Manifesto

13 Aug 2025

The numbers tell a crystal-clear story. The global workflow automation market stands at $30.6 billion in 2025, with 83% of companies calling AI a top priority in their business plans, and an astounding 75% of new applications being built using low-code tools. We’re witnessing the largest productivity revolution since the assembly line, where AI and automation could add $15.7 trillion to the global economy by 2030.¹²³⁴

But here’s the uncomfortable truth most won’t admit: while the technology is ready, the tools we’re using to harness it are fundamentally broken.

The great UX catastrophe

Traditional automation platforms have turned what should be empowering technology into a nightmare of complexity. Despite promises of “no-code” and “drag-and-drop simplicity,” the reality is starkly different.

Most workflow builders force users into a visual programming paradigm that looks deceptively simple but quickly becomes a tangled mess. Users describe workflows with 250+ nodes becoming “unresponsive” and “spaghetti monsters” that are impossible to debug or maintain. What starts as a simple “when this, then that” quickly evolves into a complex web of connections that would make an electrical engineer weep.⁵⁶

Even “low-code” platforms require users to think like programmers. You need to understand concepts like conditional logic, data types, API endpoints, and error handling. One frustrated user captured it perfectly: “I’m not convinced this is ‘no-code’ if I need to learn JavaScript to use it”. Another noted the “steep learning curve” even for returning users after just a year away.⁷⁸

Popular platforms like Zapier may seem simple on the surface, but users quickly hit walls. Setting up anything beyond basic automations requires “more in-depth setup knowledge” and often leads to “force-fitting the way you work to match your automation tool’s confines”. The promise of simplicity becomes a bait-and-switch once you need anything custom.⁹¹⁰

The rise of the citizen developer

The low-code revolution has created a new class of workers: citizen developers. By 2025, 85% of enterprises are adopting AI agents, with over 65% of enterprises having citizen development programs. But these users are hitting the same walls as IT professionals.¹¹¹²

Organizations are investing heavily in training programs - Shell trained 6,500 business users as citizen developers - but the learning curve remains punishing. Users report needing to “look up technical help” repeatedly and facing “high learning curves when troubleshooting errors”. Low-code platforms promise flexibility but deliver rigid templates, leaving users facing “limited customization options” and finding themselves “constrained by predefined templates that may not fully align with a brand’s vision”. When you need something slightly different from what the platform imagines, you’re stuck.¹³¹⁴¹⁵

Despite thousands of available connectors, users constantly encounter “integration challenges with non-Microsoft tools” and situations where “partial functionality or authentication roadblocks appear”. Each new tool in your stack becomes a potential breaking point. What appears simple on the surface reveals layers of complexity beneath, where users discover they need to understand “XML documents”, “API endpoints”, and “data validation rules”. The promise of democratized development becomes a cruel joke when users need technical expertise anyway.¹⁴⁶¹⁶¹⁷

The quiet billion dollar problem

The workflow automation market is exploding - from $30.6 billion today to $87.7 billion by 2032 - but the dirty secret is that most of this spending is going toward tools that frustrate their users.¹⁸¹

Users routinely abandon projects after weeks of struggle. One frustrated Zapier user asked: “Why is it harder not easier to use Zapier?” describing how they “spend hours trying to figure out why something isn’t working”. Another described reaching a “tipping point” with Make that led them to completely switch platforms. Companies become “dependent on the platform’s ecosystem”, creating “vendor lock-in” where “switching platforms becomes difficult”. Organizations invest months in training and setup, only to find themselves trapped when they need to scale or customize.⁸¹⁰¹⁵

Traditional platforms force users to think in terms of triggers, actions, conditions, and data transformations. But humans don’t naturally think this way. We think in terms of outcomes: “When someone books a demo, I want my team to know about it and be ready.” The gap between human intention and platform requirements is where frustration breeds. Node-based editors look impressive in demos but become overwhelming in practice. Users describe interfaces as “clunky, confusing, or difficult-to-navigate” and note that “minor visual changes can cause scripts to fail”. The visual complexity that’s supposed to help actually creates cognitive overload.¹⁹²⁰

Once you build something complex, you’re trapped maintaining it. Users report “frequent script breakage due to UI changes” and automation that becomes “fragile”. What should be a set-it-and-forget-it solution becomes a full-time job.²⁰

The AI revolution is waiting for better UX

We’re at an inflection point. 85% of enterprises are deploying AI agents in 2025, and AI-powered automation could unlock $2.6-4.4 trillion in annual economic value. But this potential is locked behind interfaces that most people can’t or won’t use.²¹¹¹

The future isn’t about connecting nodes or mapping data flows. It’s about natural language: “Send a Slack alert to #demo-bookings whenever a product demo is booked via my Calendly and add the lead into the ‘Lead Tracker Q3-2025’ Excel sheet.” This shouldn’t require a computer science degree to understand or implement.

True democratization means your marketing manager, sales rep, or operations coordinator can create powerful automations without feeling stupid. It means the tool adapts to how humans naturally think about work, not the other way around. The most sophisticated technology should feel like magic - effortless and invisible. When you have to think about how the tool works rather than what you want to achieve, the tool has failed.

Introducing Skada

This is why Skada exists. We believe automation should be as natural as having a conversation. Describe what you want in plain English, and watch it come to life. No nodes to connect, no logic to debug, no engineering degree required.

“Send a Slack alert when someone books a demo and update my spreadsheet” becomes a working automation in seconds, not hours. No iterations, no editing, no debugging - just results. We built Skada around how people actually think about their work, not how computers process information. The complexity happens behind the scenes; you get the outcome. Managed authentication across 2,000+ integrations means you spend time on results, not setup. Your tools connect seamlessly because that’s how they should work.

The future of work isn’t about learning new tools - it’s about tools that learn you. While others ask you to think like a programmer, Skada lets you think like a human.

AI workflows at the speed of thought. Because your ideas shouldn’t have to wait for your tools to catch up.

We are currently in private beta. Join our waitlist to be among the first to experience the future of workflow automation.