Clover Food Lab: The Fast-Food Chain that aims to solve climate change via Open Innovation

Fast-food chain and innovation are two topics that are not often seen together. Ayr Muir, the founder and CEO of Clover Food Lab, hopes to change this. Clover involves feedback at every step of its product and process development with the hopes of making delicious food that is environmentally friendly.

Two minutes to Food Dev. Two minutes, people! Grab your name tag, your tasting sheet, a hair cover, and meet me at the kitchen lab. –Christopher (Chris) Anderson, Clover’s VP of Food Operations, shouted from across the room as he kicked-off Clover’s weekly Food Development meeting.

The fifteen participants –which included Clover’s CEO, the VP of Ops, a few employees, a handful of customers, and me— promptly followed Chris’ indications. We gathered around a steel table with samples of three different types of corn muffin ready to be evaluated.

As I tasted the new recipes and shared my thoughts with the group, I reflected on Clover’s approach to product innovation. Constant recipe tweaks fueled by feedback was not a usual practice for fast food restaurants. Most chains operated on consistency, but Clover changed up to 80% of its menu over the course of a month.[1] How could this radically different approach to its operations lead to a successful business?

Me at this Tuesday’s Food Dev Meeting and Chris introducing the three types of Corn Bread.


Clover’s Origins

Since its origin, Clover has differentiated itself from traditional players in the fast-food category by including feedback from customers, employees, and suppliers in every step of the food development process. This culture of experimentation can be attributed to Ayr Muir, the founder and CEO of Clover Food Labs.

With a BS in Material Science from MIT and an MBA from Harvard Business School, Ayr hoped to go into the field of wind energy as he cared deeply about the effects of climate change.  However, after reading a report on meat consumption and its effects on the environment, Ayr decided to go into the food service industry instead. If he was able to use a restaurant as a vehicle to reduce meat consumption, he could have a bigger impact on climate change than transportation and energy[2]. Clover’s ambitious goal is to “take someone who loves meat – and probably a lot of them don’t think they like vegetables– and we’re trying to get that person to make their own decision to change their meals [and] change their habits.[3]

Changing the eating habits of millions of people is no small endeavor. However, Ayr believes that Clover is uniquely positioned to achieve this because they incorporate experimentation and data collection into every facet of the company.  Ayr argues that having a culture of experimentation and using open innovation methods comprises much of Clover’s competitive advantage. This will lead them to “defy all the laws of gravity for [his] industry”.



The Product Development Cycle

  1. Critique at Food Dev Meeting | Food Dev meetings happen every Tuesday at 3PM at the Clover Hub in Inman Square. As described in their open invitation on their website, these meetings are “open to the public. You can listen and taste. Or you can bring a sample of something you think we should try.[4]” The group often tastes about five recipes –a  couple introduced by Clover’s VP of Food Dev, a few brought by Clover employees, and one item from a customer who has submitted a Champion Form[5].  The goal is to try new dishes that could be added to the menu or make tweaks to existing items on the Clover menu[6]. No item has been nor will be added to the Clover menu without first going through critique at Food Dev.


  1. Testing (soft launch) at one Clover location | If a recipe gets unanimous good feedback at Food Dev meeting, Ayr approves for the item to be tested at one location for a week. In this week, Clover employees hand samples of the new item and get informal feedback from costumers. If the menu item seems to be well received, it is labeled as Experimental Item and rolled out to all locations.


  1. Experimental Launch in all locations | At any point in time, Clover will have 5-7 experimental items on its menu. These items have a special label next on the menu (see picture below). The Clover team gathers qualitative and quantitative data to determine if they should make the menu item permanent. The methods used are as informal as talking to a customer who is trying the item for the first time and noting down the answers on a shared form, or as formal as keeping track of the number of sales over all locations.
Menu highlighting the “Item in Testing”: Brussel Sprout Sandwich with its backstory and special signal.
  1. Continuous testing and iteration | Testing never ends. Ayr would argue that “Clover is a giant R&D project.[8]” The Clover team is constantly looking for ways to improve their menu via monthly surveys and “eavesdropping” on Twitter[9]. Twitter feedback actually lead to a change on the type of vinegar used on Clover’s most popular item: the Chickpea Fritter Sandwich. This change was the 34th change to this item.




