This is the second part of our article series exploring the dimensions of our CARE Check (Cognitive, Responsible, Accessible, Ethical). The last edition covered Cognitive Load. Today, we turn to the Accessible dimension.

Who Is Your Website Excluding Right Now – Without You Even Knowing?

Imagine someone wants to donate to a foundation. Not much – twenty euros, because they believe in the work or because it affects them personally. This person has a mild tremor. No diagnosis, no care assessment, just hands that sometimes don't quite cooperate.

The donation page looks clean. The buttons are slim and elegant. The person taps "Continue". Lands on "Cancel". Tries again. Misses again. On the third attempt it works – but now they're out of the flow, a little frustrated, a little uncertain. They close the page. And don't donate.

The button looked great in the UI design. Nobody made it small on purpose. And that is precisely the problem.

"This Is Fine" meme originated from the webcomic Gunshow by KC Green; strip "On Fire“, 2013

What Accessibility Really Is – and What It Isn't

Most people assume accessibility is a legally required add-on for a small minority. This misconception persists – and it is simply wrong.

According to the WHO, around 1.3 billion people worldwide live with a disability. That is approximately 15 to 16 per cent of the global population. On top of that, there are people who are temporarily impaired (broken arm, ear infection), situationally limited (glaring sunlight on a phone screen, a noisy environment) or simply getting older..

A classic example: subtitles were developed for Deaf and hard-of-hearing people. Today, around 80 per cent of all viewers switch them on – on the tube, during a lunch break, while falling asleep.

This is known as the Curb-Cut Effect: dropped kerbs were introduced for wheelchair users. Today, parents with pushchairs, cyclists, and delivery drivers rely on them every day. The same principle applies in the digital world: Accessibility is usability – for everyone.

And there are clear, regularly updated standards that define what this means in practice: WCAG – Web Content Accessibility Guidelines

Why Accessibility Barriers in Digital Products Stay Invisible

The tricky part: accessibility barriers are often simply invisible to those who aren't affected.

• Those with good eyesight don't notice when contrast is too low or text too small.
• Those without colour blindness don't realise that status indicators relying solely on red and green are unreadable for many users.
• Those working with a mouse have no idea how laborious it is to navigate a website using only a keyboard.

In practice, this means accessibility questions tend to surface only when a legal complaint is already looming or an audit has failed. And yet the landscape has been clear for some time.

Accessibility Has Been a Legal Requirement Across Europe Since June 2025 – Including for NGOs and Universities

The European Accessibility Act (EAA) (https://commission.europa.eu/strategy-and-policy/policies/justice-and-fundamental-rights/disability/european-accessibility-act-eaa_en) has been EU law since 2019 – but its requirements became enforceable on 28 June 2025. Since that date, digital products and services within scope must comply. It applies across all EU member states and requires that digital products and services meet accessibility standards – including those offered by private organisations.

For NGOs, universities, and values-driven organisations, this means: those without a prioritised roadmap are now at risk of legal action, reputational damage, and – worst of all – a loss of trust among the very donors and stakeholders whose support they depend on. If you are based in the UK, the equivalent framework is the Public Sector Bodies Accessibility Regulations (PSBAR) alongside the Equality Act 2010, which requires reasonable adjustments for disabled users. While private organisations outside the public sector are not yet subject to the same mandatory standards as in the EU, the direction of travel is clear – and best practice now means building accessibility in from the start.

The Real Cost of Poor Accessibility

Beyond legal risk, poor accessibility has a direct impact on the effectiveness of your digital channels. Here is where the costs show up:

• Conversion drops – as in our opening example: users give up before they donate, book, or apply.
• Trust erodes – when someone struggles on a poorly accessible site, they question the professionalism of the organisation behind it.
• Support requests increase – questions that an accessible product would have made unnecessary end up in your inbox.
• Reach goes untapped – not just among people with disabilities, but among everyone who is situationally limited.

A proactive approach to accessibility is not only more humane. It is also more cost-effective.

Is Poor Accessibility Preventing Successful Outcomes on Your Website?

There are hundreds of tools – the majority free – that can help you identify where accessibility issues exist in your digital products. The Web Accessibility Initiative has a curated list here: WAI Evaluation Tools

But: scanning away, generating an endless list of errors, and then frantically trying to work through them – that is how you end up with a PDF graveyard. The report sits in a drawer because nobody has the time or capacity to act on it.

A Pragmatic Start to Accessibility – Without the Overwhelm

The starting point does not need to be perfect. It just needs to be a start.

If you are engaging with accessibility for the first time, it is far more effective to understand and implement a few things properlythan to receive a full WCAG report that overwhelms – and therefore gets quietly ignored.

Understanding is key: once you grasp why something creates a barrier, you can apply that lens to future design and development decisions from the outset. This is what Shift Left means in practice – embedding accessibility early in the process, not as a last-minute check before launch.

The Accessible dimension of our CARE Check is built for exactly this. Rather than a purely technical WCAG compliance procedure, we take a UX-led look at the barriers that have real, immediate impact – most of which can be addressed without placing excessive demands on your development team. We work to WCAG 2.1 Standard, Level AA – the legally relevant standard for digital accessibility in Europe.

What you get: not an endless fault list, but a prioritised roadmap with quick wins – and the documentation you need to present to your board or stakeholders.

👉 Find out more about the CARE Check

Frequently Asked Questions About Accessibility

Does the European Accessibility Act apply to our NGO? Yes – the EAA has applied since June 2025 to private providers of digital products and services across the EU, which includes many NGOs and non-profit organisations. Whether and to what extent your organisation is affected depends on your size and the nature of your offering. A first audit will give you clarity.

What is the difference between WCAG 2.1 AA and full accessibility? WCAG 2.1 Level AA is the legally relevant standard in Europe. It covers the key requirements around perceivability, operability, understandability, and robustness. Full accessibility in an absolute sense is an ongoing process – but meeting this standard gives you the necessary, demonstrable framework.

Do accessibility overlays work as a quick fix? No. The consensus across the accessibility community is unambiguous: overlay widgets do not fix underlying code issues. They can even actively interfere with screen reader users. Sustainable accessibility requires changes directly within the product.

Where do I start if my team has no capacity? This is exactly where a pragmatic approach helps: not everything at once, but prioritised quick wins that show immediate impact – without overloading your development team.

Found this useful? The LOVE /: RAGE Newsletter is published in German monthly on LinkedIn. www.linkedin.com/newsletters/love-or-rage

It is the first of four LOVE /: RAGE issues diving deeper into the four dimensions of the CARE Check (Cognitive, Responsible, Accessible, Ethical). CARE is an ethical website check based on the DSA and BSFG, psychologically sound and practical to implement – a new service developed by BIRD UX.

Sigh, hidden picture puzzles!

Do you remember hidden picture puzzles? You could spend whole afternoons playing them! Discovering situations, imagining stories and finding something new time and time again – dreamy. But there is a time and a place for hidden picture puzzles. 

When websites or digital processes are so cluttered that they resemble hidden picture puzzles, they may well invite you to dream – but of nightmares. This is because people generally engage with digital products in order to find specific content or complete tasks. In this context of use, they are looking for guidance, not distraction, as is the case, for example, when playing hidden picture puzzles. On websites, therefore, clutter does not lead to fun, but to feeling overwhelmed. In this case, we speak of cognitive load.

What is cognitive load?

Digital products only work if they align with the way people think, perceive and make decisions. Design that respects the limits of human cognition reduces mental strain and makes complex content usable in the first place. Design that does not take the limits of human cognition into account leads to an overload of memory capacity (cognitive load), which can result in confusion, errors and, ultimately, the abandonment of processes. 

However, this overload is not just about visual overload, as in our hidden picture puzzle example. According to Cognitive Load Theory (Chandler, P., & Sweller, J.), there are various forms of cognitive load:

• Intrinsic Load (content complexity) describes the inherent, real complexity of topics (e.g. the ordering process for a customisable product). This load is unavoidable and a harsh reality we must deal with. 
• Extraneous Load (unnecessary additional load) describes an unnecessary additional cognitive load that does not stem from the complexity of a topic, but arises solely from design decisions or the absence of such decisions (e.g. when information is missing from an ordering process, yet people need it to make an informed decision or provide the correct details).  
• Germane Load (meaningful cognitive load) describes the learning-related load, i.e. the part of the cognitive load that is necessary for the learning process and which, when encouraged, enables the targeted construction of schemata/mental models in working memory or the retrieval of existing mental models (e.g. familiar patterns for navigation, searching, donation forms), which simultaneously keeps both intrinsic and Extraneous Load low.

The link between cognitive load and error rates, and what this means for your digital products

First, the bad news:the more sensitive the context, the more detrimental the consequences of high cognitive load. If, for example, people become overloaded during a payment or donation process, there is a high probability that they will abandon it. The risk of them making a mistake or suboptimal decisions is simply too high. As if the loss of the failed transaction weren’t bad enough for us, another possible consequence is a loss of trust. A double whammy of 💩, so to speak. So how do we deal with these different types of load?

Accept Intrinsic Load

If we want to – or need to – facilitate complex processes online, we must be aware of and accept the Instrinsic Load these entail. This form of load simply exists, and we can manage it, for example through good structuring and pacing of content. Breaking long processes down into multi-stage steps that are comprehensible to people (e.g. personalising the order in step 1, entering the address in step 2, etc.) is one way in which good experience design typically deals with this. Providing useful and appropriate help texts is another. These are all learned patterns (thanks to Germane Load) that we can make use of and that can keep the error rate low.

Minimise Extraneous Load

Unlike Instrinsic Load , Extraneous Load is self-inflicted and a consequence of e.g. overloaded, unstructured processes that have either been designed or not designed. This type of load is avoidable and can always exist, even in seemingly less complex processes. Imagine, for example, a simple contact form with 5 input fields. Easy, right? Not if none of the input fields tells you what kind of information should go in the field (e.g. name, email address, message, phone number) and you can’t submit the form until the correct content is in the correct field [insert sound effect 🚿👵🏼🔪 Psycho, Shower Scene https://youtu.be/HCpRt-3SMWE?si=n3v55G2HSpmma9wz]. The potential error rate in this example is incredibly high, and people would – quite rightly – abandon the process. 

