Small Teams, Big Innovations: Lean Digital Product Development in High-Tech Firms

Engineering and manufacturing companies face intense pressure to deliver new digital products — from AI-driven tools to IoT platforms — yet many struggle under teams and layers of managers with traditional product development approaches. This paper demonstrates that successful digital innovation does not require massive teams, layers of consultants and managers, or protracted market studies. Instead, evidence shows that small, passionate, and empowered teams (on the order of 10–20 individuals) can drive outsized innovation and business results when given the proper support. We provide case studies of nimble startups achieving rapid growth with minimal headcount, examples of large companies adopting “lean” innovation models, and hard data illustrating how lean teams outperform. Key findings include:

• Rapid Growth with Lean Teams: Startups like Cursor (Anysphere) and ElevenLabs reached ~$100 million in annual recurring revenue (ARR) within ~2 years using teams of well under 50 people thevccorner.comsacra.com. Others like Doss and Lace AI cracked into mature markets (enterprise software, call centers) with ~20-person teams by focusing on speed and customer-centric development (upstartsmedia.comaimresearch.co). These cases underscore how small teams can out-innovate incumbents given focus and talent density.
• Lean Innovation in Big Companies: Even tech giants and industrial firms have learned to “think small”. Amazon’s famous “two-pizza teams” (keeping teams small enough to feed with two pizzas) illustrate how <10-person teams minimize bureaucracy and move faster aws.amazon.com. Lockheed Martin’s Skunk Works is a classic example: a tiny team of 23 engineers built a cutting-edge jet in just 143 days by operating outside normal processes latimes.com. Leading industrial firms like Bosch have established internal incubators to empower startup-style teams within the enterprise bosch.us.
• Talent Over Pedigree: A common thread among successful innovators is an emphasis on hiring for skills, passion, and real-world contributions over traditional credentials. This does not mean degrees or GPAs are unimportant—they can reflect discipline and foundational knowledge—but rather that they are not the sole or best predictors of job performance. Successful innovators prioritize demonstrated ability and creative problem-solving over formal pedigree. Successful investors look for “evidence of exceptional ability” businessinsider.combusinessinsider.com. Google similarly found that academic metrics like GPA are “worthless” in predicting job performance and now has teams where 14% of employees have no college degree twentyonetoys.com. Lean teams recruit T-shaped contributors who show they can build things, often by highlighting prior projects or open-source contributions rather than relying on fancy resumes.
• Recommendations for High-Tech Firms: To turn these insights into action, engineering companies should identify employees with a knack for innovation and build small, autonomous teams around their ideas. Cut back on bureaucracy, internal approvals, and reliance on external consultants and workshops. Instead of spending large budgets trying to define the problem, invest in internal "intrapreneur" teams to build prototypes and iterate directly with customers. Foster a research-driven culture that embraces experimentation and accepts failure as part of the innovation process. Practical steps include hosting internal hackathons or creating incubators to surface new ideas, giving teams end-to-end ownership, and redefining success metrics to prioritize learning and adaptability. This paper offers a comparative analysis, visual data on team size vs. outcomes, and actionable recommendations to help high-tech firms adopt lean innovation and stay competitive in a fast-changing digital landscape.

Introduction

Many established engineering and industrial companies – including high-tech manufacturers – are finding themselves outpaced in the race to develop digital products. Traditional approaches often involve large project teams, multiple management layers, extensive planning cycles, and heavy use of consultants. However, these methods frequently yield disappointing results for software innovation. Studies have shown that larger projects in corporate environments have shockingly low success rates: The Standish Group found that while small projects succeed 62% of the time, only 6% of large projects are fully successful pmessentials.us. In other words, the bigger and more complex the initiative, the more likely it is to stumble. Long timelines and bureaucracy can be fatal in fast-moving domains like AI, software-as-a-service (SaaS), and Internet of Things (IoT) platforms.

High-tech industrial firms often carry a legacy product development with heavy stage-gate processes. When they attempt digital innovation, they may apply the same big-team, high-process formula – only to be outmaneuvered by startups operating with lean, agile methods. The result is a frustrating gap: despite sizeable R&D budgets and industry expertise, incumbents struggle to produce software offerings as compelling or timely as those from far smaller competitors. For example, while an established company might spend years and millions developing an IoT analytics platform, an upstart of a dozen developers can release a cloud tool in months and start capturing market share.

