AI’s Chicken Sandwich Problem

“AI in the enterprise has over 250 chicken sandwiches, and it only needs 3.” I started a remote presentation for a group at Davos with that line yesterday, and I’ll share the same story with everyone in this community. You might think you know where I’m going with this story, but the ending will catch you by surprise.

Most AI strategies are little more than a mandate to “do something with AI.” Let me tell you how that always ends up. I’ll also explain what’s fundamentally different this time around.

The Chicken Sandwich Wars

In 2020, McDonald’s began a “multi-year chicken journey” in response to what many in the fast-food industry called the chicken sandwich war. You’re laughing, but this was a real event that played out over two years. It has its own Wikipedia page. From 2019 to 2022, the chicken sandwich was hailed as the biggest growth driver in the industry.

Fast food companies believed it would reignite stagnant sales growth, and the best chicken sandwich would be the key to taking market share from competitors. It all started when Popeyes and Chick-fil-A got into a spat on social media that went viral. Popeyes released a new chicken sandwich that sold extremely well.

Chik-fil-A, seeing a threat, pointed out that they were the first chain to offer a chicken sandwich and that theirs was much better. Both companies cited benchmarks from taste testing and sales figures to support their claims. Each activated its influencer army to sway the masses to its side.

Begun, the chicken sandwich war has.

Organic social media posts from people who loved one or deeply hated the other went viral. You couldn’t go on Twitter, TikTok, Instagram, or Reddit without running into someone talking about chicken sandwiches.

“We’re certainly working on ways that we can update and upgrade our chicken offering in the U.S. Suffice to say, we will continue to be competing and innovating in chicken,” McDonald’s CEO Chris Kempczinski said in 2020.

By January 2021, over 20 US fast food chains introduced a chicken sandwich. Poultry farms, the picks and shovels of the chicken sandwich boom, ramped up to manage the surge in demand. Chicken prices hit an all-time high in the Spring of 2021. Popeyes began stockpiling chicken in preparation for a further menu expansion.

The Winners & Losers

Popeyes grew its revenue by almost 38% in 2019 alone and opened 865 new locations by 2022. The chicken sandwich-driven growth cycle only recently came to an end last year after an impressive 5-year run.

By the end of 2022, McDonald’s CEO was singing a different tune. He complained in one meeting that the company had over 250 different variations of the chicken sandwich, and none of them were driving significant revenue growth. He ordered a consolidation back down to just the 3 most popular versions. McDonald’s began an aggressive restructuring and downsizing effort shortly after. It focused on winning back customers with better value and growth through other parts of its menu.

The difference between Popeyes and McDonald’s is obvious to the outsider. McDonald’s top menu items are all burgers. That’s the brand. McDonald’s has the Hamburglar, not cute cows running slide decks urging people to eat more chicken. Popeye’s slogan is literally, “Love that chicken from Popeyes,” while McDonald’s used to just be “I’m loving it.”

Popeyes is a chicken company. McDonald’s used to be a burger company. Then it was the burger company with some chicken, too. Now, there’s also that fish sandwich and the weird, on-again/off-again, rib sandwich. Even Ronald McDonald looks confused these days.

Burger King makes burgers. Chick-fil-A, Popeyes, and KFC do chicken. Burger King believes the burger fast-food business model will prevail in the long run, and the others are all-in on chicken. That’s why Popeyes and Chick-fil-A benefited most while McDonald’s saw a modest bump.

If you’re a chicken business, a chicken sandwich will amplify your business model the most. For burger chains, it’s just an incremental improvement. A burger chain going all-in on chicken innovation and competition doesn’t make sense unless it plans to transition to the chicken model. That’s how companies lose their core brand identity, strategic focus, and connection with customers.

When Burger Companies Chase Chicken

Yes, there’s a lesson about AI opportunity discovery and enterprise use case selection here, but I’m headed in a completely different direction.

Think of NVIDIA in 2019-2021 being tempted by the crypto boom. GPUs were used for crypto mining, and NVIDIA got a significant boost in sales from crypto in the last two years of the cycle. If it had taken its eyes off of AI to chase that opportunity, would it be as dominant in the AI market today?

Jensen Huang took NVIDIA all-in on AI. He didn’t refuse the new sources of business, but the core strategy never wavered. He had conviction in AI and kept positioning NVIDIA as an AI business in every earnings call. The company refused to become a divided business even when demand spiked.