Leverage Germane Load 

Everyone brings knowledge and expectations – so-called mental models and schemas – with them when they start an interactive process. They have certain expectations, such as how an ordering process typically works. We can make use of this so-called Germane Load by designing in line with these expectations. We can also use Germane Load to form new schemas or mental models – e.g. by using consistent patterns that users anchor in their long-term memory. A successful example of schema formation is the hamburger menu (☰). Whilst many people initially wondered what these three horizontal lines meant, it has now become an established mental model that most people immediately recognise as a menu (https://www.nngroup.com/articles/hamburger-menu-icon-recognizability/). 

In summary, if we accept Intrinsic Load , minimise Extraneous Load and specifically promote Germane Load , we are well on the way to ensuring that people are not overwhelmed by the content in our digital processes and on our websites. But how do we even know if our designs have a cognitive load problem?

Is cognitive load preventing successful conversions in your digital products?

If you’re now wondering whether cognitive load is a root cause of problems on your site or in your digital processes, you can find clues by, for example, observing people as they use your site. This doesn’t have to take the form of a sophisticated usability test. Grab some colleagues, acquaintances, friends or family members and ask them, for example, to find a specific piece of content on your site or to go through a process (donation, registration, login, you name it). Then keep an ear out for tell-tale soundbites (e.g. “Do I have to fill all this in?”, “Where can I find…”, “What was that called again?”, “What have I done wrong now?”). Facial expressions are also a good indicator. Examples include 😳, 🥵, 🫠. If this sounds or looks familiar, it could be down to cognitive load. 

Your digital product is overwhelming people – what now? 

Now that you’re aware of the issue, you can review and optimise your pages and processes. Keep the mantra “Accept intrinsic load, minimise extraneous load, leveraging germane load” in mind.

Alternatively, you can have a look at our new audit service, which we’ve developed specifically for this scenario: the CARE-Check. The CARE-Check shows you how your website or app really works – not just technically, but for real people. You can find out exactly what the term CARE means here: https://birdux.studio/en/services/ethical-website-audit/ 

Did you find this content helpful? The German-language LOVE /: RAGE LinkedIn newsletter is published monthly on LinkedIn and can be subscribed to there: www.linkedin.com/newsletters/love-or-rage

A complicated donation process here, an unclear sign-up process there, or hidden barriers elsewhere lead to users abandoning the process – and organisations losing precisely the people they want to reach. For social impact start-ups and value-driven companies (ESG, purpose-driven), NGOs and social initiatives, universities and educational institutions, as well as political organisations and foundations, this is more than just a UX problem: it contradicts their very mission.

When optimising processes and breaking down barriers, you quickly reach a point where you have to ask yourself: where do I start, and where do I stop? 

On the one hand, you want (or need) to become accessible as quickly as possible (Hello, BSFG/EAA & DSA! ✌️); on the other hand, resources may be limited, and it’s not hard to get lost in the process. Sound familiar? Don’t panic!

CARE-Check to the rescue! 

Enabling digital accessibility and ethical standards one step at the time

With our CARE-Check (or audit), we offer a structured website review from a psychological and ethical perspectivethat goes beyond mere functionality. Depending on the CARE package you choose, the audit results and prioritised recommendations will be available to you within 5 days (Essentials package), 2 weeks (Impact package) or a maximum of 6 weeks (Deep Dive package). 

We analyse digital products across four key dimensions: Cognitive, Accessible, Responsible, Ethical – in short CARE.

The four dimensions of ethical UX 

• Cognitive: We check whether we are overwhelming our users, e.g. by offering too many options or too much information.
• Accessible: : In the audit, we examine whether we are excluding people, e.g. due to poor colour contrast, small touch targets, or a lack of WCAG compliance.
• Responsible: Are we unconsciously using manipulative design tricks, such as hidden sign-up options or misleading buttons? These questions are the focus of the Responsible Audit.
• Ethical: In this audit, we check whether our product is fair, trustworthy and respectful towards everyone.

How do you benefit from ethical UX Design?

The CARE Audit is ideal if: 

You are experiencing high drop-off rates: You have a specific area (e.g. donation checkout, course registration or application form) where users are leaving, and you want to know why.

You need quick wins: You require recommendations that can be implemented immediately – with a limited budget and time commitment.

Want to minimise risks: You want to identify potential barriers (WCAG) or problematic design patterns (DSA) at an early stage to avoid legal pitfalls.

Are looking for sound guidance: You want a clear assessment of where you currently stand in terms of digital accessibility and ethical standards.

Clear guidance in 4 steps and 5 days 

Our process is streamlined and tailored to your day-to-day operations

No-obligation initial call (15–30 mins)
We clarify your requirements. Then you decide whether a CARE Audit package is right for you.

Kick-off
Together, we define the process and the screens we will review in the CARE Audit.

Analysis
We review your application against the CARE dimensions.

Report & Handover
You’ll receive a detailed report including an executive summary, screenshots and a prioritised roadmap (from quick wins to strategic recommendations). 

