A multi-ethnic team of developers meets to discuss marketing data analysis.
getty
companies have invested 30 billion to 40 billion dollars Although generative AI has seen increased development over the past two years, only 5% of AI pilot projects are producing an identifiable ROI. According to MIT’s Project NANDA, 95% of the money companies invest in AI is invisible from a P&L perspective. For a small business owner investing $200 a month in an app or service with the expectation of self-funding, this could be worse than a CFO at a Fortune 500 company. Developing a viable small business AI plan starts with understanding your business. Most of the reasons small businesses fail to implement AI are determined by how they operated before AI.
Leadership hasn’t caught up with the tools.
New research published on July 22, 2026. Manpower Group’s human resources solutions We came to the same conclusion as other studies. The survey was developed by Everest Group and includes responses from 80 executives and senior talent leaders. Only 3% of those surveyed reported that their current leaders are well prepared to lead an AI-enabled team. In addition, McKinsey’s 2025 State of AI Survey Almost the same results were obtained. Only 1% of companies say their organization’s AI strategy is fully mature. Today, 88% of organizations have some level of AI in their business, but leaders aren’t yet ready to take advantage of it.
boston consulting group 70% of AI transformation initiatives in business organizations report failing to meet leadership expectations for results. They identified organizational culture, rather than technology itself, as the root cause of this gap. According to BCG, organizations that devoted at least 10% of their AI budget to training and change management were 1.5 times more likely to succeed than those that did not. Rather than developing new processes, AI enhances existing processes by executing them faster than humans can write or input data.
The successful 5% do things differently.
The majority of the 5% who successfully used AI process automation identified a pattern in their implementation. The process automation pilots these companies conducted were limited to processes that repeated daily. These include already established, repeatable processes such as processing invoices, setting schedules, and completing back-office functions. Broad-based chatbots may seem flexible on the sales page, but they are difficult to implement in the average company because they lack the ability to understand or adapt to the company’s specific workflow. Using the same type of AI-based software systems used by Fortune 500 companies is no substitute for creating the workflows needed to use this technology effectively. Therefore, AI implementation for small and medium-sized businesses can only deliver a significant return on investment if the workflow is documented upfront.
Founder dependency issue
Most small business owners ignore this step. According to a study conducted by National Association of Insurance Commissioners (NAIC) On the role of founder dependence in small and medium-sized enterprises. 71% of small businesses rely on one or two individuals for their overall success. The NAIC study was referenced in the national “Key Person Risk” study. When lenders consider the potential risks associated with acquiring a small business, they assess whether the company has sufficient controls, processes, and redundancies for day-to-day operations in the event the founder leaves.
Small Business Administration (SBA) lenders perform their own dependency assessments when considering acquisition transactions. Even if a business passes an SBA lender’s dependency assessment, there is no guarantee that the business will receive the full amount of funding. In return, the lender may offer you less money than you would expect under normal circumstances. Additionally, lenders may require a longer transition period for founders to remain on payroll, during which lenders may also include life insurance coverage to ensure business continuity if the person responsible for running the business leaves.
AI adoption readiness: What separates the 5% from the 95%
Source: Business AI Research Institute
A founder who bridged the gap
Anthony Godley is keenly aware of that gap. He single-handedly founded Logix BPO. In the early days, he had just one client. Godley, who became chairman in 2025, was able to grow Logix BPO into an outsourcing company that employs more than 1,000 people. “The biggest barrier to AI adoption is not technology; it’s reliance on founders. If all important decisions still come back to founders, AI will only expose bottlenecks faster,” Godley said in a written statement.
“They already have a disciplined process and a leader who doesn’t have to make all the decisions themselves,” Godley said. AI increases operational maturity. Make the decision first, then use the software. Document all approvals and escalations you need. Then document how each approval is done. Document every “judgment” call that a spreadsheet or AI could never make. Documenting this workflow provides sufficient context to the AI and helps take some of the work out of your own workload rather than producing additional output for validation. To determine the ROI of implementing an AI system in your small business, it’s important to start by documenting the actual process that produces quantifiable results.
Start with decisions, not software
Before considering purchasing your next AI product, ask yourself three questions. Who else in your organization has the authority to make decisions about how to use AI? Is your definition of success when using AI for this task documented somewhere for people other than you to read? Will your processes continue to work if you go out for a week? If your company can’t answer these questions right now, an AI subscription will help you do so better. A subscription just gives you more tools for a longer period of time to avoid these questions.
Identify the least well-documented people in your company on the most important tasks and cross-train them on those tasks by the end of the month. Not because the AI vendor said so, but because companies that consistently achieve failure rates above 95% knew who else could do this before they bought their first AI seat. Once you have identified that person and answered this way, it is time to identify how to measure or determine the AI return on investment (ROI) for your small business. Cost and purchase of purchased tools It will be the easiest part of the project, not the whole project.
As long as the 95% failure figure continues to be cited as a problem with AI, it will continue to be treated the wrong way. This is a leadership issue with an AI price tag. Companies in the remaining 5% don’t necessarily have the best algorithms. They have already built their business on a system of writing, not just one person’s personal experience, and their process-oriented culture has enabled their ability to successfully implement AI. Implementing AI was a simple part they decided to add at the end.
