tags to every link you share. `?utm_source=reddit&utm_medium=post&utm_campaign=launch` tells your analytics tool exactly where that visitor came from. Most analytics tools can parse these automatically.

First touch vs. last touch. First touch attribution credits the channel that introduced the user. Last touch credits the channel that finally converted them. For early startups, first touch is usually more useful because you want to know what's bringing people to your door.

Track source as a user property. When someone signs up, save their referral source as a property on their user profile. This lets you filter all your metrics by source later. You can see that users from Hacker News have 25% Day 30 retention while users from paid ads have 5%.

Submitting your startup to directories like PostYourStartup.co and adding UTM parameters to each listing is a great way to see which directories send users who actually stick around.

Privacy: Do This Right From the Start

Privacy regulations are real, and retrofitting compliance is expensive. Get the basics right from day one:

Cookie consent. If you use tools that set cookies (GA4, Mixpanel), you need a cookie consent banner for EU visitors. Tools like Plausible and some PostHog configurations don't use cookies, which simplifies this.

Privacy policy. List what data you collect, what tools you use, and how users can request data deletion. Use a privacy policy generator to get started, then customize it.

Data minimization. Only collect what you need. Don't track personally identifiable information unless it's necessary for your analytics goals. Aggregate data is almost always sufficient.

Respect Do Not Track. Some analytics tools respect the browser's Do Not Track signal. Enable this if your tool supports it.

Getting privacy right early builds trust with your users and saves you from a painful cleanup later when you're bigger and under more scrutiny.

The Weekly Analytics Routine

Here's a simple routine that takes 15 minutes every Monday:

1.Check signups. Up or down from last week? Any obvious cause?

2.Check activation rate. Is the percentage of users who activate improving?

3.Check retention. Look at your most recent cohort. Better or worse than the previous one?

4.Check the funnel. Where is the biggest drop off? Has it changed?

5.Check top sources. Where are your best users coming from this week?

Write down one thing you'll change or test this week based on what you see. Maybe it's rewriting the onboarding email. Maybe it's posting more in the subreddit that's sending activated users. Maybe it's fixing that checkout page where 60% of people drop off.

One change per week, informed by data. That compounds into a dramatically better product over a few months.

Start Today, Not Tomorrow

You don't need a perfect analytics setup. You need any analytics setup. Pick one tool, install it, set up your five core events, and start collecting data.

The startups that win are the ones that make decisions based on what's actually happening, not what they assume is happening. Every week without analytics is a week of flying blind, making product decisions based on gut feeling, and spending time on things that might not matter.

Fifteen minutes of setup today gives you months of better decisions ahead. Open a new tab and do it right now.