The best part: it’s quick and easy. Once you’ve booked a CARE package, you can have the finished report in your hands within 5 days of the kick-off . 

Launch offer: 20% off the Essentials package

To mark the launch, we’re offering you a fantastic deal: the first three enquiries will receive a 20% discount on the Essentials package:

• Regular price: 4.900 €
• Launch offer: 3.920 €
• Your saving: 980 €

Full details of the CARE Essentials Package, as well as the other CARE packages, can be found on the CARE Audit page.

Book a no-obligation initial call now

Would you like to know how your website or app is performing in terms of digital accessibility and ethical standards? 

Book a video call

Previously, the mails were sent out to students and staff as unformatted text only emails – functional, but hardly inviting. Our assignment included developing two newsletter templates (digest and single-topic newsletters), each for desktop and mobile, while adhering to WCAG-AA guidelines.

From Plain Text to Snackable and Structured Design Offering Quick Orientation

The RPTU newsletter before the redesign: a purely text-based email with long, unformatted text. 

Visual hierarchy and the Gestalt principle of proximity play a crucial role , especially in newsletters. The faster recipients can perceive and understand whether and which content is relevant to them, the more likely they are to engage with it. This is particularly important for information that regularly lands in inboxes – otherwise, the reflex to delete the email may dominate over actually reflecting on its content. 

When designing the RPTU HTML newsletters, we therefore focused heavily on typography, colour combinations, spacing, and whitespace, as these elements create structure and orientation. Of course, the design also needed to align with the existing corporate design while remaining WCAG 2.2-AA compliant. 

How We Brought Together 20 Topics, 3 Target Groups, and 3 Event Locations

For the digest newsletter, which thematically covers a wide range of content for diverse target groups, we created a solution with clear headline structures, icons, and colour coding.

The previous 20 topics for three recipient groups (students, staff, all) across three locations (Kaiserslautern, Landau, all, or online) were consolidated into 6 topic clusters (Events – Leisure, Events – Teaching, Events – General, News, Campus Life, Other), each colour-coded. 

The colour coding of topic clusters significantly improves orientation and information capture, helping recipients identify relevant information at a glance. Additionally, we selected suitable icons from the RPTU's icon library that serve as ideal visual anchors for subtopics (e.g., Events – Leisure/Sports).

Technical requirements such as file size limitations and the use of web-safe fonts (Arial instead of the corporate font specified in the corporate design) also required pragmatic solutions.

The RPTU newsletter after the redesign: WCAG 2.2-AA compliant, structured HTML newsletters with colour-coded topic clusters, icons for subtopics (Events – Leisure/Sports and Events – Teaching/Science), and clear visual hierarchies.

Communication Optimisation Through Thoughtful Structure and Aesthetics

This project demonstrates how thoughtful design improves communication: visual hierarchies created through strategically used font sizes and spacing guide attention, clear structures simplify navigation, and accessibility ensures that all recipients can actually use the information.

With the new mailing templates, RPTU now reaches over 20,000 students and staff across its two locations in Kaiserslautern and Landau – with newsletters that not only look better but, according to corporate communications, primarily function better.

"So far, we’ve received nothing but positive feedback. With its structure and new design, the new newsletter offers real added value."

Iris Fares, University Communications

GrowFlowFly Final Report

Free Download

Last week, the final report for GrowFlowFly was published and can be downloaded free of charge as a PDF from ReNaTe (Repository for Natural Sciences and Technology) : https://oa.tib.eu/renate/items/8038eb24-6b6f-4897-9270-c6776b4b22e2

Excerpt from the summary“The increasing urgency of transforming energy systems calls for innovative approaches to optimally utilize energy and land-use potentials while more strongly integrating positive design potentials contributed by citizens. The aim of this research project was to develop a digital participation tool and to pilot it in two structural transformation regions (NW & ST) with both experts and citizens. The project conducted a comparative acceptance analysis between conventional renewable energy technologies - wind turbines and ground-mounted photovoltaic systems - and land-efficient innovative renewables, including agri-photovoltaics, floating photovoltaics, and airborne wind energy, while also evaluating the tool’s design and participation potential. In addition to the technological parameters of the renewable energy types, key acceptance factors, landscape-structural spatial requirements, as well as user experience and gamification elements were developed interdisciplinary as framework parameters for the gamification tool.(...)”

GrowFlowFly is a prime example of what can be achieved when science, experience design, and technology work together as a team. Since User Experience Design can help convey comprehensive content and information playfully to users without prior knowledge, the usage of UX in research projects can play a decisive role. 

