Your Business Data Isn't Ready for 2026 (and it's going to hurt)

Here’s what 20 years of wrangling data for clients has taught us about the AI revolution

Remember when having a website was enough? When “digital transformation” meant moving from paper to spreadsheets? Those days are done.

What’s happening now is different. We’re watching entire industries realise their data is a mess whilst their competitors are already building AI-powered advantages. The gap between prepared and unprepared businesses is about to become a chasm.

Our Data Journey (Why We’ve Seen This Coming)

For twenty years, Devstars has been the agency people call when they need complex data problems solved. We’ve never just been a web agency, though that’s often how we’re seen. There’s always been a data science angle to what we do.

From utilising structured data to highlight SlimFast recipes to linking nearly 50,000 courses to over 400
UK Standard Occupation Codes (SOC) for Sort Your Future.

Every large website we’ve built has required careful data management. Not just storing it, but organising it in ways that actual humans can query and find what they need. Sounds basic, but you’d be amazed how many businesses still struggle with this fundamental requirement.

The DSTL Project That Changed Our Perspective

Things amplified in 2020 when we worked with DSTL (Defence Science and Technology Laboratory), a branch of the Ministry of Defence. They needed help building a platform for organising data on XR and VR technologies for training purposes.

What they had: An existing database that was becoming unwieldy. What they needed: A platform where researchers could effectively manage and query growing volumes of unstructured data.

We built them a system using Elasticsearch that introduced fuzzy logic and machine learning capabilities. The system could take masses of unstructured data and start adding structure to it automatically. Pretty powerful stuff.

That project taught us something critical: Most organisations have no idea how much value is locked in their messy data.

The AI Watershed Moment

In early 2022, we started experimenting with tools like Jarvis (remember that?). They were showing glimmers of what was coming. Then November 2022 hit. ChatGPT 3.5 launched.

I posted Tolstoy’s observation, “There are decades when nothing happens, and there are weeks when decades happen.”. That’s exactly what it felt like. A seismic shift, happening in real-time.

ChatGPT grew at a record pace, but here’s the thing: really useful business adoption has been slower. We’re all still figuring out how to use this technology safely, effectively, with proper guardrails.

2025 feels like the year of reckoning. The AI platforms are now at such a height of quality that ignoring them is no longer an option. But there’s a massive problem most businesses haven’t grasped yet.

Why Your Data Strategy Is Everything

Here’s the brutal truth: Writing dumb prompts into ChatGPT isn’t going to solve your business problems.

What actually matters:

• You need good data
• You need to own that data
• You need it secured properly
• You need it well-formatted with proper structure and metadata

When you’re using these powerful machine learning tools, you want them working on YOUR data, not the internet’s generic mass of information. Otherwise, you get:

• Hallucinations
• Generic tone of voice
• Answers that could apply to anyone
• Zero competitive advantage

The real opportunity? Creating digital twins. Building your own small language models trained on your specific data, your processes, your expertise.

The Bigger Picture: We’re Not Digitising, We’re Reinventing

Every 80-year cycle includes a brutal fight. The defenders of the old system (who benefit from it or fear change) versus the builders of the new system (who see the old one failing most people).

2025 will be seen as a pivotal year in this cycle. We’re not just adding digital features to old processes. We’re inventing entirely new economic models.

The AI Revolution (Your Mental Amplifier)

ChatGPT-3.5 in November 2022 will be remembered as the starting gun for the Age of AI. It’s comparable to the steam engine launching the Industrial Age.

We’ve crossed from “clunky and broken” to “holy shit, this changes everything”. It’s the iPhone moment for artificial intelligence. And once crossed, we don’t go back.

What to watch:

• Incremental improvements in existing workflows (predictable but limited)
• Entirely new business models only possible with AI (where the real money is)
• Legal and regulatory battles as old industries fight displacement

Your move: If you’re just thinking about “adding AI features”, you’re missing the point. Ask yourself: “What would our entire business model look like if intelligence were free and infinite?” Then build towards that.

The Energy Revolution (The Hidden Enabler)

Here’s something most businesses aren’t considering: Solar, wind, and batteries are technologies, not commodities. This changes everything about how we plan for the future.

Coal and oil are extracted, depleted, prices fluctuate wildly. Solar and batteries are manufactured. Costs drop about 20% each time production doubles (Wright’s Law). The result? Energy becomes cheap, predictable, and improves forever.

By 2030-2040:

• Energy costs approach near-zero in many contexts
• Computing becomes absurdly cheap (fuelling the AI explosion)
• Entire supply chains re-route around clean abundance

Your move: Any long-term infrastructure or platform you’re building should assume energy is cheap and getting cheaper. Design for abundance, not scarcity.

Your SMB Data Readiness Checklist for 2025

Right, enough context. Here’s what you actually need to do:

1. Audit Your Data (This Week)

• Where is your data actually stored? (Spreadsheets, CRM, random folders?)
• What format is it in? (PDFs, CSVs, databases, Post-it notes?)
• Who owns it and who can access it?
• What’s the quality like? (Duplicates, outdated info, inconsistencies?)

2. Build Your Data Foundation (Next Month)

• Centralise: Get everything into one system, even if it’s messy at first
• Standardise: Create consistent formats and naming conventions
• Secure: Implement proper access controls and backup systems
• Document: Write down what data you have and why it matters

3. Prepare for AI Integration (Next Quarter)

• Clean your data: Remove duplicates, fix inconsistencies
• Add metadata: Tag and categorise everything properly
• Create structure: Build relationships between different data sets
• Test small: Start with one use case for AI-powered analysis

4. Think Bigger (Next Year)

• Digital twin creation: Build AI models trained on your specific data
• Automation opportunities: Identify repetitive tasks AI can handle
• New business models: Consider what’s possible with infinite intelligence
• Competitive positioning: Move before your competitors figure this out

The Real Cost of Waiting

Every week you delay sorting your data is a week your competitors might be building their AI advantage. The gap between prepared and unprepared businesses is accelerating.

The businesses that win in 2025 won’t be the ones with the biggest budgets. They’ll be the ones with the cleanest data, the clearest strategy, and the courage to build for abundance rather than scarcity.

What We’re Doing About It

At Devstars, we’re not just talking about this shift. We’re building the systems that help businesses navigate it.

We’re combining twenty years of data wrangling experience with cutting-edge AI implementation. We help you audit your data reality, build proper foundations, and create AI-powered advantages your competitors can’t match.

The question isn’t whether AI will transform your industry. It’s whether you’ll be driving that transformation or watching from the sidelines.

Your Next Move

Start small but start now. Pick one data set. Clean it. Structure it. Then use AI to analyse it in ways you never could before.

The results will convince you faster than any blog post could.

Need help getting your data AI-ready? Let’s talk about where you are and where you need to be. Twenty years of solving complex data problems has taught us one thing: the best time to start was yesterday. The second-best time is now.

Stuart Watkins is Director at Devstars, where we’ve been turning data chaos into competitive advantage since 2004. We specialise in bespoke software, AI implementation, and making sure businesses are ready for what’s coming next.