Close Ad

The Spoon

Daily news and analysis about the food tech revolution

Halla

Halla’s Spencer Price: Grocers Will Create ‘Unique Grocery Store for Every Shopper’ in The Future

Next up in our Smart Kitchen Summit speaker preview series is Spencer Price, the founder of Halla.

Halla has built an AI personalization and recommendation platform for grocery store providers. According to Price, the turning point for his company and the broader grocery store industry was when Amazon acquired Whole Foods.

“When Amazon acquired Whole Foods in 2017, it sent grocers into this innovation frenzy,” said Price. “I think the main driving force for grocers to want to look at this type of tech back then was that Amazon generates over one-third of all of its product sales revenue from their recommendation with the ‘You may also like’ and ‘customers also bought’ type product suggestions. Grocers do not have a passive piece of AI that drives a third of their sales, and that is what we enable grocers to do. We give them that competitive weapon to fight back in this World War grocery.”

Price thinks that while the grocery industry is lagging behind other industries, such as entertainment, when it comes to personalization, they are looking to AI to make up ground.

“Netflix isn’t just on a one-account basis. Within an account, you have a handful of profiles in your household, and each profile sees a completely different set of suggested categories, titles within those categories, and even different cover art for each one of those titles that’s likely to resonate with you as a specific end user.

“Grocers are a little bit behind these content platforms, but I think in 10 years time, we will see a very similar thing, and it’s going to be even more exciting because if you can give every single shopper their own unique grocery store, that’s going to make for both the fastest and most efficient and of course, most inspiring shopping experience. Grocers want to move quickly.”

Price’s company was acquired by Wynshop in March. Price says the company brought over his entire team and that Halla remains an independent business unit within Wynshop.

The Spoon Talks With Halla's Spencer Price About AI's Impact on the Grocery Business

You can hear Price speak at Smart Kitchen Summit on June 4-5th in Seattle. Get your ticket today!

You can read the full transcript of our conversation below:

Michael Wolf: All right, I have with me Spencer Price, the CEO of Halla, now a part of Wynshop. It’s been a while since we caught up. We wrote a first article about you guys since then, and you guys have changed a lot since then.

Spencer Price: We have changed a lot since then, yes.

Michael Wolf: At that point, you were very much focused on being a personalized recommendation platform based on a lot of different data. I still think that’s a lot of what’s pretty true, but you guys did evolve since then.

Spencer Price: Yeah, so 2018 was a transitional period. We had developed, as you said, a personalized recommendation engine centered on food and beverage products. And we had a mobile app that would recommend restaurants and even specific dishes from those menus to users or groups of users with varying taste preferences, dietary restrictions, et cetera. And that was 2017 to 2018.

When Amazon acquired Whole Foods in 2017, it sent grocers into this innovation frenzy. There was a demand for us to gut the tech from the app, license it B2B, and we ended up sunsetting the mobile app, which feels like a lifetime ago now. And all we’ve done is deploy personalized recommendations, search and substitutions for online grocers ever since.

Michael Wolf: I didn’t know that that had such a big impact. It makes sense, in retrospect, the acquisition of Whole Foods by Amazon. But like you said, there was this frenzy and a wake up call to existing grocers, and that sent you in a completely different direction.

Spencer Price: Exactly. We had some innovative nimble online grocers as well as some legacy retailers that knew they needed to step up. I think the main driving force for grocers to want to look at this type of tech back then was that Amazon generates over one-third of all of its product sales revenue from their recommendation with the ‘You may also like’ and ‘customers also bought’ type product suggestions. Grocers do not have a passive piece of AI that drives a third of their sales, and that is what we enable grocers to do. We give them that competitive weapon to fight back in this World War grocery.