To close the gap between idea and execution, organizations need to rethink how teams are structured and how culture supports innovation.  The principle that “small is powerful” proves true in many high-performing product teams. Real-world case studies and data show that small, passionate teams—when given the right autonomy and support—consistently outperform expectations and drive meaningful results. Some large organizations have successfully adopted lean startup principles internally to speed up software development, proving that size doesn't have to slow innovation. This paper outlines practical recommendations for engineering and manufacturing firms on how to build and sustain lean, high-impact teams—and how to create a culture where they can succeed.

The Power of Small, Empowered Teams

Why do small teams tend to outperform larger ones in innovation? The answer lies in agility, focus, and motivation. Small teams minimize complexity. Communication overhead grows exponentially with team size – sometimes called the Ringelmann effect, where individuals in a bigger group become less productive on average. In contrast, a tight-knit group of 5–15 people can iterate quickly and align easily on a shared vision. Amazon institutionalized this idea with the “two-pizza team” rule: no team should be larger than can be fed with two pizzas (typically about 6–10 people). Amazon credits these autonomous, focused teams for its agility and customer-centric innovation. Small teams reduce bureaucracy by design – there are fewer handoffs and less need for meetings, reporting, and management layers. As Amazon’s internal documents note, keeping teams under about ten people “minimizes lines of communication and decreases the overhead of bureaucracy and decision-making,” aws.amazon.com.

Small teams are also more engaged and mission-driven. Members of a lean team feel personal ownership of the product and know their contributions matter. A Gallup study cited by Amazon found that organizations with fewer than 10 employees had engagement levels of 42%, compared to under 30% in large organizations(Forbes). High engagement fuels extra discretionary effort – people in a startup-like setting often work with an urgency and passion that larger teams struggle to match. In many cases, lean teams operate almost like SWAT units or “tiger teams”, laser-focused on a specific goal or product, which concentrates their energy. They can rapidly prototype, test with users, and iterate, whereas a big team might be bogged down creating slide decks for steering committees.

History provides striking examples of the disproportionate impact of small teams. During World War II, Lockheed Martin’s Skunk Works pioneered the idea of an autonomous mini-organization dedicated to radical innovation. In 1943, a Skunk Works team of just 23 engineers (supported by ~105 technicians) designed and built America’s first jet fighter prototype – the XP-80 – in an astonishing 143 days, beating an already aggressive schedule latimes.com. They achieved this by bypassing normal bureaucratic procedures: the team was physically separate, managerially independent, and empowered to make swift decisions. That project’s success (and many others under Skunk Works leader Kelly Johnson) cemented the principle that small, flexible teams can accomplish feats that larger groups find impossible under tight timelines. The lesson remains pertinent even in today’s software-driven projects: lean, skunk-works style teams can out-innovate far bigger organizations when freed from bureaucracy.

Modern research on software projects reinforces this. The Standish Group recommends breaking large initiatives into “mini-projects” with team sizes of 7 to 11 people, noting that this drastically improves success rates pmessentials.us. In essence, size matters – but bigger is not better for innovation. By keeping product teams small and cross-functional, companies can significantly boost their speed and success. The following sections will demonstrate this through real startup case studies and look at how even large companies have reshaped their approaches to leverage the power of small teams.

Startup Case Studies: Rapid Growth with Lean Teams

One of the clearest indications that small teams can deliver big results is the remarkable performance of recent startups. In the past few years, numerous tech startups have achieved rapid revenue growth and market impact with teams of only a dozen or two people. These cases serve as living “experiments” proving that much can be accomplished with little bureaucracy and headcount, given modern technology leverage. Below, we highlight several such startups and their outcomes:

Cursor (Anysphere)

