Payment for services, core agitators, color spaces
I can’t believe that I felt motivated to write this in 2024, especially after the kind of year we had, but here we are.
Draft is a business. (I know, right?) Like other businesses, we sell things that we believe to possess intrinsic economic value. And we’ve made enough money off our work over the past 12 years to think that we might be onto something, especially considering all the money that we’ve made others in the process.
In design as in the rest of life, you get what you pay for. That’s the whole point of design as an investment, one that is provably low-risk when executed appropriately.
In short, if you want a free unresearched “audit” that will make you feel good about yourself while hurting your business, I’m sure there are plenty of people out there who would be happy to oblige. If you want something that will reliably make back your investment because it’s grounded in real-world evidence, you know where to go.
Only two more weeks until our Value-Based Design Workshop, and we’d love to have you join us. Hit reply if you have any questions!
This week, for paid members
- This week’s paid lesson shows you how to define the main problem that motivated someone to hire your product. It’s JTBD at its core!
- Our design of the week shows one of the less-great ways to provide a discount code for a sale – and what you should do instead.
- And our fortnightly teardown is for pillow brand Pluto. How do they create bespoke products for a diverse array of needs, and what can they do better?
Want in? Join us now – named one of the best ecommerce communities going on the web.
Already a member? Log in here and take a look at what’s new.
Links
- We don’t do a whole lot of tree testing at Draft, but if you want to know the difference between it and card sorting, NN/g has you covered.
- Conversion has begun testing prioritization & confidence intervals using machine learning. In one sense, machine learning for experimentation is nothing new, since it has driven most of the “black box” confidence calculations that we’ve ignored in favor of actual statistics for over a decade. But using machine learning for prioritization is different, and novel, and their claims that it may reliably beat the overall win rate are very curious. Two questions immediately come to mind. First, are there any trends that we can back out of such a lift across all industries, such that we can learn ourselves as value-based designers? And second, most people don’t really know how to prioritize, so does higher-quality value-based execution tend to reliably beat this model? Related.
- How to create icons. Related. Also related.
- Filter clarity.
- I actually had no idea how to visualize color spaces before this post, and now I think my conception of the idea is more on the “less bad” front. I still don’t know how a different color space impacts the rendering of an image, or what color spaces to favor for what kinds of work. I possess perfect color pitch in the real world, so this definitely matters to me as an alive being! It’s sort of wild to me that something as simple as color generates this much mathematical effort, with so many conflicting models.
- Univers. Frutiger is nicer, fight me.