(Web) Analytics Complete Course
0. Create a website
- wordpress
- github pages
1. Implement Web analytics on website
Can be Google Analytics 4.
Alternatively a self hosted Web analytics solution such as Matomo, Fathom or something similar.
1b. OPTIONAL - extend implementation skills
- add button tracking
- add conversion tracking
- use a tag manager
2. Debugging Web analytics
- browser developer tools
- browser plugins (omnibug)
- wireshark
3. Run basic reports in Web analytics interface.
4. Directly query collected data.
If GA4 export to bigquery and query
If self hosted get access to database and query.
5. Create reporting
- looker studio
- self hosted eg redash, superset
In data analytics there are 3 stages
1. Pre work
2. The work
3. Post work
Almost all content produced is about #2 - the work.
While this is critical, without being competent in the other areas you will severely limit your impact and enjoyment of working in data analytics.
What is pre work?
Pre work includes things like requirements gathering, prioritisation, and expectation management.
It may even include a rough plan on how to approach the particular project or piece of work.
This is important because it allows you to fundamentally understand the problem, meaning you can work around any blockers that arise end stilled up at the desired place.
It also allows you to apply your effort to the places that you think have the biggest impact.
All while keeping your business partners up to date with timelines and progress
What is the work?
This is everything hands on tools collecting and slicing and dicing data to creating visualisations.
Excel, SQL, GA4, BigQuery, Snowflake, pandas, tidyverse, duckdb, powerbi, D3, spark etc are all included in the work.
The process of turning data into information and wisdom.
SQL is the virus of the data world.
Multiple attempts to eradicate it have failed.
Not only does SQL have a strong foothold in the land of databases, SQL has slowly been infiltrating other platforms and processes.
SQL queries can be applied to local csv files using sqlite-utils.
Got lots of csv files on a cloud storage provider. DuckDB allows you to query it with SQL.
Already got a pandas dataframe? Manipulate it with SQL using DuckDB.
Data inside a Google sheet can also have SQL applied to it using the Query function.
Wherever data is SQL e eventually makes its way to it
Oddbean new post about login | logout
Notes by 497dd658 | export