How to Make Your AI Investment Actually Work This Year

Mar 3, 2026 - 16:00

A depiction of a robotic hand interacting with a laptop screen

By Satish Thiagarajan, the founder of Brysa, a Salesforce and data consultancy based in the UK. His company advises media, industrial, and services clients on using Data Cloud and Agentforce to turn signals into action. His work focuses on closing the loop between insight and execution in sales, marketing, and service.

Making AI investment work is an endless struggle. Year after year, tech leaders see their intentions crumble away into systems that never quite deliver what they promised, and with AI, the problem is becoming increasingly obvious. Not because people aren’t good at their jobs, but because they’re being caught out by scope. AI investment promises so much, but when you get drawn to a system for what it can do, rather than what your company needs, you’re never going to achieve worthwhile returns. 2026 calls for an entirely new approach.

Understanding the AI Maturity Model

Before a company can even consider AI investment, it first needs to work out whether the organisation is actually ready for AI. The AI Maturity Model can help you to work out what’s best for your current situation.

Stage 1: Foundation

You’re in Stage 1 if your data is inconsistent, systems are loosely connected, and teams still rely on manual fixes. Automation is basic, and AI efforts are fragile at best.

This is where most organisations sit, even those claiming to “do AI.” At this stage, you’re not AI-ready. The priority is data quality, integration, and operational discipline. Investing in advanced AI before fixing these foundations typically leads to disappointing results.

Stage 2: Enhancement

In Stage 2, data is largely clean, systems are integrated, and AI supports forecasting and decision-making. Intelligent workflows improve speed and efficiency, but humans remain central to execution.

AI is delivering value, but it’s augmenting work, not transforming it. The danger here is mistaking efficiency gains for true maturity.

Stage 3: Transformation

You reach Stage 3 when AI can act autonomously within clear guardrails, workflows adapt in real time, and systems continuously optimise without constant human intervention.

Teams shift from execution to strategy and oversight. AI is no longer a tool layered on top; it becomes a core operating capability. Few organisations have genuinely reached this level, despite how often it’s implied.

5-Part Framework to Making the Best Out of Your 2026 Tech and AI Investment

Reaching Stage 3 of AI maturity requires more than technology; you need a structured way of thinking about your approach to AI.

Start with outcomes

The AI mistake made by the majority of organisations is to buy AI first and then look for problems to justify it. The result is underused tools and weak ROI. If you want your AI investment to truly pay for itself, every purchase must link to a clear business outcome. And this can only be achieved when you ask the right questions:

How will this improve revenue?
How will this reduce operational friction?
Will this increase productivity or decision quality?
And how will this support an improved customer experience?

When outcomes are explicit, decisions improve. You stop chasing AI features and start funding measurable growth and differentiation.

Build vs buy vs partner

You can’t do everything in-house. But buying everything doesn’t always make sense either. And not every rollout should be done alone. For best results, you need to work strategically, because getting it wrong can cost years. Working to a simple rule can help:

Build when it’s a true competitive differentiator, and you have the talent and time to sustain it.
Buy when the function is standardised and non-differentiating, such as CRM, analytics foundations, and service management software.
Partner when speed is critical and internal expertise is limited.

Execution risk is often underestimated. Partnering isn’t outsourcing accountability, but rather accelerating maturity while avoiding costly mistakes.

Think people-first when seeking AI implementation

The reason tech investments fail is usually because people don’t use it, or don’t use it properly. So, blowing all of your budget on the best tech, without thinking about how it’s going to help your people, and how you’re going to help your people to use it, is asking for failure. We usually recommend allocating roughly 70% of investment to technology and 30% to people enablement, which includes role-based training rather than generic onboarding, contextual, in-the-flow learning, and clear ownership and accountability for everything relating to the system adoption. And this matters, because AI maturity is as much a cultural shift as a technical one.

Invest in ecosystems

Point solutions are easy, but they create integration debt and slow you down. AI maturity requires platform thinking:

A shared data model across functions
Intelligence embedded in workflows, not bolted on
The ability to evolve without constant redevelopment

That’s why platforms like Salesforce and Microsoft Dynamics 365 matter. They provide a unified data foundation across teams, built-in AI (Einstein), low-code flexibility, and continuous innovation.

Don’t ignore integration architecture

Integration isn’t glamorous, but it’s essential, and without it no AI investment can reach its full potential. Without strong integration:

Data becomes inconsistent
AI insights lose reliability
Automation fails at scale

Reaching Stage 3 requires a clear single source of truth, real-time (or near real-time) data sync, and an API-first architecture built for flexibility. Early integration supports that.

What next? When you follow the five-step plan, you gain a clearer view of where you stand on the AI maturity curve, and what you need to do to reach Stage 3 and gain the best for your business. The next step isn’t to rush into new tools or pilots, but to take stock and prioritise, choosing your investments deliberately. This may not be something you can do entirely inhouse; external support can bring in fresh eyes and the insight you need to remove the risk from wasted AI investment. But either way, you’re now on the right path to genuine business impact.