Michael Wolf: I love that; World War grocery sounds like a movie, starring you guys apparently. But I mean, look at the last 18 months, right? I think the world’s woken up to AI. It’s permeated all the press and the pop or consciousness largely due to the exposure of things like ChatGPT and generative AI to everyone. It seemed like like six, seven years ago, a lot of people were building ontologies and had a custom code, to make their AI to get certain outputs. But now, with generative AI, you can basically do prompts and get a lot of the same results. And these large language models just keep getting bigger. Can you talk about how your business has changed by incorporating larger langue models and generative AI?

Spencer Price: Yeah, so the way that generative AI has taken shape thus far has, of course, been through chat bots. One of the things that those, at least from a consumer-facing standpoint, one of the things that chat bot ask technology with LLMs, Gen. AI, et cetera, plays into e-commerce at large and potentially grocery down the line is conversational commerce.

We don’t see that as being a particularly exciting use case, particularly in this category where people are adding usually a couple dozen items to their basket. They’re not saying, you know, I need help finding the right sweater that matches these pants. It’s a household you’re shopping for with different dietary restrictions, taste preferences. And that’s where language models don’t necessarily perform best. That’s where recommender systems have decades of tried and true proven methods.

And so that’s still a foundational component of our science. However, for one of our solutions, search, LLMs allow for a much more robust level of understanding natural language. So we had our own raw sort of NLP models that we developed in -house a few years ago, that we’ve been fine tuning, and now we can incorporate some of these open source transformers and LLMs to take our vertical eyes, rather than a generalist sort of assistant, our vertical eyes recommender systems and layer them with this cutting edge technology that allows for the generation of synonym lists and a better understanding of things like typos. But the risk with using just generative AI to try to develop these highly specialized models in a category that’s clearly so nuanced and personal is the hallucinations. I was recommended a beef and banana soup from chat GPT. And I got to tell you that that feels a ways away. I did not. It was terrifying to be honest.

Michael Wolf: Did you make it?

Spencer Price: I did not. It was terrifying, to be honest.

Michael Wolf: Well, I’ve been talking to a lot of folks who are in this area of food and beverage that are trying to deploy AI centric solutions. And like you said, a lot of the LLMs have this problem with hallucination. They’re oftentimes, they’re ingesting the world of the broader internet, but they don’t necessarily go deep on things like food and beverage. So I’ve heard companies that are building special, small language models that can couple into large language models. They’re doing kind of these transformers that provide the intelligence. Sounds like you guys have your own kind of approach to that. And you’re using LLMs as the conversational smart interface that is just so much more savvy than it would have been in the past. And then diving deep into your knowledge set.

Spencer Price: Precisely. We are using these new state of the art technologies, both as sort of a research platform to understand what we can benefit from and leverage and also where the watch outs are, like the example I just shared. One thing that you’d imagine might be really nice, whether it’s with a small language model specific to what we’re doing, or using the best of these large language models.

One use case that probably strikes you as obvious is groceries have a notoriously dirty data problem. And so maybe there’s a way to clean up these product catalogs and inventories and descriptions and attributes. The challenge is you can’t run the risk of things like health claims, nutrition facts, or marketing descriptions being completely wrong. And we’ve seen a lot of inaccuracies in using it for that.

So everything we do with LLMs has a human in the loop to make sure that none of those inaccuracies end up facing a user. But by and large, what sets us apart is layering in, as you said, our knowledge base, which is an ontology of every single product, but more than that, the essence of each product, knowing that orange, for example, is a distinct flavor. It is a product and it’s also a color. And LLMs are not built to have those nuances at play to the level of sophistication that you need them to be. Does that make sense?

Michael Wolf: Yeah, it does. What are you most excited about if, 10 years down the road, you’re building systems that use technology like AI in terms of the grocery shopping experience? What do you think will change the most?

Spencer Price: So I think that personalization historically took a lot of different shapes, and then they all kind of converged five to 10 years ago by having truly individualized browsing experiences on content platforms. Netflix isn’t just on an account basis, but within an account, you have a handful of profiles in your household, and each profile sees a completely different set of suggested categories, titles within those categories, and even different cover art for each one of those titles that’s likely to resonate with you as a specific end user. Spotify acquired Echonest, and they were able to map out all the different attributes down to subjective metrics like the danceability of every single track in their library, now they have the most robust music recommendation engine in the world, and people love them for that, and I’ve never left as a result.

