How Operations Leaders Are Integrating AI in 2026 Without Risky Bets or Delayed ROI

It's 2026, and the competitive landscape in automotive, mobility, and manufacturing has shifted. Peers are pulling ahead, not through massive overhauls, but by methodically weaving AI into core processes. They're cutting routine work by 20–30%, spotting risks earlier, and freeing teams for strategic moves that win business. The pressure is building for executives to get AI up and running as fast as possible without massive overspending.

But the real pressure isn't coming from the board demanding instant transformation. It's the directive from the market: Get moving or your competition will.

You want to deliver measurable gains without expensive solutions that drag on for months, expose data risks, or force awkward process changes. Get it wrong, and you're explaining delays, budget overruns, or underused tools while competitors keep advancing.

I've watched the market navigate this crossroads. Some chase comprehensive enterprise platforms hoping for quick wins only to hit integration hurdles that stall progress. Others start smaller enhancing tools already in daily use, building momentum slowly while reducing risk.

This article cuts through some of the noise, providing an overview of how successful executives are using AI today with their tools. Done thoughtfully, this delivers real efficiency without the pitfalls.

Common Paths Forward — And Where They Falter

Leaders today have options:

• Enterprise platforms (Copilot suites, Einstein add-ons, full ERP AI layers): Powerful when your data aligns perfectly and scale justifies the investment.
• Pure DIY: Building everything custom from the ground up. Big learning curve and slower, but with higher possible up-side in the end.
• Hybrid: Starting with low-risk enhancements to existing tools, layering on more as proof builds.

The enterprise route can promise broad coverage, but mismatches are common. Data formats don't line up, requiring heavy cleanup and customization. Security policies clash. Extensive training is needed to get teams up to speed on new processes. Months pass before value emerges, and teams revert to old ways in the meantime.

Pure DIY demands deep technical resources or a lengthy learn, trial, and error process that few companies can stomach in the short run. However, it can lead to perfectly customized solutions that meet the exact needs of the business if done correctly.

The hybrid approach many are succeeding with: Begin where the impact is immediate and control is highest in tools like Excel or note apps. Then, keep doors open for larger integrations later. This builds quick wins, proves concepts internally, and informs smarter decisions on bigger investments.

Regardless of the path chosen, all processes should start in the same place: Data security.

Security and Fit Come First — Always

Any path starts here. Proprietary data including customer projections, cost structures, supply chain details, and even names can't flow unchecked into third-party models.

You should always assume that anything put into a large language model (LLM) AI program could be monitored, stored, and used for training by the third-party owner of that model. Be sure to consult with a security expert about your individual needs, but here are some ways to integrate AI safely with your data:

• Anonymization: Replace identifiers with generics before any external query.
• Strict boundaries: Use AI for high-level reasoning or code generation, not raw data analysis.
• Controlled environments: Favor local tools or vetted APIs with clear data policies (again, consult with an expert to ensure proper compliance).

Get this wrong, and risks outweigh gains. Get it right, and you unlock safe, tailored advantages.

Practical Starting Point: Enhancing Excel with AI-Assisted Automation

One very common starting place is a tool widely used and understood: Excel. Excel handles most forecasting, reporting, and scenario work in operations and almost everyone has used it. Its VBA macro capability has been there for years, automating repetitive tasks precisely. It already has a lot of functionality built into it that make it a perfect starting place to implement AI.

The best way to do this is to use AI to help you code custom macros that automate the cleaning, filtering, and structuring of raw data that lives in your systems. AI can help you create a system that, with one click of a button, prepares all data for you so you can move straight into analysis.

A process I've seen deliver consistent 20–30% time savings on monthly cycles:

1. First, make sure you’re working on a copy of the data set — preserve the original in case of mistakes.
2. Break software macro tasks into small steps. For example, maybe first you want to remove extra data. Write that code, ensure it works, and then move on to the next task building one step at a time verifying as you go. This reduces mistakes and builds understanding of the software. Go back to the original data set each time when testing to evaluate the whole macro works correctly.
3. Prompt clearly with the AI: "Write VBA to flag rows where margin falls below 15% and calculate year-over-year variance."
4. Review the code: Ask for line-by-line explanations to understand and refine.
5. Test thoroughly. Pull in another data set to check. For example, if you are running monthly reports, pull in the data from the previous month and run the macro on that data to verify the code works across different data sets.
6. Add next layer: Once the macros are working, connect the data to charts, sensitivity tables, outlier alerts, and more to create dynamic elements for analysis.
7. Break down macros into separate buttons as needed to accomplish specific tasks.

Doing this, I have been able to generate reports and graphics that rivaled (and sometimes beat) our enterprise software like our CRM program.

Another Reliable Win: Turning Conversations into Actionable Insights

Meetings and customer calls generate gold, but capturing details without disrupting conversation flow is tough. No one wants to stop a discussion to jot down notes, and no one wants to hold a recording device or notepad when talking to a colleague.

Instead, do this right after the conversation:

• In your phone, start an email or note file and use the speech-to-text function to dictate everything you remember about the conversation and anything else that comes to mind.
• Cover key points, commitments, implications, and more. Include action items and follow up tasks in the notes as well.
• Then, email the message or note file to yourself for analysis. If using a third-party AI program, strip out proprietary data and ask it to summarize and structure the notes.

The result is a complete and professionally structured recording of every meeting you have. Ideally, you structure the notes so they can be directly uploaded into a CRM system. With this simple routine, administrative drag drops sharply and focus stays on building relationships.

Building Toward the Right Long-Term Fit

The key theme with these strategies: they get quick results that generate real returns for the business without having to dive into expensive solutions. Once you establish a foundation of working process, you’ll be better positioned to see what gaps you have and the best ways to fill them whether it be a full enterprise solution, a DIY approach, or a hybrid solution.

If you want more detailed information including a step-by-step process to streamline any business process with technology that mirrors this approach, my book Business Operations Unlocked: 7 Steps to Slash Costs and Streamline Processes with Technology covers these topics and much more including hardware solutions, ideal IT structures, automation tools, and more.

Grab the free preview chapters and get notified at launch (coming soon, before February closes):

https://www.jasonunlocked.com/pages/bou-prelaunch

For weekly strategies like this, join the newsletter:

https://www.jasonunlocked.com/pages/subscribe

Have any tips or stories to share? Reach out and let me know how you’ve been able to take on this challenge today.