I think the naming might have yet another effect: Customers now feel an association to one of the tiers. For a studio customer to buy the 'freelancer' tier might feel unprofessional / dishonoring. I'd like to know how much effect just the naming alone has vs. the price / feature changes.
I had more people signing up for the higher tier plans when I a/b tested the plan names. Don't remember what the numbers were exactly but it was interesting to see behavior change because of the plan names.
Have you watched Dan Ariely's TED talk on irrational decision making? Your revised pricing tiers reminded me of the discussion about tom, jerry, and "slightly ugly" jerry around the 15:00 mark. If you haven't seen it, it's interesting that you've stumbled on the effect independently.
You might find that further tweaks push more people to the agency tier by creating a "slightly ugly" agency instead of a "slightly ugly" studio -- freelancer.
I've wanted to a/b test my plan name but don't know of any good alternatives. I have a free WordPress plugin and have a paid-for version which adds more features so it's called generically 'Premium version'.
The people who buy it come from different backgrounds and I don't want to segment the product.
It's a recipe plugin and the premium version adds a nutrition tab to satisfy the nutritionists and category listing pages for people who want to organise their recipes more than what the free version allows.
Don't take it the wrong way, but you're trying to squeeze your users into the rigid frame of your preconceived notions about who they are. Changing people is a very difficult job. I understand you put a lot of effort into your product and feel entitled to steer it as you see fit, but it's far more productive to do what the topic post suggests - ask the users who they are and what they care about. In other words, reconsider your decision not to segment.
mmh how about "Hobby" and "Passion"? I got nothing. To be honest I'm not sure whether those features are a good way to diversify, but then again I probably don't know enough about your product.