More about the participating companies and organisations

• The network partners and their contact persons: Technische Universität München (TUM) – Prof Dr Sören Schöbel-Rutschmann, Martin Luther University Halle-Wittenberg (MLU) - Prof Dr Gundula Hübner (Environmental and Social Psychology), RWTH Aachen University (RWTH) – Prof. Dr Dieter Moormann (Institute of Flight System Dynamics), aratall GmbH, Osnabrück – Stefan W. Kauling, https://www.aratall.com/
• Associated project partnersTU Delft, Projekt MEnergie (TH Köln et al.), Kreis Düren, Gemeinde Aldenhoven, Energieavantgarde Anhalt e.V. Dessau, Airborne Wind Europe aisbl - Brussels, regstrom e.V. Halle, Saline Technikum (Berufliches Bildungswerk e.V., Halle Saalkreis), Next2Sun Agri-PV - Dillingen/Saar, Krinner GmbH Agri-PV - Straßkirchen, BayWa r.e. - Projektentwicklung PV DE München, Stadtwerke Düren

More GrowFlowFly News on our Studio Blog: 

• The game for the energy transition - MDR reports on the first tests of the GrowFlowFly project – Juni 2024
• Together for a sustainable future! Start of cooperation with the TU Munich in the joint project GrowFlowFly – November 2023

Quick recap from Part 1: We analysed why Duolingo's "Did you just come back to ghost me again?" notification works so well. Our conclusion: It plays with self-irony instead of guilt, a consistent product personality, honesty, and perfect timing. Check out part 1, "Seriously?" – Duolingo's UX Writing | Part 1: The Anatomy of a Working Notification here.

Why the same tone can motivate and manipulate - depending on context

Here we’ll answer the crucial question: Why does this tone work with Duolingo, but would be toxic with Instagram or TikTok?

Duolingo is a learning tool

• It's about an intrinsic goal: an example of a "pleasurable troublemaker" according to Prof. Matthias Laschke, a small, benevolent disruption that specifically encourages me to stick with it and control my own behaviour reflectively (cf. also Deci & Ryan, 1985)
• Adds value: The app helps me with my personal goals
• Alignment with user goal: The app wants the same thing I do - progress
• Finitude: There is a goal, instead of endless engagement

For comparison: Social Media is an Attention Trap

• Extrinsic goal: The app wants my time, I want, e.g. distraction
• Questionable value: More scrolling ≠ more quality of life
• Misalignment: The app wants my attention to sell it
• Endlessness: There's never enough, never a "finished"

The ethical boundary lies in purpose. : the same UX techniques can be motivating or manipulative – depending on whether they help me achieve my goals or distract me from them. (Cialdini, 2008; Fogg, 2003). Self-Determination Theory, as proposed by Deci and Ryan (1985), provides a crucial psychological foundation. It emphasises that people are particularly motivated and satisfied when their needs for autonomy, competence, and social connectedness are met. This is why honest and transparent design aimed at user well-being is both ethical and effective.

For Practice: A Framework for Ethical Engagement

We have developed a simple and practical question test to check whether an engagement strategy is ethically justifiable:

The 3-Question Method:

1. Does it help the user achieve their own goal?
   • Yes - with Duolingo (learning a language)
   • No - with endless social media doomscrolling
2. Would it still work if we transparently communicate what we're doing?
   • Yes: "We're reminding you because regular practice helps with learning"
   • No: "We're showing you emotional content so you stay longer"
3. Would we want our own children/partners/friends to be addressed this way?

If you can answer all three with "yes," you're probably on the right side.

These three questions are not random, but are based on established frameworks from UX Ethics, such as Sharon Lindberg's Design Ethics at Work (research project on ethical considerations in design) and Ethical Persuasion models (e.g., the TARES Test by Baker & Martinson, 2001, as a conceptual construct showing how to evaluate transparency and user well-being in Persuasive UX).

Our goal was to translate these theoretical principles into practical question logic that product teams can apply directly.

What Designers and Developers can learn from this

1. Tonality is strategic, not decorative

Many teams treat UX writing as an afterthought: "Quickly write a text for the button." But tonality is part of the user experience – sometimes even the crucial part.

When implemented, this means:

• Define your product's personality before you write the first text
• Create a tone-of-voice guide with concrete dos and don'ts
• Test different tones with real users – what is perceived as funny by the team can be annoying for outsiders

2. Know your Product's Context

The tone works with Duolingo. Would it work with a banking app? Disaster! With a meditation app? Completely misses the mark!

Ask yourself:

• What is the user's emotional state in this moment?
• What is the overarching goal of my product and my users?
• What type of relationship should my product have with users?

3. Honesty > Tricks

Users aren't stupid. They notice when you're trying to manipulate them. Duolingo's directness works precisely because it's not disguised..

Here are some alternatives to deceptive patterns:

• Instead of "Only 3 left in stock!" → "Others are also looking at this right now"
• Instead of "Your friends miss you!" → "You have 3 new messages"
• Instead of "Last chance!" → "The offer expires on dd/mm/yy"

4. Humanisation requires Consistency

One funny notification is nice. A consistent personality across all touchpoints is an example of good UX Design.

A checklist for your product:

• Does your app sound the same in onboarding, notifications, and error messages?
• Would users recognise your product's "voice"?
• Does the tonality fit your brand – or does it feel forced?
• What role should your product take on?

Practical Tips: How to develop the right Tone

Step 1: User Research with Focus on Language

Understand how your users think and talk:

• How do your users talk about your product?
• What metaphors do they use?
• What relationship do your users want with the product?

Step 2: Define your Product Personality

Imagine your product as a person:

• What job would they have?
• How would they talk in real life?
• What would they NEVER say?

