Marketers are never starved of ideas. There's always something you can do. There's always some new thing you can try.
This is becoming more and more true with the rise of no-code.
Before, your crazy ideas might need to be vetted by someone in engineering before it becomes reality.
Now, sometimes you can just spin something up in Zapier or go straight into Webflow or HubSpot and make a bunch of changes yourself without being forced to have any rigor in your thinking.
Net-net, this is a good thing: less friction tends to lead to better outcomes. But less friction also usually leads to more waste.
Before you act on your hunch, or get your team pumped up about some "test" you want to try, introduce a bit of productive friction by asking these 3 questions:
1. What does impact on the business look like?
Connect the dots to how this will improve the overall business. It doesn't always have to tie directly back to revenue, but you should at least be able to explain the relationship to business outcomes that matter.
2. What are we going to learn?
If you are planning to run a test, write down what you expect to learn beforehand. Take a shot at some hypotheses. Take a shot at predicting the results. Don't just blindly turn on paid social ads "to see what happens" because you're going to end up not knowing how to interpret the results fairly.
3. Why would anyone outside of marketing give a shit?
This is an exercise in cross-team communication as much as it is a forcing function for working on valuable things.
To have a good answer for this one – in addition to working on something worthwhile – it requires you to (a) have good relationships with stakeholders that matter and (b) know what they care about so you can communicate your activities in a way they can appreciate.
Sometimes this happens organically in a startup, where you are 1 level away from the CEO or are on the c-suite yourself. Sometimes you have to engineer this, like when you are working at a huge company. (See also: The Hourglass Framework)
If you or your team don't have clear answers for these, that's a red flag that you're about to waste a bunch of time, and you should move on to the next idea.