Lauri Kien Kotcher, CEO and Co-Founder of Different Day, is an experienced business leader, brand builder, board director, and strategic advisor whose career spans artificial intelligence, consumer products, retail, healthcare, financial services, and private equity. Before founding Different Day, she served as CEO and board member of quip, where she led a broader shift from oral care to oral wellness, introduced new products, reworked the company’s direct-to-consumer subscription model, and reduced operating costs. She previously led The Shade Store and Hello Products, guiding the latter through significant growth and its acquisition by Colgate in 2020. Her earlier leadership positions include senior roles at L Catterton, Godiva Chocolatier, Lehman Brothers, and Pfizer Consumer Healthcare, as well as 15 years at McKinsey & Company, where she became a partner and co-led the firm’s North American consumer goods practice. Kotcher also serves on Freshpet’s board of directors, bringing extensive experience in growth strategy, innovation, marketing, mergers and acquisitions, and organizational transformation.
Different Day helps middle-market organizations turn operational challenges into practical, customized artificial intelligence applications that integrate with their existing technology environments. Rather than focusing on experimental AI projects or general-purpose software, the company develops end-to-end solutions designed to reduce manual work, improve visibility, accelerate revenue growth, strengthen margins, and uncover useful business insights. Its work spans areas such as demand forecasting, supply and inventory planning, sales intelligence, financial aid reconciliation, compliance, contract analysis, customer relationship management, and performance reporting. Different Day positions its team as a partner from initial problem discovery through deployment and knowledge transfer, enabling clients to operate and continue improving the applications internally.
You spent decades leading transformation across McKinsey, Godiva, hello products, The Shade Store and quip before co-founding Different Day. What convinced you that middle-market companies needed a new kind of AI partner, and which operational problems did you believe existing technology vendors and consulting firms were failing to solve?
Shockingly, business leaders often live in darkness. As a CEO and CMO running high-growth middle-market consumer brands and retailers, I could never get a clear read on the full scope of all the moving parts of the business. Our data often lived in disconnected silos by channel, geography and functional owner. My instinct was that AI could finally connect those dots and help figure out what we were missing, but I had no idea who to call or where to start. Different Day became the answer to that question.
Middle-market companies have fewer resources than large enterprises, but they may also have less bureaucracy and technical debt. How should their approach to AI implementation differ from the enterprise playbook?
Middle-market companies have an extraordinary superpower they need to embrace. They should aim to leapfrog their bigger competitors since speed is their real advantage. Our view is that we can turn most middle-market businesses into AI-native organizations, as if they were being launched in 2026 from the ground up. Large enterprises are quietly afraid that AI will let smaller rivals level the playing field, using AI teammates to move faster without adding headcount. Fewer people mean fewer people deciding what to do, how to do it, and what the goals are, so a smaller organizational chart becomes an asset, not a limitation.
When evaluating a company’s operations, how do you distinguish a workflow that genuinely benefits from AI from one that would be better addressed through conventional automation, improved software or process redesign?
I believe in the naked approach: strip your workflow down to the basics. Simplifying the process should come before the AI. Start with a clean sheet of paper – what would you do here if the legacy software and workarounds didn’t exist? Most existing processes were built to route around clunky tools, not because they’re the right way to work. As contracts come up for renewal, ask what features you use, and with few exceptions, AI can build something more tailored than what you have today. Though there are always systems worth keeping in place, like your ERP or CRM, we design the exact add-ons you need and nothing more.
Different Day says it can begin with a 72-hour prototype and move to a production-ready handoff within weeks. What makes that pace possible, and what technical, security or organizational conditions could slow an implementation down?
It seems like an over-promise, but it’s a reflection of our unique advantage as a company founded by people who understand business AND technology. Our team is AI-native, so the build itself moves far faster than traditional software development. What doesn’t change is discovery: who owns the process, who else needs to buy in, what outcome you’re targeting and by how much, what’s failed before. Skip that work up front, and it resurfaces later; in integrations, in implementation, at a much higher cost. The pace is only possible because we front-load the understanding instead of coding.
Many middle-market businesses operate across disconnected spreadsheets, legacy systems and inconsistent data sources. How much data preparation is usually required before AI can produce dependable operational results?
The disconnected data most middle-market companies live with is actually the opportunity, not the obstacle. Our tools surface exactly where the gaps are, and the company decides whether and how to fill them. Before AI, those gaps were invisible; someone just muscled the spreadsheets into a decision or decided on incomplete data without realizing it. Now teams can see the gap and weigh their decision accordingly instead of assuming the data was solid.
What metrics should executives establish before beginning an AI project, and how quickly should they expect to see evidence that an implementation is generating a meaningful return on investment?
AI may be new, but the classic business disciplines prevail. Where AI runs into trouble is when people think all the old rules are gone. Every AI project needs one clear goal going in, whether that’s sales, brand visibility, margin, decision speed, or more output without more headcount, along with a realistic timeline for hitting it. Timelines vary with the number of data connections, internal and external systems involved, and how much the tool needs to learn before its output is solid. You should see some form of result by the three-month mark; more complex projects take longer. The key is to design the project in stages, so you’re not waiting for one big-bang outcome.
Different Day highlights applications spanning demand forecasting, supply planning, contract intelligence, financial-aid reconciliation and real-time sales signals. Which use cases are producing the most repeatable results today, and which remain highly dependent on a company’s industry and operating model?
Don’t underestimate AI’s repeatability factor. In many ways, businesses are structurally more alike than we realize, from universal contracts to matching customer needs with internal capabilities. Contract intelligence and financial-aid reconciliation are the most repeatable today; both have clear parameters and clear outputs, so the tool knows exactly what it’s solving for. Demand forecasting and supply planning are harder: more data sets, more channels, and far more variation from one company’s operating model to the next. Something like real-time customer notification falls in the middle, depending on the company’s current tech stack and channels.
How should a business decide whether to buy an off-the-shelf AI product, build internally or develop a customized application with an external partner? What questions help expose unnecessary complexity or vendor lock-in?
Avoid AI clickbait and be skeptical of LinkedIn posts that make building sophisticated tools internally sound easy. Build internally only when the tool is genuinely simple, such as a dashboard that pulls a few spreadsheets into one clean view for a handful of colleagues. The moment you need real-time data pulling from internal and external systems, that’s out of reach for a non-technical team with DIY tools. Off-the-shelf AI has the same ceiling as off-the-shelf SaaS always did. It’ll give you a read on generic data, but it won’t give you the nuance specific to your business and your competitors.
What are the most common mistakes you see companies make when moving from an impressive AI demonstration to a system that employees must trust and use every day?
I’m seeing an epidemic of “prototype-itis”: companies falling for the visible 10 percent of the iceberg. The prototype looks impressive, but the real work happens below the surface: data testing and back-testing, testing every system linkage, and securing the data. Then comes change management: teaching people how to use it, where its limits are, when a human needs to stay in the loop, and what questions to ask of its output. AI should be fit for a purpose, not a replacement for human judgment, especially in a middle-market company where that judgment took years to build.
As AI becomes embedded in core operations, how do you expect the roles of managers and frontline employees to change, and what should middle-market leaders do now to ensure AI elevates their teams rather than simply adding another layer of technology?
“Managers” once applied to directing people. Rapidly, it has become the management of agents and systems. Start by mapping the team’s biggest pain points and isolating the manual, repetitive work; that’s where AI belongs first, before you hand it more complex decisions. Walk before you run, especially in a middle-market company where trust must be earned one win at a time.
Thank you for the great interview, readers who wish to learn more should visit Different Day.