In his book, The Third Wave, futurist Alvin Toffler argues that pure consumers are a phenomenon of the Industrial Age and that they will be replaced by prosumers, consumers who take part in the production process and coproduce many of their own goods and services[10]. Clover Food Lab appears to be the epitome of the type of company Toffler envisioned. However, after learning about Open Innovation, reading all about Clover’s history, and attending a weekly Food Dev meeting, a few details still make me question how true the previous statement truly is.


  • Food Dev meetings –a tool for innovation or a tool for deep costumer loyalty? Even though the meeting is open to the public and employees, when I attended the meeting, I really got the sense that it might be more of a “show”. All the ultimate decisions were made by Ayr and Chris. They didn’t really ask much of what we thought.
  • Clovers method for Open Innovation—does it scale? While up until now Clover has mostly been based in Massachusetts, the company has clear plans for national expansion. When this happens, it will be impossible for Ayr to attend every single Food Dev meeting or keep close track of operations at each store. Will the current process of development fall apart then?


[Word Count: 794]




[2] ttps://


HBR Case: Clover Food Lab: Building Out the Team by Ghosn, Payton and Huberlie

[4] Clover Website:







Can the maker of Post-it notes solve healthcare pricing?


Toutiao: Challenging the “One-Feed-Fits-All” Notion of Information Dissemination

Student comments on Clover Food Lab: The Fast-Food Chain that aims to solve climate change via Open Innovation

  1. Very well written article on such an interesting topic! I actually went to Clover last night and had no idea…

    I think the answer to your questions are related: When the Food Dev Meeting is purely used as a method of developing the menu, only one such meeting is required to update the menu of all (or at least many similar) Clover locations. These meetings could either be held centrally in Massachusetts, or Ayr and Chris could hold these in different locations as they are traveling there for other reasons. However, when the meetings are more of a promotional nature, simultaneous meetings would have to be hold in all regions. These will not always be able to be physically attended by Chris and Ayr, thereby losing some of its appeal.

    1. A well-crafted and enlightening post – you present a case that we can certainly all learn something from! You pose two very interesting questions. To the second question – does this model of open innovation scale – I would argue yes. Through maximising the use of technology and gathering extraordinary amounts of data, Clover has ultimate transparency over their business and is thereby able to make their model consistent across the globe. My challenge with this business model, and similar companies such as Joe and the Juice who claim to be pioneering a more creative approach within the food industry, is the tendency to confuse creativity with technology. I remember being struck by a Boston Globe article which explained the overblown efforts to be innovative: ‘Sensors and computers monitor the soil-moisture levels of restaurant plants and water them automatically, sparing workers a menial task and giving them time better spent interacting with customers or honing kitchen skills.’ In seeking to be innovative, have they gone too far?

      Really enjoyed reading this, thank you for sharing!

  2. Congrats for the article. Very pleasant to read.
    It seems that the folks at Clover have a good process in place. As you suggested in your conclusion, rolling the concept nationally or overseas will be a big challenge but I think they can do it. The key is to have local regional independent entities that will roll up the process in their region.

  3. Your question around the potential for scale in Clover’s open innovation model resonates with me. While I love the open innovation model to guide product development at a restaurant, it seems hyper localized and best suited to operations that are small in scale. I am curious to understand if Clover has experienced any risks to-date from open innovation, and if they have thought through control mechanisms that would be required for this model to scale.