Cursor is an AI-powered coding assistant developed by the startup Anysphere, founded in 2022. Despite being a newcomer, Cursor’s rise has been meteoric. In just about one year from launch, Cursor became one of the fastest-growing software products ever – reaching $100 million in annual recurring revenue (ARR) without any marketing spend entrepreneur.com. By early 2025, it had over 1 million users and had doubled to $200M ARR entrepreneur.com. This growth trajectory outpaced even OpenAI’s ChatGPT in monetization. Crucially, Anysphere accomplished this with a very lean team. Reports indicate the company had on the order of 20 employees when achieving the $100M ARR milestone thevccorner.com, illustrating extreme productivity per person. “Almost all of Cursor’s revenue” came from individual developers paying for the tool themselves (a bottom-up adoption model) entrepreneur.com. This underscores that a small, focused team closely attuned to user needs can build a product compelling enough to penetrate the market organically, without big salesforces or marketing departments. Anysphere’s internal culture appears to enable this speed: their job postings emphasize a flat organization with a “talent-dense” small team that values crazy ideas and shipping code quickly anysphere.inc. They explicitly ask applicants to share “a project you’re proud of” when applying, signaling that concrete skills and passion are prized over credentials anysphere.inc. By assembling a tight group of such talented engineers and giving them autonomy, Anysphere was able to leapfrog far larger companies in the coding tools space in a stunningly short time.

ElevenLabs

ElevenLabs, founded in 2022 by just two engineers, is another startup that demonstrated explosive growth with a minimal team. The company specializes in AI-driven voice synthesis – essentially a platform for ultra-realistic text-to-speech and voice cloning. In under two years, ElevenLabs went from zero to serving millions of users and a run-rate nearing $100M in ARR thevccorner.com. By October 2024, it had reached an estimated $90M ARR (up from $25M at the start of that year), and notably achieved profitability along the way sacra.com. This kind of growth (over 3X in 9 months) is exceptional. Even more telling is that ElevenLabs accomplished it with a tiny workforce. For much of its early journey the team was reportedly only around a few dozen people. One analysis notes ElevenLabs got to ~$100M ARR with roughly 50 employees in total thevccorner.com. The company’s co-founders focused on hiring top-tier AI research and engineering talent regardless of location or pedigree – they themselves are young Polish entrepreneurs whose idea attracted global attention. With a lean staff, ElevenLabs could iterate rapidly on their voice engine and deploy improvements weekly. They also leveraged a community of creators and developers who built content with their tools, effectively scaling outreach without a large sales organization. By early 2023, ElevenLabs had already won adoption by 1 million registered users in 170 countries through a self-service model sacra.com. This case exemplifies how in the digital era, a small engineering-driven team can quickly create a worldwide platform by combining cutting-edge research with relentless product iteration. The founders noted that being lean forced them to be “hyper-focused on the core technology and user feedback,” avoiding distractions that often come with bigger operations.

 Doss

Not only consumer-facing AI startups, but even enterprise software companies are proving the power of lean teams. Doss is a two-year-old startup (founded ~2021) aiming to disrupt the staid world of enterprise resource planning (ERP) software with AI. ERP is dominated by giants like Oracle NetSuite and SAP, yet Doss has made inroads by moving faster and innovating on user experience. With “a team of about 20 employees,” Doss reached $1 million in ARR within its first two years by mid-2023 upstartsmedia.com. That may sound modest compared to the likes of Cursor, but consider the context: ERP is a complex B2B market where winning even a handful of corporate customers is difficult. Doss’s small team managed to sign up over 30 business customers and deliver a full AI-driven ERP product, an achievement that typically takes larger enterprise software firms far longer with many more resources upstartsmedia.com. By radically automating aspects of the sales and onboarding process (using their own AI tools), the Doss team accelerated growth without needing a big support staff. Their success attracted $18 million in new venture funding to take on entrenched incumbents, upstartsmedia.com. In essence, Doss is a “David vs Goliath” story: a lean startup attacking a market of billion-dollar giants. Their early traction, accomplished with only ~20 people, shows that even in enterprise domains a focused team can build a credible product quickly by leveraging modern AI and cloud architectures. It also highlights how such teams are unencumbered by legacy – whereas an incumbent might be stuck supporting old customer installations, Doss could design a cloud-native, AI-first solution from scratch and iterate rapidly.