In online shopping, we’re talking about products now, not content. We’re a little bit behind these content platforms, but I think in 10 years’ time, we will see a very similar thing, and it’s going to be even more exciting because if you can give every single shopper their own unique grocery store, that’s going to make for both the fastest and most efficient and of course, most inspiring shopping experience. And we’re not there yet, but we have all the rails to get there in a lot less than 10 years. Depends how much. Grocers want to move quickly.

Michael Wolf: That’s exciting, getting Mike’s grocery store tailored towards me. That’s perfect. Tell us about the Wynshop deal. You guys got acquired, which is exciting news for you. What does that mean?

Spencer Price: So our biggest channel partner to reach retailers and have our personalization technology directly embedded into an e-commerce platform was with Wynshop. And they’re the leading provider of e com platform technology on a white label basis to grocers all over the continent and a handful of international accounts as well. And we’ve been working with them for a few years. We love the team. We think they have a clearly differentiated product and they got to know us, our team and our tech. And it was just a pretty perfect match, to be honest, to have what we’ve developed baked in as more of a base level set of functionality, as well as being able to offer premium levels of functionality for these grocers that they can opt into if they want.

So yeah, about six weeks ago, we joined the team. They brought on all the day to day, all the personnel, we remain an independent business unit within Wynshop, but obviously it’s not like there’s any walls up. We work with everybody there very well. They put some resources behind us and yeah, the goal is both to service their existing accounts and future customers as well with the tech we’ve built and the new stuff we’re building.

Michael Wolf: All right, well, Spencer, congratulations. You worked hard for years to build the product and then create a opportunity for you. So I’m looking forward to talking more with you in Seattle in June at the Smart Kitchen Summit. How can people find out more about Halla and Wynshop?

Spencer Price: Yeah, well, thank you so much for the opportunity and the congratulations. You can still find us even though we don’t go by holla .io, we’re just holla now, at halla .io and winshop .com, W -Y -N, shop.

Michael Wolf: Cool. Hey, well, Spencer, thanks so much for spending time with me, man.

Spencer Price: Thank you so much, Mike. Look forward to seeing you in June.

Halla Raises $4.5M for Its Food Recommendation Platform

Halla, the AI-powered production recommendation service for grocery retail, announced today that it has raised $4.5 million in Series A1 funding, led by Food Retail Ventures. This brings the total amount raised by Halla to $8.5 million.

Halla’s platform integrates with a grocer’s existing digital commerce solution to provide customized product recommendations and substitutions for out of stock items to consumers. But Halla’s platform doesn’t just rely on previous purchases to make its recommendations. The company says it uses more than 100 billion shopper and product data points to predict what a shopper is looking for. Halla’s product video embedded below illustrates how Halla’s the system looks at all kinds of data about a shopper as it makes a recommendation in real time.

Grocery has been the beneficiary of a ton of funding this year, with $10 billion going into the sector as of July this year. Most of of the funding and attention has been around speedy grocery delivery services, but money has been doled out to startups working up and down the grocery stack. Hungryroot, and online grocer that uses machine learning for predictive recommendations, raised $40 million last month. In April, Trax raised $640 million for its computer vision-based inventory management system, and Shelf Engine raised $41 million for its perishable inventory management platform.

A big reason for all this money flowing into grocery is the pandemic. Fears around COVID-19 sent record amounts of people into online grocery shopping last year, creating new logistical and fulfillment issues for retailers. But the pandemic also highlighted flaws in our existing grocery supply chain, as evidenced by the panic hoarding and empty shelves that happened at the beginning of the outbreak. All of this is to say there are a bunch of new issues for grocery retailers to solve post-pandemic, which means plenty of opportunities for startups.

Halla said that a “top-5 U.S. grocer” is currently running Halla in more than 1,100 e-commerce storefronts. Halla will use the new funding to double the number of stores it’s in, and double its headcount by 2022.

