Purism vs. pragmatism, add-to-cart behavior, holiday testing
This should be the first letter you’re getting from our new mailing list provider. Hit reply & let us know if you’re running into any issues. Thanks!
I was lucky enough to grab a meal with Joel a few weeks ago. We chatted about how our businesses were going, and he did the very reasonable thing that people normally do all the time, which is to ask if I’m writing another book.
I said “no,” which at the time was true. I’ve already made three books. I still feel like there’s a lot to talk about in my latest. And I go years between books. This feels appropriate, since books are big questy things, something you should only write when you have something big and new to say.
Then I got on a 4-hour flight back to Chicago and outlined two books.
To be clear, outlines mean nothing. I write outlines like these all the time. Sometimes they turn into zines; sometimes they turn into letters or paid lessons; most of the time they turn into nothing. But I also hadn’t outlined a book in three years. The gesture means something within the practice. Something is stirring, and I know I need to create the space to explore it.
And really, I shouldn’t be surprised about this? While its principles remain evergreen, the world has utterly transformed since I released Value-Based Design. And I’ve changed, too.
On my own front:
- I know more about how to serve others’ businesses. A couple of months ago, I mentioned that I want to focus exclusively on teaching & training by the end of 2022. In light of that, I’ve spent a lot of time understanding the pains that people are facing, not only in their jobs but also in their experiences with technology.
- I’ve grown deeply suspicious of technology. This probably makes sense for someone who knows way too much about how it works! And also I’m not on social media, so I spent way less time doomscrolling during the past two years than pretty much everyone else. This has given me a perspective that I don’t think is widely shared by my peers, let alone the rest of humanity.
- If the world is transforming, creative practices must transform alongside it. The weirdest thing about the current moment is the unity of it. I have experienced considerable loss, just as many of you have. I’ve spent years reflecting, just as most of you are. Any future work should resonate with the current moment.
So I’ll probably have something to announce early next year – and in the meantime, I’m curious to hear what you’re looking for.
This week, for paid members
- Our design of the week talks a lot about state & scope in mobile navigation. What state is the navigation in, and what scope does that create in the product catalog?
After a busy week pre-BFCM, things were a little quieter than par as we embraced rest. Lots of stuff is planned for the next two weeks, though!
Want in? Join us now – now named one of the best ecommerce communities going on the web.
Links & analysis
- From WiderFunnel: the split between purism & pragmatism in experimentation practices. This resonated deeply with me, since I tend to be a bit of a methodological pragmatist, but a statistical purist. In what ways do you demand certainty in your value-based design practice, and what are you more flexible about?
- A terrific post on calculating MDE for new experiments. Note that MDE should be a factor of your sample size; you can calculate that here. More from Evan Miller.
- Just as value-based design has a clear economic impact, there’s also the impact of making bad decisions. This case study discusses multi-armed bandit experiments, which are great if you have the traffic. (Change.org has the traffic. Your tiny store does not.) If you don’t, just keep running experiments.
- How to test during BFCM promotions. I generally don’t recommend running experiments during this period, but if you make more than $10M it might be worth it.
- A framework for prioritizing research. At Draft, we follow a roadmap to make sure we’re researching the right things at the right times – but we also adapt our schedule to individual clients’ needs. This post takes a broader look on all research activities.
- If you run your own experimentation framework, you’ll dine out on this insightful post on re-randomization. It’s easy to forget about all of the math that goes into determining whether customers receive the control or a variant – and it’s very easy to forget about how complex this can get when you’re running multiple variants, or multiple experiments, across a single page in a given experience.
This month’s free lesson: What behavior & feedback should adding to cart provide?
You tap an “add to cart” button. What happens?
Beyond “the product ends up in your cart,” this should be an easy question. It is not.
What happens if you have a sidebar cart? A tooltip cart? A modal cart? What about animation transitions?
Let’s talk about some design guidelines that most stores should follow when adding a product to cart:
- Adding to cart should display the cart contents. This is a core principle of feedback, showing the customer that they did something, and reading back what they did. A popup shouldn’t simply say “thanks” and leave the customer on the product detail page.
- The cart should show your products, potential upsells, and the ability to checkout – all above the fold. This is a heavy lift for modal overlays, which is why we generally don’t recommend using them. Sidebar carts are preferable.
- Adding to cart should show the actual cart. Some stores have “in-between” pages with upsells or other calls to browse. This is a mistake. Show the cart, on a page called “cart.” (Shopify stores have
/cartas the default URL for a reason.) Upsells should be in the cart if you’re going to put them there. - Provide the full order cost in the cart. This, too, is the result of quality feedback: people should know how much they expect to pay.
- Provide prominent, persistent feedback. Adding to cart is a significant action that requires significant, prominent feedback. A single line of z-indexed text over your primary navigation that fades after 5 seconds won’t do it.
Go through your store’s add-to-cart behavior with these guidelines, and make sure that it fulfills each one. If it doesn’t, that’s a big redesign opportunity once traffic dies down in January.