Yeah browser notifications are really terrible; and especially when there's already been so much thought and work done on mobile notifications... lots of easy lessons to be learned.

However web "notifications" (think pop-ups / modals / product tours / banners) are also still pretty bad and can learn a lot from best practices around how to engage users around notifications; which is the content of this article.

Yes the goal of a product IS to solve a need. But are you suggesting that all habit-building products should be avoided? Games??

Andrew, totally agree that as a goal that's useless! Your goal should be to create value. However this needs to be broken-down, and a lot of helping people do things THEY want to do involves creating good choice architecture and helping them help themselves.

That's where some of this thinking comes from. It's not about fooling people into being hooked onto something irrelevant, but how to create an engaging product that's enjoyable.

Do you believe it's not acceptable to use psychology in creating engaging design?

Yup for anyone who needs help on finding it, go here

The strategy adopted by InVision here is pretty interesting from a business perspective: provide a good integration with Sketch, the main used tool in this space, until they were comfortable launching their own version.

Wonder what the dynamics will be for the Sketch relationship now that they're competitive. Also wonder if Sketch might have considered going into the prototyping space!

TL/DR: great user onboarding is more than good design. Our framework for success consists of:

1. Making a single person / team responsible and accountable
2. Being aware of some basic user psychology (BJ Fogg model)
3. Figuring out the optimal path to Aha
4. Using all channels (email, chat, product tours, docs) to guide and teach
5. Analyzing and iterating continually

One of the key things to remember is that a product's success isn't down to just design.

BJ Fogg explains that behaviors occur when a person is motivated, has ability and receives a prompts. To apply that to products, consider motivation ~ value proposition; ability ~ interface; and prompts ~ triggers.

The relationship between these is akin to an inverse exponential.. so as the value proposition (Y axis) gets stronger, then worse interface (X axis) still generates action. Conversely when there is clear / intuitive interface but low value prop then action doesn't occur.

I think Pokémon nailed the value proposition and and some aspects of interface (making AR accessible) which is a large component to their success (notice they use few prompts).

There are a lot of very active Slack communities where you can post questions and have conversation. I find that this leads to much more engaged and high-quality feedback because it's less transactional than a review site.

Here's a pretty decent list

Great job with this. What do you think is the half-life of these trends? Which are fleeting and which will stick around for 2-3 years?

This is a great primer, thanks! As we transition to AI there is a lot of pseudo-AI being implemented, and that's a good place to start.

This is effectively responsiveness. Facebook's newsfeed many years ago was pseudo-AI because it responded to whether people liked or shared a post. Now it's true AI because it automatically learns what type of content works for each user and presents that.

Within products, designers and PMs can begin by introducing more reactiveness. For example this could be to show contextual guidance, or only to show the most basic features and reveal more as user discover.

There are also now tools to help build stuff like this so worth exploring.