COVID-19 Got Shoppers to Order Groceries Online, but Will They Keep Coming Back?

As COVID-19 has forced shoppers indoors, the growth curve of online grocery has suddenly accelerated, with downloads of popular grocery apps increasing by as much as 2000%. Shoppers who were previously hesitant to buy groceries online have found themselves now doing so out of necessity. It’s a bittersweet victory for online grocers who have long struggled to gain real traction.

But as the old adage says, “this too, shall pass.” The current crisis will subside, and when it does, it’s anybody’s prediction if grocery shoppers will retain their newfound affection for buying foodstuffs online. I believe that the outcome of that question will be determined by the kind of customer experience that online grocers deliver to shoppers now.

Here are three levels of effort that online grocers might want to focus on to make sure that experience will keep bringing shoppers back after the crisis has passed.

1. A functional platform and robust delivery capability is crucial

Online grocers with holes in their game need to scramble. They had better eliminate technical glitches, ensure that their back-end can support purchases in large volume, secure agreements with suppliers for gluts in demand, and validate that their pickup and delivery services are built to flexibly scale.

Even some very established players have seen troubles during this spike of new traffic. British online grocers like Ocado and others came face-to-face with this problem recently, as sites and apps repeatedly crashed under the weight of new users. Some ultimately had to turn new customers away, or create “virtual queues” just to use the service. Morrisons quickly updated its payment terms to make sure that deliveries from small suppliers would not be stymied by cash flow problems.

American retailers have rather famously faced logistics problems. Instacart, Amazon and others have been unable to meet their typical delivery commitments while Walmart, Target, and Amazon are all facing severe inventory problems on high-demand items.

While it’s understood that these are unusual times, online grocers should take a lesson and develop contingency plans for higher traffic. It’s a good time to take a hard look at both the technology and the support systems that you will need to service your new customers.

2. Make sure your shopping experience is easy 

Adoption of online grocery shopping to date has largely been driven by tech-savvy millennials. Provided that your online grocery shopping website and app have a half-decent user interface, these young shoppers will natively understand how to navigate them. But that may not be so true of those who have hopped on the bandwagon recently. Either way, ease and enjoyment of the online shopping experience can make or break loyalty to your online platform.

If your online shopping experience is not already as easy and smooth as it could be, now’s the time to change that. Products must be easy to find through search functions and intuitive shopping categories. Consider carefully how you are organizing product by category and sub-category, by brand, price, and even by lifestyle or dietary choices. Think about how to best present holiday and promotion items. If shoppers feel that it is difficult to find what they want, they’ll soon switch back to what they know.

Don’t neglect sign-up, check-out, and re-order, either. Think of online grocery as a means to “always have the customer in the store.” You can make shopping easier by proposing automatic delivery of common staples each shopper routinely purchases. The more seamless these routine functions are, the more customers will appreciate the convenience of shopping online.  

3. Make online shopping about personalization, imagination and discovery  

So what about online grocers who already have a functional platform that makes it easy for shoppers to get what they need? If you are one of these grocers, you are positioned to create not just a functional shopping experience, but an extraordinary one. Online grocery should not be just “a store on a website.” By making your online platform emulate something that it is not, you’ll miss out on making it something much better.

That something should include smart, personalized recommendations to customers. When a shopper puts fresh chicken breasts, bouillon cubes, and vegetables in their cart, you could recommend egg noodles for the chicken noodle soup they are making, but you could also suggest freshly squeezed orange juice or hot tea, honey, and ginger for whoever is nursing a cold. That vegan shopper who is always buying garbanzo beans and tahini will probably appreciate you suggesting the imported middle-eastern spices that just came in. And for that customer you know is looking to shed a few pounds, you could prioritize delicious, healthy foods that help him or her choose wisely. 

