Running Restaurants Founder Jaime Oikle speaks with Stephen Klein, co-founder and CEO of Saivory, about how AI is transforming the restaurant industry. Stephen explains how Saivory helps restaurants reduce reliance on costly third-party platforms like DoorDash by improving first-party ordering experiences through natural language AI prompts. The conversation covers the importance of owning customer data, personalized marketing, AI-driven smart landing pages, and an agent AI platform that delivers real-time marketing recommendations. Stephen also discusses the shifting customer discovery landscape and the need for restaurants to future-proof their digital presence.

Watch

Listen

Or Read – the interview was condensed to an article format with key highlights. Be sure to listen or watch for the full interview.

AI, Ordering, and Customer Relationships: Tips for Restaurant Operators

AI is already changing how guests discover restaurants, how they place orders, how catering inquiries are handled, how offers are personalized, and how operators build direct customer relationships.

That was the center of a recent Running Restaurants podcast conversation with Stephen Klein, co-founder and CEO of Saivory, an AI-powered restaurant technology company focused on first-party ordering, customer engagement, and smarter digital experiences.

The big takeaway for independent restaurant operators is simple:

AI is not valuable because it is trendy. It is valuable when it helps restaurants get more direct orders, reduce friction, capture better customer data, and bring guests back more often.

That matters because the restaurant industry has a painful digital problem.

Too many operators are renting customers from third-party platforms instead of building relationships they actually own.

The Third-Party Problem

Third-party ordering platforms can be useful. They offer reach, convenience, and visibility. For some restaurants, they bring in orders that may not have happened otherwise.

But there is a major tradeoff.

The restaurant often gives up margin, customer data, brand control, and long-term relationship value.

That is a serious problem.

When a guest orders through a third-party platform, the platform usually owns the customer relationship. The restaurant gets the order, but not always the information needed to remarket, personalize offers, or build loyalty.

That may seem acceptable for a few incremental orders. But over time, it can weaken the business.

Independent restaurants need to ask hard questions:

How many orders are we getting directly?
How many are coming through third parties?
How much margin are we losing?
How much customer data are we failing to capture?
How many repeat visits are we leaving to someone else's platform?

The goal is not necessarily to abandon third-party platforms entirely. The better goal is to make sure your direct ordering channel is strong enough that you are not dependent on them.

First-Party Ordering Is About Control

First-party ordering simply means guests order directly from your restaurant, usually through your website, app, branded ordering page, or catering portal.

That control matters.

Direct ordering gives operators several advantages:

• better margins
• ownership of customer information
• stronger loyalty opportunities
• more control over the ordering experience
• better ability to promote catering, bundles, and high-margin items
• more freedom to communicate after the order

This is especially important for catering and larger group orders.

A guest ordering lunch for four people may tolerate a basic ordering flow. A guest planning food for 40 people needs more guidance. They need portioning help, menu suggestions, dietary accommodations, delivery clarity, and confidence that the order will be right.

That is where AI can become very useful.

AI Can Make Ordering Easier

One of the most practical AI use cases in restaurants is simplifying complex ordering.

Think about a catering customer who says:

"I'm hosting a party for 40 people, and 10 are vegetarian."

A traditional ordering system forces that customer to calculate quantities, guess portion sizes, scroll through menus, compare options, and hope they did not under-order or over-order.

An AI-powered ordering experience can take that plain-language request and recommend a cart.

That is powerful.

It reduces decision fatigue. It helps the customer feel more confident. It can suggest the right mix of entrées, sides, desserts, beverages, and add-ons. It can also account for dietary needs, order history, popular combinations, and event type.

For the restaurant, that can mean:

• fewer abandoned carts
• higher average order values
• better guest satisfaction
• fewer order mistakes
• more repeat catering customers

This is the kind of AI that matters. Not gimmicks. Not novelty. Real friction removal.

First-Party Data Becomes a Real Asset

The best restaurant marketing starts with knowing who your customers are.

That is why first-party data is so important.

When customers order directly, join your loyalty program, fill out a catering form, respond to an offer, or interact with your website, you can begin building a more valuable customer database.

That database can include:

• name
• email
• phone number
• order history
• favorite items
• frequency of visits
• catering occasions
• event size
• location
• dietary preferences
• offer response
• average spend

This information helps restaurants stop blasting the same message to everyone.

Instead, they can start communicating with more relevance.

A catering customer should not receive the exact same message as a weekday lunch regular. A family meal customer should not receive the exact same offer as a corporate office buyer. A vegan customer should not be buried in a promotion that highlights only meat-heavy menu items.

Better data creates better marketing.

Personalization Turns Data Into Revenue

Personalization is where AI can help operators move from generic marketing to smarter marketing.

Examples:

• Send lunch offers to people who usually order lunch.
• Promote catering to customers who have placed large orders before.
• Remind past catering buyers before major holidays or office events.
• Feature plant-based items to guests who have ordered vegetarian meals.
• Promote family bundles to guests who typically order multiple entrées.
• Offer a comeback promotion to guests who have not ordered in 60 or 90 days.

This is not just "nice to have" marketing. It is a direct revenue opportunity.