  4. As a vegetarian, Clover’s mission certainly resonates with me deeply. Reducing demand for meat is an important step in reducing our carbon footprint, conserving water, and correcting the ethical missteps of factory farming and animal cruelty. I am pretty skeptical about the use of open innovation in this model however. Food and taste preferences are so subjective – everyone could feel something different and there is no “right” answer. While it might be a useful way to get ideas for innovative foods, I’m skeptical about how it would scale. All food companies solicit feedback from customers, this isn’t really anything new (only the format in which they solicit feedback, the community tastings, feels new). The problem they all run into is that once you reach a certain scale, getting 50,000 recommendations for what flavor chips you should make becomes almost useless. People don’t agree on things and don’t always even know what they want! I’ll wait and see what happens with this restaurant chain!

    PS: It is unfortunate that a man named “Ayr” rebelled against his namesake and did not end up in wind energy…

  5. Going to the Food Dev meetings sounds like a great experience! While you may be right, and these may mostly function as a “show” to customers, it sounds like the ideas that employees bring could be very valuable. Since they are working with products every day, and potentially have the ability to bounce ideas off of customers and other employees, they might have a better grasp on what people are interested in when compared to the executives of the company. Also, since you mentioned that customers have the ability to submit a product with the Champion Form it sounds like the open innovation process is doing its job and allowing anybody to submit their idea. Even if this is more of a formality, it would only take one customer getting an idea correct in order for it to have been worth it.

    I think the way I see something like this scaling is more as a funnel towards a meeting that would then involve the key decision makers. However, the issue with this would be that it would take longer to complete meetings and get products into stores, which seems to be important to that process. I am curious to see if they are able to accomplish scaling this aspect of the business or if they feel that continuing it at only a few location is adequate to get the results they need.

  6. Very interesting read! My main concern about the Clover Food Lab model is that cuisine may not readily lend itself to the open innovation model. I say this because different people have different pallets, hence to iterate the menu endlessly to appeal to the average tastes of various crowd-sourced tasters may have the unwelcome effect of diluting the brand and what it stands for. Attempting to appeal to everyone’s pallets runs the risk of creating menus that customers do not necessarily hate, but at the same time do not feel very strong about, since the law of averages means that the product will not oscillate too widely in one direction or the other from the mean. This might limit the brand to tourist-types who just visit to experience the “cool” innovation, but not necessarily because they absolutely love the food from experience.

  7. Great post! In general, I really like Clover’s approach to continuous improvement and using a data driven method to test out new food concepts, which is dramatically different than the way most food establishments handle their menus (little data, mostly gut feeling). That being said, I am not sure how helpful the open innovation part of their model is other than providing a super interesting way for customers to connect with the brand. It is nice to crowd-source ideas for sandwich flavors or recipes, but does that actually materially impact the end product? I feel like this is a little bit like when Doritos challenges customers to submit a new flavor for chips and builds an ad campaign around it– ultimately more of a promotional tool rather than something that provides a competitive advantage in product development.

  8. Amazing mission and process — not to mention customer focus. While Clover’s mission based approach is clearly behind their innovation process, I think it also helps that trends in the restaurant industry (such as better choices, more customization and increased engagement from guests) are also fully aligned by having a robust and lightning fast innovation process in place. I too have questions regarding the ability for the company to scale this process, however. For the company to continue to engage its customers and markets in the same way, I can’t imagine this innovation process being centralized…but would replicating this on a local level make any economic sense at all? I’m not sure that it would.

  9. Very interesting post and topic. As a consumer, I never think about the process that goes into menu selection, but when I take a step back and think through the process, I realize there are an overwhelming number of complexities. For example, not only do you need to decide the items on your menu, but you need to determine what type of ingredients you will use, from where you will source those ingredients, what quantities you should buy, shelf life, etc.

    I wonder if restaurants can benefit from the use of machine learning to help them understand consumer preferences and predict trends. This forecasting would enable the Company to identify the ingredients they need to purchase, providing more visibility down the supply chain. This visibility could reduce the amount of waste and create increased negotiating leverage with farmers / suppliers.

