Alteryx Academy is the free training arm of Alteryx, a drag and drop analytics platform that had a strong run in enterprise data preparation and is now under pressure from several directions at once. The training itself is well produced. Interactive lessons, a structured path to the Designer Core certification, and the Weekly Challenge series, which is the best thing here by some distance. The Weekly Challenges deserve credit.
Each is a small real problem with a dataset and a required output, and you build a workflow to get there. Solutions are posted and discussed, so you can compare your approach against others and see the idioms experienced users reach for. That format, small problems with published solutions and community discussion, is a genuinely effective way to learn a tool, and quite a few paid platforms do not offer anything as good. The certifications are free, which is unusual and worth acknowledging.
Designer Core is achievable with a few weeks of focused work and Designer Advanced is a real step up. Alteryx made a decision to give away certification rather than monetise it, and the effect is that the credential is widely held. That is the problem with it as a signal, of course. When a certification is free and passable in a fortnight, it tells a hiring manager very little beyond that you have used the product.
The skills you learn are legitimately transferable at the conceptual level. Joins, unions, filters, aggregation, parsing, reshaping between wide and long, handling nulls, building a repeatable pipeline rather than doing it by hand each month. Those are data preparation fundamentals and understanding them properly will help you in pandas or SQL or dbt. Someone who learns to think in Alteryx workflows has learned to think about data transformation as a sequence of operations, and that is worth something.
Here is my actual concern, and I do not think it is unfair. Alteryx is a proprietary tool with a licence cost that puts it well out of reach of individual learners, and it is competing against Python with pandas, against SQL running in a warehouse, against dbt, and against cloud native pipelines, all of which are cheaper, more flexible and more widely demanded. The trend in analytics job postings has moved towards code for years. Investing forty hours in becoming good at Alteryx is a reasonable choice if your employer has already bought it and you need to be productive on it.
As a strategic bet on your own employability, it is a weak one, and the same forty hours in Python and SQL would leave you in a much stronger position. The drag and drop interface has a subtler cost. It is approachable, which is the selling point, and it lets people who cannot program build real pipelines, which is a genuine democratisation. It also lets people avoid learning to program for years while feeling productive.
I have met analysts who were skilled Alteryx users and could not write a for loop, and when their organisation moved off the platform they were in a difficult position. If you use Alteryx at work, learn Python alongside it rather than instead of it. The AI and machine learning modules are the weakest part. The Intelligence Suite and the assisted modelling features let you fit models without understanding them, which produces results faster and understanding slower.
What you learn about model selection, validation and interpretation through a guided interface is thin compared with doing it in scikit-learn where the choices are visible. If machine learning is the goal, this is not the route. My three point three is for genuinely good free training and an excellent practice format attached to a platform whose trajectory I would not bet on. Take the certification if it is free and relevant to your current job.
Do not build a career plan on it.