Partner Story · a contributed guest article by Gen Gacer, founder of Leiva Assistants — published through our free Submit Your Story program. The argument and examples below are the author's own, edited for clarity and style. No payment was involved.
By Gen Gacer, founder of Leiva Assistants
AI has created a strange pattern inside growing businesses. A team runs into a bottleneck, and the first response is to shop for software. Reporting takes too long, so they buy an AI analytics tool. Leads slip through the cracks, so they add an AI sales assistant. Customer response times are slow, so they introduce a chatbot. A few weeks later the company has another subscription — and the original bottleneck is still there.
I see this regularly in my work with founders and remote teams. In most cases the technology is not the problem. The work itself has never been clearly organized. Responsibilities overlap, instructions live in someone's head, and exceptions are handled differently depending on who is online. Putting AI on top of that does not create efficiency. It simply gives the confusion somewhere new to go.
That may explain part of the gap between AI adoption and actual business results. According to McKinsey's 2025 global survey, 88% of respondents said their organizations were regularly using AI in at least one business function — yet only 39% reported an enterprise-level impact on EBIT, and most were still stuck in pilot mode. IBM's 2025 CEO study found a similar gap: only about 25% of AI initiatives delivered the expected return, and just 16% had scaled across the enterprise.
Businesses are clearly using AI. The harder part is turning that use into money saved, time recovered, or revenue gained.
What Sundial gets right
What stood out to me in AI Business's recent profile of Sundial was not the technology. It was the problem the company had chosen to solve. Sundial focuses on decision latency — the time lost while a business waits for analysts, reports, and dashboards to produce a usable answer. Its AI runs on expert-built analytical playbooks and ships evidence alongside its conclusions.
That matters because speed alone is not enough. A faster answer has little value if no one trusts it or knows what to do with it. There is also a recognizable workflow behind the product: a business question comes in, the relevant data is examined, the findings are backed with evidence, and someone uses that to make a decision. The AI has a defined role inside the process. It is not being asked to compensate for the absence of one.
The ROI starts before implementation
Before automating a workflow, a business should know what that workflow currently costs.
Take weekly performance reporting. Suppose two employees each spend five hours gathering figures, checking spreadsheets, and preparing a report. At a loaded labor cost of $40 per hour, that process costs roughly $20,800 per year. If an AI-assisted workflow cuts the work from ten hours to three hours per week, the potential annual labor saving is $14,560. Now assume the software costs $3,600 per year and implementation another $4,000. The first-year net benefit lands at $6,960 — a return of roughly 92%.
That sounds worthwhile, but only if the redesigned process actually works. If employees still have to correct inconsistent data, confirm which spreadsheet is current, and chase managers for missing information, much of that saving evaporates. The AI generated the report quickly, but the company never solved the workflow around it.
This is why I would not judge an AI initiative only by how fast it produces an output. I would also look at the hours of human review required, the number of corrections made, the time between request and completion, and whether the output leads to a faster or better business decision.
Ownership matters more than another feature
Customer support is a good example. An AI system can categorize incoming messages, spot common issues, and prepare suggested responses — real time savings. But someone still has to decide when a refund is appropriate, which cases require escalation, and who is accountable when the AI recommends the wrong thing. Without those rules, the company has automated the easiest part and left the expensive decisions unresolved.
So before adding AI to any process, I want to know: What starts the work? What information does the system need? Which decisions can the AI make? When must a person step in? Who owns the final outcome? How will the business know whether the change worked? If a team cannot answer those questions, it probably is not ready to automate that process. It is the same instinct behind starting with your services, not your internal functions — describe the work before you hand it to a machine.
Start with one boring workflow
Founders do not need a complete AI stack to see a return. They need one frequent, expensive, measurable problem — support-ticket triage, lead qualification, weekly reporting, meeting follow-ups, content repurposing. Watch how the work is done for a week or two. Record how long it takes, where the delays happen, and how often someone has to fix a mistake.
Then simplify the process before introducing AI. Remove unnecessary steps. Decide who owns the result. Document the rules and the exceptions. Run a focused 30-day pilot, then compare the outcome with the original baseline. Did it save real labor hours? Did it cut response time? Did it improve conversion, capacity, or customer experience? Did the saving exceed the cost of the software, implementation, and oversight? If yes, scale it. If not, fix the workflow before buying another tool.
AI can create meaningful leverage, but it cannot define accountability or decide what a successful outcome looks like. That work still belongs to leadership. The businesses that get the most value from AI will not be the ones with the most tools. They will be the ones that give those tools clear, well-designed work to do.
About the author
Gen Gacer is the founder of Leiva Assistants, where she helps founders and entrepreneurs build sustainable operations through effective delegation and structured remote-team support. Her work focuses on reducing founder dependency and improving execution as businesses scale, and on creating stable, growth-oriented careers for remote professionals. Leiva Assistants is an operations and remote-staffing partner that supports founders and growing teams with administrative, operational, and execution-focused roles — helping them scale without adding unnecessary complexity or overhead.
---
This is a contributed guest article published through AI Business's free Submit Your Story program. The views, figures, and examples are the author's own and reflect her perspective; they are not verified claims or endorsements by AI Business, and no payment was exchanged for publication. Want to contribute? Submit your story.