  10. Thank you for sharing your experience with Clover. As an amateur cook myself, I really enjoy the concept of contributing to the menu and exploring other’s people ideas. It is both a cheap innovation process and a way to connect in a meaningful way with their clients. However, I wonder the problems the company might have. The first thing that comes into mind is consistency. How can they keep an food identity if they are changing so much? What would a client do if their favorite is no longer available. Instead of changing 80% couldn’t they have just one or two rotating options on the menu?

  11. I admire Clover’s mission of addressing climate change by making vegetarian food tasty for all consumers. Despite eating there pretty regularly, I didn’t know they had this responsive, open development process. I see parallels to the Agile software development process here — the fast development of a prototype, limited launch, and rapid iteration toward improvement. It works for Clover. I always like seeing what new items they’ll have on the menu when I stop by, and now I see that their constant menu cycling is related to their open innovation approach. Thanks for the recommendation for the Alvin Toffler book — it looks like a good read. The “prosumer” idea seems like a handy concept to keep in mind for future reference.

  12. Amazing to discover that there are others in our section that have realised that it is the food supply chain and what we eat that is the largest contributor to climate change and actually not the more widely assumed transportation/energy sectors. Although I do think Clover is using open innovation in an interesting way, I side with you in that it will be hard to scale a model that is based on consumers participating in weekly tasting sessions before a product is launched. I wonder if making it more of a digital voting process might be more scalable. Clover could for example open for customers to send in menus suggestions and open up for a voting contest for a sub set of the submitted suggestions each quarter, awarding the winner of the contest free Clover food for a year. In my view, that might be a more scalable model for engaging customers in open innovation once the chain goes national.

    If Clover were to stick with the current appraoach to open innovation, I believe they would need to run a more decentralised version of it as they scale with potentially different menues in each region. While this might work for Clover from a marketing, customer and business perspective, I wonder if they would still be able to achieve their goal of having an impact on the climate impact of the fast food industry as it might lead to high levels of food waste and a less climate friendly supply chain at scale.

  13. Fascinating topic! Similar to other commentators, I also doubt how the company can scale with this model of experimentation and open innovation. It seems that it might work while the company is local, but I think maintaining this model combined managing supply chain costs of a growing national business will be difficult. I think the company can establish a more controlled process for open innovation that maintains the open funnel for ideas, but might be slower and take longer roll out to different stores in order to better manage costs of constantly changing the products.

  14. Thank you for sharing!

    I’m pretty familiar with Clover’s growth story, their food, and the food dev process. It is, however, the first time I realize this weekly food dev meetings may in fact be a type of open innovation. Ayr’s comment that Clover is a giant R&D project is also particularly interesting. But I completely agree with the two concerns you expressed at the end. Throughout my virtual interactions with Ayr (instagram, blog) and simply from my understanding from following this business for the past several years, I have a strong sense that Ayr is quite a controlling leader and often has very strong opinions. I am not surprised to hear that at the food dev meeting, he and Chris didn’t seem to care much about the participants’ feedback. As for whether it is scalable. I don’t think so. They can probably continue to run the food dev meetings only at the Inman location, but as they go national, the cost of testing and changing the menu frequently will go up significantly, especially given how obsessed Ayr is with QC and his various principles.

    Now this all makes me hungry!

  15. I found this to be such an unexpected application of open innovation. However, I don’t know how realistic the founder’s ambition is given this approach. The founder’s mission is to have a greater impact on climate change, yet I don’t believe he will accomplish this vision given the structure of his current supply chain. Having 80% turnover on your menu , with 5-7 experimental items at any given time, simply cannot be sustainable or scalable. Given the variability of the menu, sourcing such a range of material inputs has to be a massive challenge and a logistical nightmare. This business would likely have to have a myraid of suppliers, each of which the company would have to negotiate and monitor. This is bound to lead to quality control issues, high variability in lead times, and difficult inventory management. This is very difficult to scale, and would require a substantial investment in fixed costs like SG&A and procure operations. As he scales, the diversity and variability on his menus will have to decrease, which means that the company’s competitive advantage will also decrease. He would also have to consider franchising, which will introduce further quality and customer experience control issues. In conclusion, I think this is an interesting idea, but I don’t think it is going to have a meaningful impact on climate change.