NVIDIA also wasn’t a CPU company that also sold GPUs. Intel flirted with that strategy for a couple of years before abandoning it. AMD has split its focus for some time now, and it hasn’t seen the same gains as NVIDIA.

SaaS companies are the technology world’s McDonald’s, and if they’re not careful, they’ll follow Intel’s decline. They diversified their product lines across domains to continue growing beyond their core business. They are not AI companies, even though they have always had a few AI items on the menu. Now that AI demand is spiking, SaaS companies are offering 250 versions of their AI sandwich.

Each SaaS company has multiple platforms and product lines, each with its own AI brand and assistant. Some have AI clouds, AI apps, AI foundational models, AI agent builders, AI analytics, AI simulations, and on and on. But they’re still not AI companies. Their pricing models charge for subscription apps, platforms, modules, consumption, or some hybrid version.

SaaS and most legacy technology companies still have app-based or cloud-based business models. Apps were sold as individual products. The cloud is just treated like a different product category that they can hang their app products on to bundle them. It’s not really a novel business model, just an incremental change, but it has worked well for them.

The Death Of Products & The Rise Of Platforms

I spent about 20 minutes explaining the old and new paradigms, but for the purposes of this article, I’ll get right to the point. That’s why SaaS companies haven’t been very successful with AI. There’s no such thing as an AI product in the app paradigm. AI is a platform-first technology, so its ‘products’ follow platform-first monetization paradigms.

But there aren’t really distinct products in the AI paradigm, and you can’t just hang all your apps on an AI interface, then call it good. Users get different outcomes from a single starting point or single pane of glass. There’s no reporting product like Power BI or Tableau. There’s just the AI.

Reporting isn’t even a distinct function in an AI platform. Data visualization is built into most workflows, but it isn’t the core of any of them. No one’s job is just reading reports. We use data to improve or inform whatever our actual job is. Our roles require us to do something with the data or leverage the data to improve a work product and business outcome.

Hidden in every app-first, task-centric workflow, you’ll find several similar steps. No one’s job is just reading resumes. The job is to hire qualified candidates. No one’s job is filling out expense reports or reading marketing copy. There are outcomes on the other side of every task and workflow.

Apps charge customers by the steps and workflows they support. AI charges based on the outcomes it supports (and eventually delivers with increasing autonomy). SaaS companies have a product catalog. AI platforms have an outcomes catalog.

AI’s Winners & Losers

Platforms are the new product, the only product, for AI-first companies, and only AI-first companies will be the big winners of the AI platform wars. AI will provide incremental opportunities and growth for SaaS companies, as chicken sandwiches did for McDonald’s and Burger King.

However, going all-in on AI innovation without transforming from a SaaS product into an AI platform business doesn’t make sense. SaaS companies have the advantage of data, but they haven’t been able to capitalize because they’re still anchored to the app-first product paradigm.

You can call this ‘the innovator’s dilemma’, but I see something different. It’s a faster-moving disruption cycle and a clean break from legacy monetization paradigms. Why do we need all these apps? In the near future, I’m just going to open Claude. Not Claude Code for engineering, Claude Cowork for nontechnical workflows, Claude Classic for search and research, etc.

We will just start Windows. Not Word for writing, VSCode + Copilot for engineering, Outlook for email, Teams for messaging, Power BI for reports, and even more etc.

We will just run Gemini, ChatGPT, Salesforce, SAP, whatever, not an endless app sprawl. One AI platform. One AI interface. Multiple intents supported and outcomes delivered. Zero friction from context switching and app hopping.

SaaS and AI seem similar, like chicken sandwiches and hamburgers, but they aren’t. The single pane of glass is a significantly better user experience. Think of how AI search is ending the click-through to websites, and AI Q&A is flatlining traffic to sites like Stack Overflow. It’s the same behavioral shift with apps instead of websites and a much higher reduction in effort.

The incremental bumps in revenue from AI products have lulled SaaS companies into believing they’ve weathered the worst of the storm. In reality, they have just felt the first waves.

If you’d like to have me talk about chicken sandwiches and how to grow revenue with AI, send an email to seminars@v2ds.com.

If you’re ready to learn more about AI & Agentic Platform Monetization, have a look at my new course.