Step 3: Test in extreme Situations

A good tone shows itself in crises:

• How does your app talk when something goes wrong?
• How does it communicate errors?
• How are users addressed who have been away for a while?

Step 4: Iterate with real Feedback

User exposure will change your tonality. Test, learn from it, and adjust accordingly.

What Duolingo also teaches us: The Power of Microcopy

This notification we are discussing is about 15 words long. But it shows what microcopy can do when it's done well:

• Re-engagement without pressure
• Brand building through consistency
• Emotional connection through humour
• Behaviour change through the right tone

The best UX texts are the ones you show your friends and say, "Look how funny this is!"

Conclusion: UX Writing is UX Design

Duolingo's owl is not a marketing gimmick. It's a carefully developed personality that has a positive effect on the user experience. Every notification, every button text, every error message was designed so that it:

• Fits the Duolingo brand
• Understands their users
• Communicates honestly
• Motivates action – but without manipulating

This is not a coincidence, but a strategy.

And in the end, good UX writing is like good UX overall: invisible when it works and unforgettable when it has an impact.

This is part 2 of the two-part series on UX Writing and Design Ethics. 👉 Part 1 can be read here: The Anatomy of a Perfect Notification

At BIRD UX, we have been developing the strategic foundation for such nuances since 2011: We help our clients develop consistent UX strategies in which content and text are not added as an afterthought, but are considered from the beginning as a fundamental component of the user interface. Tonality is not a question of copywriting – but of product strategy. Sounds like we can help you with your challenges? Feel free to get in touch!

Literature

• Baker, S., & Martinson, D. L. (2001). The TARES Test: Five Principles for Ethical Persuasion. Journal of Mass Media Ethics, 16(2–3), 148–175. https://doi.org/10.1080/08900523.2001.9679610
• Cialdini, R. B. (2008). Influence (5. Aufl.). Pearson
• Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. New York, NY: Plenum
• ​​Fogg, B. J. (2003). Persuasive technology: using computers to change what we think and do. Morgan Kaufmann Publishers
• Hassenzahl, M., & Laschke, M. (2015). Pleasurable Troublemakers. In S. P. Walz & S. Deterding (Eds.), The Gameful World (pp. 167–196). The MIT Press. https://doi.org/10.7551/mitpress/9788.003.0011
• Lindberg, S. (2024). Design Ethics at Work. https://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-233188

Article Image Character from https://duolingopress.lingoapp.com/

As a women-led studio, we’d like to do our part: We’re offering two non-profit organisationsworking to end gender-based violence a free UX and Service Design consultation. We are calling this offer Orange Hours 🧡.

Orange Hours 🧡

Our offer

✅ A tailored consultation via Zoom (2–3 hours)
✅ Together, we’ll take a close look at your digital offer — for example, a quick UX audit of your website or the service design of your online counselling platform.
✅ You’ll get concrete, actionable recommendations. If you have designers or developers who’ll be involved in implementing these, they’re more than welcome to join the session.
✅ Our focus: Accessibility & Safety by Design — because safe, accessible digital spaces can save lives.

Who it’s for: Two non-profit organisations working in the field of counselling, prevention, or protection against gender-based violence.
Application deadline: 20 November 2025, 18:00 (C.E.T.)

Participation details

Participation details Each organisation will receive a dedicated 2–3 hour consultation session via Zoom. We can offer two possible time slots per organisation – please choose one: 

• Organisation 1:
  • EITHER Thu, 27 Nov 2025 (start between 10:00 and 14:00) 
  • OR Fri, 28 Nov 2025 (start at 10:00) 
• Organisation 2:
  • EITHER Thu, 4 Dec 2025 (start between 10:00 and 14:00) 
  • OR Fri, 5 Dec 2025 (start at 10:00)

IMPORTANT: Please note that we unfortunately won’t be able to accommodate other dates.

How to apply for Orange Hours

Send us an email at hello@birdux.studio
subject line: 🧡 Orange Hours Application – [Name of your organisation]

and include the following details:

• A short intro about your organisation (Name + 2–3 sentences: What do you do? Who do you support?)
• Which digital product or service you’d like to improve (e.g. website, app, online counselling tool, booking system)
• Your challenges: What’s worrying you about your digital offer? What would you like to improve but aren’t quite sure how?
• How many people do you reach on average (monthly/yearly)? 
• Why now? (e.g. Are you planning a relaunch? Have you received negative feedback? Do you have resources to make changes soon?)
• Which consultation slot works best for you? 

Application deadline: 20 November 2025, 18:00 (C.E.T.)

A heartfelt thank-you to all the organisations working every day to end gender-based violence. Your work is essential — and deserves every bit of support it can get. We’re looking forward to your applications and hope that, through the Orange Hours, we can help two organisations make their vital work a little more accessible. 

Because safe, accessible digital spaces can save lives – and sometimes it’s the small improvements that make the biggest difference.