  16. Nicely written. I am skeptical of Clover and Ayr on several levels. First, I question whether starting a vegetarian restaurant is actually an effective way to reduce meat consumption. It does not seem like a way to get people to eat less meat so much as it is just a new vegetarian restaurant. There is nothing particularly innovative about that. If reducing meat consumption is the goal, I would argue that working on direct analogues for meat (like the impossible burger), innovating on methane capture, or reducing the green house gas from fertilizers are all more direct levers. Second, it is troubling that his PR quote for Clover is that reducing meat consumption has more impact than working on energy or transportation. That is just a bad fact. Agriculture as a whole contributes 13-24% of emissions worldwide, energy and transportation are both higher than that. That means agriculture in general and meat in particular are a huge contributors to climate change and very worthy of work, but it doesn’t help at all for Ayr to start by bending the facts. I know these have less to do with TOM, but they are important to clarify.

    Looking at Clover’s open innovation in particular, I agree that it seems like they are mostly using the customer tasting as a PR approach. I wonder if they are gathering more comments from monitoring twitter and other social media than they were in the in-person tastings. Even if they are gathering that feedback, however, the bulk of their approach seems like internal product development and extensive customer testing. That seems less like open innovation and more like traditional market testing. I’d be intrigued to see what would happen if they gave more weight to the “Champion Forum” and really doubled down on customer recipe development.

  17. Super interesting restaurant concept! I did not realize that their menu designated experimental items. On a small scale, I understand how this methodology works well to incorporate customer feedback so that they feel heard. On a large scale, I’m unsure of how scalable this model is. Similar to Chipolte having fresh items and the resulting health issues how do you control the food quality/health while incorporating synergies of a change i.e. similar menu, food sourcing, etc? Would it be easier to start having one local item per location that is designated experimental so that it can be approved by corporate while also engaging the local customer base? To me then power of this is gaining customer loyalty and also interest by the nature of this type of collaboration.

  18. I think this is an incredible idea, but I agree with your first comment – it appears more as a marketing technique than truly an open innovation platform. Unfortunately, this approach clearly isn’t scaleable; if the objective is to truly change customer perspectives on food as a whole, it will be hard to do that with such a limited reach given how intensive their innovation process is. I would focus on finding more electronic ways to provide feedback and iterate more consistently.

  19. I am struck by how iterative Clover’s menu selection process is. Clover changes 80% of its menu in a month. Is that enough time to evaluate any item and conclude that they are a successful component of the menu? Moreover, are customers attracted to Clover because their food is getting more delicious as part of their open innovation process, or are they attracted to the constant variety? It would be interesting to test whether their menu today is better than their menu a year ago. In other words, is the customer feedback leading to a better product? If that is not the case, then Clover may want to explore the accuracy of their data. It may also lead one to question whether customers actually know what they want.

  20. This sounds like a fantastic initiative – thank you for sharing! The question you pose about scalability seems very real! There are a couple of challenges I foresee. Firstly, will Clover alter it’s menu for national trends (across all it’s stores when it does go national), or would there be complete autonomy at a store level? Both seem to come with it’s fair bit of challenges. Secondly, does the customer truly know what they want? Often, the value add in this industry is being able to bring new tastes to the customers and not necessarily be as input-driven. With these two points in mind, a consideration is that maybe Clover is better off staying as a local brand? Their innovative approach and aspirations for scale seem to come at a heavy trade off!

Leave a comment