Henry Michaelson | Co-Founder, President & CTO at Halla 
While studying computer science at UC Berkeley, Henry co-founded Halla, a taste intelligence company that enables retailers to predict the personal preferences of their shoppers, all in real time. He is responsible for constantly improving Halla’s machine learning algorithm and for internal leadership, especially with respect to technology. Henry’s previous projects include machine learning based classification of supernovae in the UC Berkeley Astrophysics department, a speaking role in the Warner Brothers blockbuster comedy Project X, a three year stint as lead guitarist for Joe Banks, and a patented algorithm that has distributed over $7M in awards to mobile gamers.

Halla’s CEO is Trying to Netflix-ify the Grocery Aisle

Anyone who’s visited the cereal aisle of a grocery store knows that every time we shop for food, we’re bombarded with choice. Which product will be the best on for our tastes, our diets, and our family’s preferences?

That’s why I typically end up reaching for the same box of cereal (and jar of peanut butter, loaf of bread, etc.) every time. But some companies — like startup Halla — are trying to make grocery shopping a more individualized experience.

That’s why we can’t wait to hear Halla’s CEO Spencer Price speak at Customize on February 27th. To learn more about how Price is leveraging data to make grocery shopping a super personalized experience, check out his Q&A below. Then use code SPOON15 to get 15% off your tickets to Customize here!

Tell us a little bit about what Halla does.
Halla is a taste intelligence company that enables retailers to predict the preferences of their shoppers in real-time. Our enterprise APIs power highly-personalized shopping experiences across all retail environments, driving double-digit increases in basket size and customer retention for retailers across the globe.

There are lots of companies out there that facilitate online grocery shopping — how is Halla unique? 
We know that great recommendations lead to more purchases and better experiences, but existing options for retailers are costly, complicated, and ineffective.

Halla is different. We’re obsessed with understanding food. Because when we know food, we know people. And when we know people, we get to help retailers deliver delightful shopping experiences through deeper customer insight. 

Unlike anything else on the market, Halla is delivering food recommendations with more precision due to our proprietary data that’s able to break food items down to the molecule. With a deep understanding of each item we’re then able to make smarter recommendations based on the human experience, such as delicious food pairings or individual dietary restrictions.  

What are the benefits of a personalized grocery shopping experience, both for the retailer and the consumer?
Think about the Spotify ecosystem, versus Apple Music or YouTube. Think about the Netflix interface, versus any other streaming platform. When personalization is deeply integrated into the user experience, it leads to significant increases in conversion rates, customer retention, and all-around engagement.

So, shoppers will see only the items that are relevant to them, and will be inspired to discover new products — all while helping complete their cart as efficiently as possible. The retailer sees bigger baskets, happier customers, and has finally created a shopping environment that drives true shopper loyalty.

What do you think personalized food or drink will look like 5 years down the road?
While there are many efforts and initiatives in “standardization” — as opposed to personalization — like we see in protein bar brands, all-in-one shakes, and the likes of Soylent, there are still over 30,000 unique products in the average supermarket, and counting. The need for “product navigation”, “product discovery”, and personalization solutions will only continue to grow. How precisely this will manifest remains to be seen.

One choice that is easy to make is whether or not you should join us at Customize (hint: you should). Use code SPOON15 to get 15% off your tickets here.

FutureProof Retail to Add AI-Based Grocery Recommendations From Halla

FutureProof Retail, which provides mobile checkout technology for grocers, announced today that it will incorporate Halla’s AI-based product recommendations into its service.

FutureProof Retail uses mobile phones to create a line-free checkout process for supermarkets. Unlike other cashierless checkout solutions that install cameras and use computer vision to identify what you purchase, FutureProof moves everything to a whitelabeled app from the retailer on your phone. You scan barcodes to add items to your cart, and when you’re done, you hit a checkout button and a store employee does a quick check of your bag to make sure you aren’t shoplifting. You can see it in action in this video:

Express Checkout El Rancho Awareness Video from FutureProof Retail on Vimeo.