See you soon! 🧡
Jenny and Steffi from BIRD UX

Links and more info about Orange Weeks

• https://www.unwomen.org/en/what-we-do/ending-violence-against-women/unite/orange-the-world
• https://en.wikipedia.org/wiki/16_Days_of_Activism_against_Gender-based_Violence 
• https://www.hilfetelefon.de/en/ 
• https://www.der-paritaetische.de/alle-meldungen/tag-gegen-gewalt-an-frauen-und-orange-days-2025-veranstaltungen-der-inforeihe-kinder-jugend-und-familie/

For months, I successfully ignored Duolingo. Learning French again? Nice idea, but somehow lost between work, teaching assignments, and psychology studies. Yesterday: Downloaded an app update, opened it briefly, closed it again. So... yeah: did nothing.

Dann hab ich später auf mein Handy geschaut  und da ist sie, diese Notification: “Did you just come back to ghost me again? Now do a French lesson!” Ich musste echt kurz lachen. Dann kam die UX Designerin in mir durch: Why does this work so well and "catch" my attention?

From my own experience: What other apps say

To understand why Duolingo's approach is special, let's look at how other apps communicate with inactive users:

• Garmin: Status "unproductive" after a 15 km run on Sunday at 7 am (Um... thanks?)
• To-Do-Apps: "12 overdue tasks are waiting for you" (I know. That's why I'm not opening the app, haha.)
• Social Media: "You're missing what your friends are posting right now!" (FOMO as a business model.)
• Meditation-Apps: "Your 30-day streak is lost 😔" (Note the irony that a meditation app creates stress...)

So, in summary, most apps use a combination of guilt, artificial urgency, loss aversion (= you're "losing" something), and social pressure. Duolingo does something different.

The Anatomy of Good UX Writing: What Duolingo Gets Right

1. Self-irony instead of guilt-tripping

„Did you just come back to ghost me again?”

The app says what I'm thinking myself – before I can formulate it. It knows my behaviour (downloading an update, opening it to... well... do nothing) and comments on it with a wink. No accusation, but rather "I see through you, and that's okay."

Why this works: People react defensively to accusations, but openly to humour. The self-ironic tone signals: "We don't take ourselves too seriously – and we don't judge your behaviour either."

But wait – isn't this also guilt-tripping?

At first glance, you might think: The "ghosting" accusation, that's also guilt induction! But there's a crucial psychological nuance here.

Classic guilt induction (à la Garmin)

• Focus is on failure: "You are unproductive"
• Identity level is attacked (even if implicitly): "You are inactive/lazy/unmotivated" (addresses the person instead of the behaviour)
• Focus on past and loss aversion: Emphasis is on what you've already lost.
• Comparison with others: "Others have already achieved this and that..."

The emotional effect is usually shame and resignation, so rather negative. Duolingo's approach doesn't judge you:

• Keeps focus on your behaviour: "You're ghosting" (not "You are a ghoster")
• Stays on the situational level: Describes what you do, not who you are (Gollwitzer, 1993, 2006) (vgl. Gollwitzer, 1993,2006) 
• Is present-oriented: "Now do..." (call to action)
• No comparison with others: Just you and the owl

It does have an emotional effect, but a positive one: You feel caught, but you have to smile. That leaves a good feeling rather than a bad one.

The crucial difference: Duolingo names the behaviour playfully, without judging the person. That's a huge psychological difference. I can change behaviour, but not my identity. Additionally, the owl presents itself as the "victim" ("...ghost me"), which humorously turns the tables. I don't feel bad about myself, but smile about the situation. This doesn't trigger defensiveness, but opens me up to the call to action.

2. Anthropomorphisation* with personality

The Duolingo owl isn't a neutral learning AI. It's a personality – perhaps sometimes slightly passive-aggressive, always persistent, but ultimately – and this is the thing: on your side. It reminds us of people we know: the friend who asks, "When will we finally see each other again?" or the trainer who says, "Great that you're here – now actually do something!"

That differs from other apps: The owl is consistent in its personality. Not sometimes sweet, sometimes strict, sometimes desperate. It has a recognisable character, and that's exactly what creates a relationship.

*Anthropomorphisation = psychological attribution of human characteristics to a non-living object 

3. Honesty over manipulation

„Now do a French lesson!“

No false urgency ("Only 2 more hours!"), no fake scarcity ("Last day!"), no emotional blackmail. Instead: a direct request. Almost cheeky, but very honest.

Psychologically clever: This almost cheeky directness feels refreshing in a world full of manipulative notifications. We're so used to deceptive patterns that honesty is surprising and positive.

4. Timing and context

The notification didn't come randomly. It came exactly after I opened the app and closed it again – without completing a French lesson. That's precise behavioural tracking, but in service of a meaningful intervention. (Fogg, 2003)

The context makes a difference. With a shopping app, the same timing would be creepy ("You looked but didn't buy?"). With a learning tool, it's legitimate. Read more about context in Part 2 (coming soon!). 

What we've learned so far

Duolingo's notification works because it:

• comments on behaviour instead of judging identity
• Has a consistent personality that creates a relationship
• Is honest instead of manipulative
• Understands context and is perfectly timed

But here comes the crucial question: Why does this tone work so well with Duolingo – and would be toxic with Instagram or TikTok? Where is the ethical boundary between charming motivation and manipulative retention?

