Surfacing CTA right in the context of the flow.

Designing the timelines of possibilities over multiple tracks of time.
Like psychohistory from Foundation, but for your company.

Visually playing with different types of selectors.

Creating Context.ai review template. We tried to incorporate their logo into the graphics to play with depth illusion. While the styles are staying flat, the overlapping of shapes gives a sense of multi-level 3D effect.

The downside of most of tracking apps are inputs. It's very hard to keep the habit of filling the data and we're building UI, that showcases big buttons, easy inputs and graphic representations.

Data visualisation is hard. There are many ways to represent data, especially if it sits across multiple axises (pun intended) of information.

When we were helping Decipad, we created multiple catchy graphics that were meant to highlight the complexity of numbers storytelling. We used dark background, hard shadows and "messy" layout to translate that feeling.

One of the internal explorations to present services.
Websites, apps, metrics respectively.

Building systems is hard. One of the exercises is to test in which case you can and should break the given system. Why all border radiuses are the same, the rest of the UI adopts different tactics to convey the meaning: from filling the background to show selected state, to creating external labels to get attention.

There are multiple visual ways to make promotional rectangle more interesting and this is good example: 1. Colour "underglow" that match icon and button 2. catchy icon 3. small angle of icon 4. play of asynchronous corners

Another play with context logo, to put it as a "paper texture" on the back. While the effect is simple, it gives a sense of depth and interest without interfering with main CTA buttons.

Some systems are impossible to design in the most simple way — more than that, they would suffer if we do. So usually, if something requires a lot of inputs and layers, we try to separate them into different "floors" and questions what's the best approach for each particular selection.

Once we were challenged to design 4 second GIF explainer of a complex data visualisation product. We were not only showed how the product works, but also included small "keystroke" tutorial, so anyone could learn the power of "=" in Decipad.

Managing long text is hard. But there are multiple tactics to make it interesting. Mostly are taken from historical ways of designing newspaper layouts: list views, factoids outside of the body paragraph, references. But digital medium allows us to go further: links and mentions, popups, interactive galleries, dynamic numbers.

One of the early explorations for "dense interface" concept. As a studio, we notice how much of information is revolving around us and simple, white, bland interfaces are not always helping to get a sense of it all. So we're challenging the opposite — how busy can we make the interface, to accommodate busy people's lives.

We're always trying to see if we can push the boundaries around UI — make the progress bar wrap a card, build an unconventional button glow, separate content and navigation on different "sheets" etc.

It's hard to come up with unique graphics for startups — new branding project are dropping every day and it always feels like you have less and less options to differentiate. But sometime the colour, the perspective, the take can just feel very right, and this isometric exploration was a big success for us.

We have several ways to challenge our own interfaces. For example: would the interface work if we would remove all rectangles and styles and would keep typography only? What kind of principle can we invent to make UI consistent?

In this case, we delivered couple of unique treatments to make the UI really focused on most important things. You can notice green colour on all interactive component, but we also had to make a decision to make presets grey and login buttons black and white. Although they break the principle, they still look correct and appropriate.