With today’s announcement, when people scan an item Halla’s recommendation engine will kick in to suggest a complementary product right there on the phone’s screen. So if you scanned a bag of Tostito’s chips, the app would recommend salsa. But Halla’s recommendations are also dynamic and adapt as you shop. As we wrote when the company raised $1.4 million in May of this year:

…if you are using a grocer’s app with Halla I/O built in, the app will serve up intelligent recommendations as you continue to shop online. Buy salt, it could recommend pepper. By salt and noodles and beef, and it might guess that you are making a bolognese and recommend tomato sauce.

Halla developed its recommendations based on data from anonymized grocery transactions to see what items are typically purchased at the same time, as well as restaurant and menu items (menu descriptions are, after all, typically a list of ingredients). Halla had initially started off making recommendations for restaurants as well, but pivoted away from that to focus on grocery.

It’s important to note that Halla will only provide one recommendation per item. “As of today our recommendations are focused on complementary products with one goal in mind, what is most likely to be purchased,” Halla Co-Founder and CEO, Spencer Price, told me by phone this week. Price said that this focus on a single product is important because shopping is a very emotional experience for people, and one that can’t be junked up with lots of pop-ups on a phone screen with lists to scroll through.

FutureProof works with Fairway Markets in New York and other regional grocery chains, though Price was unable to provide a timeline or location for where its recommendations will go live through this partnership.

Almost as important as the news itself is the fact that FutureProof has publicly named Halla as a partner. There are other AI-based food recommendation engines out there like Spoonshot and Analytical Flavor Systems, but they are pretty quiet about their clientele.

I’m not fully convinced about the broad adoption of FutureProof’s cashierless implementation. Manually scanning products and having a human check your bag before you leave a store seems to bring friction into a process that is supposed to be frictionless. Regardless, for those who do use it, FutureProof’s app requirement seems like a good vehicle for Halla’s technology because of its immediacy and visual cues. I imagine the company will look towards announcing partnerships with smart shelf displays like AWM Smart Shelf in the near future.

Halla Raises $1.4M Seed Round, Pivots to Focus on AI-Powered Grocery Recommendations

Halla, a startup that uses machine learning and AI to power food recommendations for grocery shoppers, announced today that it has raised a $1.4 million seed round led by E&A Venture Capital with participation from SOSV. This brings the total amount of money Halla has raised to $1.9 million.

Halla has a B2B platform dubbed Halla I/O that helps recommend relevant food products to shoppers. As we wrote at the time of Halla’s launch last year, the “company created an entirely new model and a new taxonomy that doesn’t just look at what a food item is, but also the molecules that make it up, a map of attributes linked to other food as well as how people interact with that food.”

So if you are using a grocer’s app with Halla I/O built in, the app will serve up intelligent recommendations as you continue to shop online. Buy salt, it could recommend pepper. By salt and noodles and beef, and it might guess that you are making a bolognese and recommend tomato sauce.

If you read our coverage of Halla last year, you’d notice something different about the company now. Initially, its go-to market strategy included both grocery stores and restaurants. But in the ensuing year, Halla has abandoned its pitch to restaurants, choosing instead to focus exclusively on grocery retail.

“What we’ve found is that the market timing was screaming ‘where tech meets grocery,'” Halla Co-Founder and CEO Spencer Price told me by phone recently, “The restaurant space is more crowded for building recommendations.”

But all that work in the restaurant space didn’t go to waste. “The truth is we were able to keep all of our learnings from restaurant and made our grocery recommendations stronger,” Price said. “One core learning is that restaurant dishes and menu items, as long as they have descriptions, are just recipes without instructions.”

Halla now has more than 100,000 grocery items and one hundred million unique grocery transactions from retailers across the country in its data set, informing its machine learning algorithms. Price is quick to point out that Halla does not have any personally identifiable information on people. “We can make recommendations to customer X without knowing who customer X is,” Price said

Though a grocery chain can move a lot of product and provide a lot of data for better purchasing recommendations, grocery chains as a whole do not move quickly. To get them to adopt a new technology is like turning a battleship — they need a lot of time to execute. “They’re not looking for speed,” Price said, “but a reliable solution.”