 Lace AI

Lace AI is another young company that illustrates what a small, driven team can do in a short time. Lace AI provides an AI platform to turn customer conversations (like call center transcripts and chats) into revenue insights, effectively bringing AI to sales and support analytics. They began commercial sales only in late 2023. Yet by the end of 2024 – in roughly one year – Lace AI had expanded to over 100 business customers and grown its ARR by an astonishing 1,000% (a tenfold increase) aimresearch.co. This breakout year was achieved by a team of just ~20 people in the company aimresearch.co. Such exponential growth with limited manpower is a testament to focusing on a scalable product and a specific niche. Lace’s leadership recounts that in the beginning, “we were one small team, working off a single stack-ranked list [of priorities]. Alignment and speed were everything.” launchub.medium.com. By ruthlessly prioritizing features and staying very close to customers’ needs, the small team could deliver updates continuously, fueling rapid adoption. After securing a $14M funding round in 2024 to scale up, Lace plans to roughly triple its team aimresearch.co – but notably, even after tripling, it will still be on the order of only 60 people, showing how far a tiny startup can go before any “bulk” is needed. Lace AI’s story also underscores the importance of supporting a research-driven culture: many of its team members have AI research backgrounds, and the company’s approach is to implement state-of-the-art language models into a real-world solution faster than larger competitors can. This agility comes from having a small, tightly-knit engineering core that iterates on research ideas without heavy management overhead.

Other Notable Examples

The above are just a few examples among many recent startups proving the efficacy of lean teams. To illustrate the broader trend, Figure 1 visualizes several startups and the outcomes they achieved with small team sizes. Companies like Midjourney – an AI image generation service – reached an estimated $200M in ARR with only ~10 employees on staff thevccorner.com. Another AI company, Mercor, got to ~$50M ARR in two years with ~30 people thevccorner.com. In fact, a review of emerging AI-driven startups in 2023–2024 found multiple instances of 8–15 person teams achieving tens or even hundreds of millions in revenue within 1–2 years thevccorner.com. These outcomes would have seemed unbelievable a decade ago, but they reflect the unprecedented leverage of modern technology (cloud infrastructure, open-source, AI) combined with lean methodologies. A small team today can rent virtually unlimited computing power, tap into global distribution via app stores or the web, and build on open-source frameworks – allowing speed and scale that previously required armies of developers. The takeaway for incumbents is clear: a lack of large staff or big budgets is not necessarily a disadvantage. On the contrary, constrained, agile teams are often more inventive and efficient. By studying how these startups operate – fast release cycles, obsessive focus on user feedback, and hiring only “A-players” – larger companies can learn how to emulate their success in-house.

Figure 1: Rapid growth achieved by startups with lean teams. Several AI-driven startups (Midjourney, Cursor/Anysphere, ElevenLabs, etc.) reached high annual recurring revenues (ARR) in a short time with very small teams (often under 20–30 people) thevccorner.com. This illustrates the outsized impact a lean, talented team can have in the digital product space (Image courtesy of Ben Lang).

Lean Innovation in Large Organizations

Small, fast-moving teams are not just the domain of startups. Forward-thinking large companies have deliberately adopted lean team models to spur innovation internally. These examples show that even in a corporate setting with thousands of employees, it is possible – and often advantageous – to create startup-like units that operate with autonomy and focus.

One of the most cited examples is Amazon’s “two-pizza team” structure, championed by CEO Jeff Bezos. As mentioned earlier, Amazon organizes its workforce into many independent teams typically fewer than 10 people, each owning a specific product or service. The rationale is to combine the strengths of a big company (resources, infrastructure) with the nimbleness of a startup. Bezos observed that as companies grow, there is a risk of becoming slow and bureaucratic, so Amazon constantly fights that by keeping teams small and decentralized. These teams have end-to-end responsibility – for example, one two-pizza team fully owns the development, deployment, and operation of a microservice on Amazon’s platform aws.amazon.com. They don’t “handoff” a project to another group; they build, launch, and run it like a small business. This model has been credited with enabling Amazon Web Services (AWS) to innovate rapidly and roll out new offerings faster than competitors. Each AWS feature or service often starts with a tiny team acting as an internal startup. Amazon’s approach has also shown scalability: when a team’s scope grows too large, they split into two autonomous teams (much like a cell dividing) rather than accrete more people aws.amazon.comaws.amazon.com. This way, they maintain agility even as overall company headcount grows. The success of Amazon – now one of the most valuable companies in the world – demonstrates that lean teams can work on a massive scale if the culture and structure support them.

