Why tech vendors are key to solving AI's adoption problem

Many companies successfully complete AI pilots, but a lot struggle to scale AI application across their organizations. Only 15% of AI initiatives ever manage to truly scale.

Research from Boston Consulting Group suggests that only 26% of companies have developed the capabilities needed to move beyond pilots and generate meaningful value from AI tools. Meanwhile, McKinsey estimates that while almost every organization is experimenting with generative AI, only a small minority are seeing material financial impact across the business.

The technology is not the bottleneck.

Large language models offer a powerful solution that supports employees in various tasks, such as research, content generation, document analysis, meeting summary or task management. Despite these advances, teams still battle to turn AI capability into measurable business outcomes.

It is a translation problem.

There is a lack of depth that prevents these tools from meeting the needs of a specific industry and organization. General models are good at understanding language and information. But they have a hard time connecting them to workflows, regulations, governance requirements and operational nuances that rule industries, such as accounting, legal service, healthcare or manufacturing.

There needs to be an expert to bridge the gap between the capabilities of AI and the requirements of a profession.

1. A new role at the intersection of the technology and an industry

Technology vendors have traditionally been measured by the quality of their products. This is no longer enough.

Today, their role goes far beyond delivering a functional platform. They now actively contribute to the digital transformation of an organization. This includes sharing best practices, providing ongoing technical support, designing workflows, optimizing for governance and long-term application. All to support smooth product deployment.

Different industries are moving towards this same idea from different angles. When there is an intermediary that focuses on customer success, it translates capability into outcomes. Early examples of this are the rise of new jobs such as Customer Value Managers and Forward Deployed Engineers. These are professionals whose primary objective is ensuring the technology generates value for employees and the business overall.

Every industry is beginning to create its own version of the role.

In accounting, we see the emergence of so-called ‘Accountancy Engineers’. They are trained accountants with deep accounting expertise and technical capability. Rather than working from a specification document, they sit next to a firm’s team and co-design the workflow with them, building the automation alongside the people who will actually use it.

They build workflows tailored to how the profession works, effectively automating repetitive tasks and processes that answer to the challenges of accountancy. Just as important, they train the firm’s own accountants to become the practice’s internal AI architects, so the capability stays inside the firm long after the initial rollout.

For instance, they work to connect backend accounting systems with operational tools and cloud infrastructure and ensure that automated financial processes align with regulatory reporting and data governance requirements. This avoids generic automation bolted onto old corporate systems, resulting in a tailored integration.

And every pattern they uncover in the field, a workaround, a recurring bottleneck, a workflow nobody had written down, feeds straight back into the vendor’s own product roadmap, the same feedback loop that made Forward Deployed Engineers indispensable at Palantir.

AI adoption rarely fails because the model is not capable enough. It does not go through because employees don’t know where the technology fits within their daily work.

2. Supported human integration for smoother AI adoption

General AI models are horizontal by design. They struggle to generate specific materials and in-depth data analysis tailored to a single industry, since they were built to serve every industry at once.

Firms that engage with vendor partners see successful integration because they receive industry-specific guidance and training that empowers staff to firmly handle AI tools.

In general, people are hesitant when encountering new tools and processes. Leaders worry about governance, security and regulatory compliance. Employees wonder whether they can trust AI-generated outputs.

On the individual level, teams experiment with consumer AI tools, creating fragmented workflows and introducing the growing problem of “shadow AI”, where employees expose sensitive company information through personal AI accounts.

Without governance, training and clear operational guidance, organizations risk creating thousands of disconnected AI users rather than a coordinated AI strategy.

Technology vendors make a difference by collaborating with product managers, financial controllers, and cross-functional engineering teams.

Together, they design an integrated system architecture that includes existing technology, develop governance frameworks that reduce risk, and create education programs and processes that support positive engagement across a company. This ultimately generates trust and safety, ensuring a successful adoption.

Vendors provide the missing layer between technological possibility and organizational reality.

3. From one-off projects to long-term implementation

AI adoption does not have a finish line.

The technology almost changes every week. New models emerge. Capabilities improve. Costs change. Workflows that were impossible six months ago become routine today. That means AI cannot be treated like a traditional ERP rollout or cloud migration.

Technology vendors are uniquely positioned to provide continuous support. They know the ins and outs of the product and understand the challenges faced by organizations, which means they can tailor the approach to find the right solutions. Plus, they monitor the rapid pace of AI innovation while understanding how those advances affect specific industries.

Instead of trying to interpret an increasingly complex AI landscape themselves, teams can rely on vendors to get a clear picture of technical breakthroughs and how they can be relevant to the business and employees.

4. Sell business outcomes, not a model

The vendors who only sell AI models are selling half the answer. AI adoption fails when teams are left to figure out the product on their own, with no one in the room translating the platform into the firm’s own workflow.

Every organization will eventually have access to powerful models. What will differentiate them is their ability to combine AI expertise with industry expertise.

Technology vendors who succeed will be the ones who live between the technology and the profession, translating AI capability into trusted, scalable business outcomes. Accountancy Engineers, and every version of that role now emerging across other industries, is what that translation looks like in practice.

We’ve featured the best AI website builder.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit