Doug Huber, principal at Foodservice Consultants Studio, believes foodservice consultants shouldn’t wait on the sidelines when it comes to artificial intelligence. “It’s a big body of knowledge and moving so fast that you’ve just got to jump in and start paddling around,” he says. “You might be uncoordinated at first, but you can even tell your AI who you are and ask how you can use it.”

Doug HuberHuber encourages consultants to treat AI, or large language models (LLMs) like ChatGPT, Microsoft CoPilot, Gemini, Claude and NotebookLM as collaborative tools. “It’s not about replacing humans — it’s about making humans superhuman at what they’re doing,” he says. “Until you start interacting with AI, you don’t really know what it can do. It has a very large base of knowledge that we as humans can tap into and goes beyond just searching things in Google, where you have to go down the rabbit hole yourself.” He recommends starting with an online course like Google’s AI Essentials to learn prompting basics, ethics, and how to “keep humans in the loop,” as the term’s being used in the context of AI.

Next, explore the different LLMs and some prompt engineering. “Give the prompt a persona, a task, and some specifics, like whether you want it concise or a deep dive,” Huber adds.

Once you’ve settled on your preferred LLM or LLMs, here are some ways Huber suggests using them.

Conduct Basic Research and Strategic Framing

LLMs are great for quick research when preparing for a pitch or learning about a client or institution. “You can ask AI: ‘I’m joining the team for this university project — how does my proposal fit in with their marketing plans, facility goals and overall strategic vision?’” Huber says. “That context helps you position your consulting role more effectively.”

Huber offers a sample prompt for that: “I’m a professional FCSI consultant designing a kitchen for XYZ University. Please do a marketing analysis: what are 3 to 4 key differentiators in their positioning and strategy?”

Then upload PDFs, master plans, URLs and other documentation to the LLM to train the model specific to that project. To start tailoring your prompts even more to your style and business, Huber recommends uploading any and all marketing documents, templates, sample proposals you’ve created for your company or firm so it can “learn” more about you.

Summarize Long RFPs and Scope Documents

It’s possible to feed LLMs large documents and have them summarize and pull out the key points. “I once had a 110-page RFP that would’ve taken six or seven hours to properly review. AI pulled out all the foodservice-relevant content in minutes. I reviewed the output, made some adjustments and had my scope of work defined in under an hour,” Huber says.

While Huber doesn’t always use AI to write his actual proposals because his firm has templates set up for that, he’ll use it in more step-by-step ways to get close to the proposal writing level. He’ll use AI to summarize the scope of work and then create an outline of a proposal draft that he can build upon.

Prepare for Meetings or Client Calls

LLMs can also act as meeting coaches and help consultants prepare for meetings with architects and clients and more. “The goal is to customize your offer to the specific needs of the architect or end user,” Huber says.  “I tell it: ‘you’re a foodservice consultant preparing for a client meeting—what are the top 10 questions I should ask?’”

Huber can then upload the meeting invite, client background, presentation deck, and ask AI to suggest additional talking points or refine your delivery.

Solve Business Problems and Improve Strategy

Huber’s used his well-trained AI tools as a business coach, primarily using the classic 1-3-1 technique. “I asked AI to first identify the issue or problem, give me three possible suggestions and recommend one approach — and explain why.”

“I’ll upload details about my firm, past proposals, success rates, even competition—and ask AI to blend strategies or create a new one,” Huber says. He then stores this knowledge as an “artifact” inside his chosen platforms — essentially as a live document with everything AI has generated for his firm.

Eventually you can train your AI models to start thinking or making suggestions for you, as you would — though you always want to check its work, just like an intern or beginner associate. Huber says, “I once recorded a conversation with a client about a refrigeration rack system, loaded it to NotebookLM, had it summarize it, and then uploaded drawings and all the information I had for the project and started asking it questions like ‘how do you suggest I rough this in?’ and ‘how many circuits does this require?’”

Bottom line: Don’t be afraid of AI; learn how to embrace it. But, Huber says, “be polite — don’t forget to say ‘please’ and ‘thank you,’ it will work better for you!”