To this end, the biggest thing Halla’s funding buys them is time. “We’ve bought some runway,” said Price. The company now has some breathing room to take its time and conduct even more tests with slow-moving retailers. Halla is in tests with unnamed grocers right now, and offers its recommendation on an API pay-per-call model.

AI-based B2B food recommendation is almost its own mini-industry. Spoonshot, Analytical Flavor Systems, and Tastewise all use vast data sets to make product predictions and recommendations to restaurants and CPG companies. Other companies like AWM Smart Shelf are using a combination of prediction and smart digital signage to make in-store grocery purchase recommendations.

With online grocery shopping reaching a tipping point, people buying food via apps adding one or two more items to their cart because of intelligent recommendations could mean a nice sales boost for the grocery industry.

Halla Launches AI-Powered B2B Food Recommendation Platform

Let’s say you’ve bought a chicken sandwich. Based just on that purchase, what do you think is a better recommendation for another meal you might enjoy: chicken piccata, or a hamburger?

This is a typical type of problem that LA-based startup, Halla, is trying to solve with its new artificial intelligence (AI) powered recommendation platform. The new Halla I/O (which stands for “Intelligent Ordering”) software integrates with a restaurant or grocer’s existing website or app, so it’s invisible to the end user. People buying groceries or meals through those apps would continue to do so, but would receive recommendations through Halla’s algorithm.

Halla I/O does this, according to Co-Founder and CEO Spencer Price, by combining data and psychographics around how people think and interact with food. Let’s go back to the chicken sandwich example. According to Price, a traditional data science approach would recommend chicken piccata, because it’s, well, chicken.

But, “The human experience of eating food is different,” Price said. He went on to describe how the chicken sandwich is also meat-centered, covered on both sides with bread and eaten by hand. And when you think of it that way, a hamburger recommendation makes sense.

Price didn’t say that the chicken piccata was wrong, per se, but just provided it as an example to illustrate how Halla I/O’s AI works differently. Using a proprietary dataset from more than 10,000 grocery items, 20,000 ingredients, 175,000 recipes and 20 million restaurant dishes, Price said that his company has created an entirely new model and a new taxonomy that doesn’t just look at what a food item is, but also the molecules that make it up, a map of attributes linked to other food as well as how people interact with that food.

With that information, Price said Halla I/O can provide more contextual and more meaningful recommendations. On a surface level, this means Halla I/O can make very basic recommendations. If you are buying salt, it might recommend pepper. But if you put salt and ground beef and onions into your cart, Halla I/O might infer that you’re making a bolognese. Not only will it recommend more items needed to make a bolognese, Halla I/O will show you recommended recipes and automatically add all the ingredients for that recipe to your cart. If you’ve shopped with that app before, Halla I/O will know if you purchased tomato sauce recently, and automatically remove that from the recommendation.

Though Halla I/O works for both restaurants and grocers, right now Halla I/O is only in pilot programs with four grocers. Interestingly, one of the pilots was with Green Zebra in Portland OR. Instead of online cart recommendations, Halla provides shelf recommendations. For example, they may see that lots of people are buying bananas and peanut butter, and set up a physical display to take advantage of that.

Halla is at the nexus of two trends we’re following at The Spoon: personalization and shoppable recipes. If its software works as promised, not only will it deliver better recommendations, but the ability to automatically recommend and adjust cart items based on recipes could be another step towards customized meal kits.

Halla is just seven employees, and has raised $500,000 in angel funding. It’s going to need more than a scrappy startup spirit, however, as the AI-powered recommendation space has plenty of competition. Startups like Dishq, Foodpairing and Plantjammer are all running their own algorithms to deliver more relevant recommendations.

Pretty soon, grocers and restaurants will need an AI-based recommendation engine to recommend the best AI-based recommendation engine.


An earlier version of this article stated that Green Zebra does not have online ordering capabilities. The grocer does through Instacart, and we have updated the post.