What are the four types of data analytics?

Descriptive (what happened), diagnostic (why), predictive (what might happen), and prescriptive (what to do). Most small businesses only need the first two done honestly.

What is the difference between data analytics and data analysis?

People use them interchangeably. Analysis is the act of looking. Analytics often means the system: tracking, tools, and repeating the look.

What are data analytics? Definition, types, and how to use them without lying to yourself

Q: What are data analytics?

Data analytics is the practice of collecting, cleaning, and interpreting data so you can make a decision — what happened, why it happened, what might happen next, and what to do.

Data analytics is the boring superpower: take a pile of events and decide what to do on Tuesday. Not a dashboard for the board. A sentence that says which page made a customer.

Definition

Data analytics is collecting events, cleaning the mess, and turning them into a decision. Pageviews are not analytics. “This URL made $400 and this one made $0” is analytics.

If you cannot change behavior with the number, you built interior design. Pretty charts. No steering wheel.

The four types, in human

Descriptive: what happened. Traffic, revenue, unsubscribes. Start here. Stay here longer than the course told you to.
Diagnostic: why. That landing page converted because the ad matched the H1. That one died because you asked for a phone number.
Predictive: what might happen. Fine when you have history. Comedy when you have 40 users and a neural net.
Prescriptive: what to do. “Kill the channel. Double the swap. Change the button.” This is the point. Everything else is homework.

What to track if you ship on the internet

Unique visitors to each campaign URL.
The conversion the page promised (email, trial, purchase).
Revenue or kept subscribers, not just joins.
Where they came from, without lying to UTM parameters.

That is enough to run a funnel. Vanity: bounce rate you do not understand, “engagement,” a heat map of people rage-clicking your logo.

I use DataFast when the question is “which article made a customer.” Affiliate link. The review is here. Google Analytics can count. It is worse at the money sentence.

How people lie with analytics

Reporting sessions as humans.
Counting waitlist emails as customers.
Ignoring refunds.
Changing the definition of “active” the week the chart looks bad.
A/B testing a button color with 70 visitors and calling it science.

If you are building in public, post a number you would still defend on a bad week.

A weekly ritual that fits in twenty minutes

Pick one question. “Which source sent keeping subscribers?” Open the tool. Write one sentence. Do one thing. Close the laptop. Analytics as a hobby is how you avoid shipping Tuesday’s issue.

For a list, watch who is still there on day 14. For a product, watch who paid and did not refund. For SEO, watch which URLs get impressions with no page — then write the page. That is why this one exists.

FAQ

Do I need a data team?

No. You need unique URLs and one conversion. Hire analysts after you have a question they cannot answer in a spreadsheet.

Is Google Analytics enough?

It is enough to see traffic. It is not enough if you care which blog post made a sale. Pair it with something that sees revenue.

What’s a good conversion rate?

Depends on the traffic. Compare against yourself. Screenshots on Twitter are not a benchmark.