The more relevant the message, the more likely the guest is to act.

The mistake many restaurants make is treating all customers the same. AI and better customer data make it easier to segment, personalize, and follow up in ways that feel more useful to the guest and more profitable to the restaurant.

The Discovery Journey Is Changing

Restaurant discovery is changing fast.

Guests still use Google, Yelp, Instagram, TikTok, TripAdvisor, maps, and word of mouth. But AI assistants and voice-driven search are becoming part of the discovery process too.

People are asking more specific questions:

"Where can I get catering for 30 people near me?"
"What restaurant has vegan options and outdoor seating?"
"Best casual dinner spot near the theater."
"Where can I order lunch for an office meeting tomorrow?"

This creates a new challenge for restaurants.

If AI tools and search engines are giving direct answers, your restaurant needs to be structured in a way that these systems can understand.

That means your digital presence must be clean, complete, and consistent.

Operators should pay attention to:

• Google Business Profile
• website structure
• online menu accuracy
• ordering links
• catering pages
• location pages
• reviews
• hours
• dietary information
• schema or structured data
• social profiles
• third-party listings

This is no longer just SEO. It is visibility across the new customer journey.

Smart Landing Pages Can Improve Conversion

A strong restaurant website should do more than look good.

It should convert.

One interesting idea discussed in the AI ordering and marketing space is the use of smarter landing pages. These are pages that match the guest's intent more closely.

For example, someone searching for "office catering near me" should not land on a generic homepage where they have to hunt for catering information.

They should land on a page that clearly answers:

• Do you offer catering?
• What types of groups do you serve?
• What are popular packages?
• How do I order?
• How much notice is needed?
• Can you handle dietary needs?
• Is delivery available?
• How do I get help?

The same logic applies to family meals, private events, vegan options, holiday ordering, brunch, happy hour, and online ordering.

AI can help generate, test, and improve landing page content. But the strategy still needs to be operator-led.

The goal is simple: reduce the number of steps between interest and order.

Agent AI Could Help Operators Make Faster Marketing Decisions

One of the more interesting emerging areas is agent-style AI.

Instead of simply answering questions, these systems can analyze data from multiple places and recommend actions.

Imagine AI reviewing:

• website traffic
• online ordering behavior
• ad performance
• search trends
• customer segments
• menu performance
• email results
• local competitor activity
• POS sales

Then it might recommend:

"Your lunch traffic is down on Wednesdays. Promote the office lunch bundle to customers within three miles who previously ordered between 11 a.m. and 1 p.m."

Or:

"Your catering page is getting traffic, but conversion is low. Add package pricing, delivery details, and a shorter inquiry form."

Or:

"Customers who order this entrée often add this side. Feature that pairing in your online ordering flow."

That kind of intelligence can be valuable because most operators do not have time to dig through multiple dashboards.

The danger, of course, is blindly trusting recommendations.

AI should support decisions, not replace judgment.

The best operators will use AI to surface opportunities faster, then apply real-world restaurant sense before acting.

What Independent Operators Should Do Now

This can all sound complicated, but the starting point does not have to be.

Here is a practical action list.

• First, audit your direct ordering experience. Try placing an order on your own website from a phone. If it is slow, confusing, ugly, or hard to use, fix that before chasing advanced AI.
• Second, review your customer data capture. Are you collecting emails, phone numbers, order history, catering inquiries, and loyalty signups? If not, you are letting future sales disappear.
• Third, improve your catering and large-order flow. Make it easier for guests to get help, understand portions, and place confident orders.
• Fourth, clean up your digital presence. Your website, Google listing, menu, ordering links, hours, and photos should all be accurate.
• Fifth, start using segmentation. Do not send every offer to every guest. Match messages to behavior.
• Sixth, use AI tools to speed up marketing execution. Draft email campaigns, create subject lines, build landing page copy, brainstorm offers, and analyze customer segments.
• Seventh, keep the human touch. AI can help with ordering and marketing, but hospitality still wins the relationship.

The Big Warning

AI will not save a restaurant with weak fundamentals.

If the food is inconsistent, the service is poor, the website is broken, the menu is confusing, or the operation cannot execute catering reliably, AI will only expose those weaknesses faster.

Technology is an amplifier.

It can make a strong system stronger.
It can also make a weak system look even messier.

So operators should not think of AI as magic. They should think of it as leverage.

The right technology helps good restaurants become easier to find, easier to order from, and easier to return to.

Final Thought

AI is going to reshape restaurant ordering, marketing, and customer relationships.

But the real issue is not the technology itself. The real issue is ownership.

Who owns the customer relationship?
Who owns the data?
Who controls the ordering experience?
Who gets to invite the guest back?
Who benefits from the repeat order?

Independent restaurant operators need to pay close attention to those questions.

Third-party platforms may still play a role, but restaurants cannot afford to build their digital future entirely on rented relationships.

The stronger path is to build better first-party ordering, capture better data, personalize communication, simplify the guest experience, and use AI where it creates real value.

Because the restaurants that win in the next stage of digital hospitality will not simply be the ones using AI.

They will be the ones using AI to build stronger direct relationships with their guests.