Part 2 of this article will cover:

• Why the same tonality motivates with Duolingo – and manipulates with Instagram
• A 3-question framework for ethical engagement design
• 4 practical steps to the right tone for your product

Follow us on LinkedIn (www.linkedin.com/company/birdux-studio) or Bluesky (bsky.app/birdux) so you don't miss Part 2. 

Literature

• ​​Fogg, B. J. (2003). Persuasive technology: using computers to change what we think and do. Morgan Kaufmann Publishers
• Gollwitzer, P. M. (1993). „Implementation Intentions: Strong Effects of Simple Plans.“ American Psychologist.
• Gollwitzer, P. M., & Sheeran, P. (2006). „Implementation intentions and goal achievement: A meta-analysis of effects and processes.“ Advances in Experimental Social Psychology.
• Gollwitzer, P. M., & Sheeran, P. (2006). „Implementation intentions and goal achievement: A meta-analysis of effects and processes.“ Advances in Experimental Social Psychology. Rationality in action: Contemporary approaches (pp. 140–170). Cambridge University Press. (Reprinted from „Econometrica“ 47 (1979), 263-91)

Article Image Character from https://duolingopress.lingoapp.com/

Founded in 2011—when specialized UX studios were still a rarity in Germany—we started out with a clear mission: to design well-thought-out, user-centered solutions for an increasingly complex digital world. Because we know that only when users are happy does the business have reason to celebrate—and that's exactly what has been driving us for 14 years.

We’re proud to be a women-led studio, and to have evolved in the ever changing landscape that is Experience Design. 

Our rebrand from The Geekettez to BIRD UX Studio this year shows that 14 years later, we remain as curious, and committed to the ongoing challenges that good and thoughtful UX can solve.

Thanks to everyone who’s been part of the flight so far. Here's to designing what's next.

Photo Credit:  Alexander Grey on Unsplash

What Are Deceptive Patterns / Dark Patterns?

The DSA defines Dark Patterns in its recitals as practices in online interfaces that are "designed to distort or substantially impair the ability of users to make autonomous and informed choices or decisions."
In simpler terms, these are deceptive design techniques intended to trick users into making choices that may not be in their best interest, but are beneficial for the service or platform provider.

What Exactly Does the DSA Prohibit?

Article 25 of the Digital Services Act prohibits online platforms from designing user interfaces in a way that misleads or manipulates users. However, this is where things become complicated, as the law employs intentionally vague terms that necessitate interpretation on a case-by-case basis. Expressions like: “Unduly prominently displayed”, “Excessively difficult”, “Deceptive default settings”...all leave room for debate.

• So, when exactly is something “unduly prominent”?
• When does a cancellation flow become “excessively difficult”?
• And when does a “default setting” cross the line into “deceptive”?

The DSA offers some examples—such as repeated prompts to make the same selection or deliberately complicated unsubscribe processes—but ultimately, the exact threshold is often unclear. A grey area, indeed.

Who Decides When the Line Is Crossed?

The DSA’s open-ended language allows for flexibility in addressing emerging manipulation techniques, but it also introduces inevitable ambiguity.
In practice, enforcement and interpretation fall to:

• National courts and supervisory authorities
• The European Commission, which may issue further guidance
• Consumer protection organisations, which can initiate legal action

Assessments are typically based on an objective standard: Would an average user’s ability to make a free and informed choice be significantly impaired? iCommon evaluation criteria include

• Level of effort required
• Clarity and comprehensibility
• Number of clicks
• Visual design choices
• Comparison with alternative options
• Observable effects on user behaviour

Ultimately, the decision often relies on expert opinions, user testing, and comparisons with industry norms, creating both flexibility and legal uncertainty for designers and businesses alike.

What Does This Mean for Us as Designers?

We believe this legal "grey zone" presents both a challenge and an opportunity. On the one hand, it offers some room for interpretation. On the other hand, it forces all decision-makers—from product managers to executives—to take ethical design seriously. It pushes us to pause and ask:

• Are we helping users make informed, autonomous choices?
• Or are we nudging them towards decisions that serve the business model more than the user?

Final Thoughts

Yes, the Digital Services Act leaves room for interpretation when it comes to Deceptive Patterns. But rather than viewing this as a legal loophole, we see it as a call to action for designers, marketers, product owners, and leadership teams alike.
Because in the end, it’s not just about compliance or avoiding penalties. It’s about building long-term trust in a digital world where transparency and user autonomy are no longer optional, but essential pillars of any future-proof digital strategy.

Sources:

• VERORDNUNG (EU) 2022/2065 DES EUROPÄISCHEN PARLAMENTS UND DES RATES vom 19. Oktober 2022 über einen Binnenmarkt für digitale Dienste und zur Änderung der Richtlinie 2000/31/EG (Gesetz über digitale Dienste) (Language DE): https://eur-lex.europa.eu/legal-content/DE/TXT/PDF/?uri=CELEX32022R2065 (especially Recital No. 6, Article 25)
• Dark Patterns nach dem DSA (Language DE):: https://haerting.de/wissen/dark-patterns-nach-dem-dsa/
• The Digital Service Act (Language EN):: https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age/digital-services-act_en