Traditional industrial companies have also seen the light. We discussed Lockheed Martin’s Skunk Works as a historical example; its legacy lives on as many corporations set up similar innovation labs. These are intentionally kept separate from the main hierarchy to allow rule-bending experimentation. For instance, Boeing, Northrop Grumman, and other aerospace firms have all implemented “rapid innovation cells” modeled after Skunk Works to develop advanced projects on faster timelines. In the consumer electronics realm, Apple’s original Macintosh team in the 1980s was famously sequestered in a separate building with a pirate flag, embodying a rebel startup culture within Apple that produced the groundbreaking Mac computer with a relatively small team.

More recently, industrial giant Bosch created an internal startup incubator called “Grow” in 2012 to foster exactly this kind of lean innovation. Bosch’s Grow division gives small teams of intrapreneurs a home to develop new business ideas with minimal bureaucracy. These teams function like startups, even pitching for funding from Bosch’s board as venture capitalists. Bosch reports that the program has led to multiple new products and even standalone spin-off companies over the years bosch.us. Notably, Grow’s structure protects these teams from the normal corporate processes, acknowledging that innovation requires freedom to experiment beyond the usual rules. Other large companies have launched similar programs – Google’s famous “20% time” policy let employees devote a day a week to passion projects, which birthed Gmail and Google News. At the same time, Adobe’s Kickbox initiative gave any employee a funded kit to prototype an idea. The common theme is that empowering small, independent teams within a big organization can dramatically increase innovation output. It allows big firms to harness the creativity of their people without the stifling effects of hierarchy.

It is also instructive how these lean teams are managed and evaluated differently. Instead of traditional KPIs suited for steady-state operations, they use startup-style metrics – e.g. prototypes built, customer feedback cycles, usage growth – to measure success. This protects them from early shutdown due to conventional ROI calculations. For example, when General Electric (GE) adopted Eric Ries’s Lean Startup methods in its FastWorks program, some of its appliance and healthcare units were able to cut product development times by over 50% by empowering small cross-functional teams to test assumptions quickly rather than following 12-18 month plans durichitayat.netxmpro.com. While not all attempts have been perfect, the trend is clear: large companies that embrace lean, agile teams (backed by executive support) are more successful in building new digital products than those that stick to monolithic departmental projects.

In summary, the experience of Amazon, Bosch, Lockheed, and others shows that any organization – no matter how large – can benefit from thinking small. By structuring internal innovation efforts as collections of startup teams, big companies can regain the speed and boldness of a startup. The next section will delve into one of the most critical enablers of such teams: hiring and talent practices that prioritize ability and passion over credentials and hierarchy.

Talent and Hiring: Skills Over Pedigree

Having the right people is the lifeblood of any innovation team. One clear insight from successful lean teams is that whom you hire (and how you hire) matters far more than how many you hire. Small teams by necessity need excellent, versatile talent – there is no room for “filler” roles or B-players when only 10 people are building a product. This is why startups and forward-looking companies place heavy emphasis on a candidate’s skills, creativity, and drive rather than formal qualifications or years of seniority. In contrast, traditional hiring in engineering firms often filters for specific degrees, titles, or experience at well-known companies, which can cause them to overlook high-potential innovators who don’t fit the usual mold.

Silicon Valley’s leaders have been outspoken about this. Elon Musk bluntly stated, “there’s no need to have a college degree… or even high school” to work at his companies, as long as someone can demonstrate “evidence of exceptional ability” in the field businessinsider.combusinessinsider.com. In practice, Musk’s hiring interviews famously focus on having candidates explain how they solved hard problems in the past, looking for a track record of scrappiness and ingenuity. This mindset – prove what you can do, don’t just tell me what’s on your diploma – has permeated many top tech companies. Google, for example, analyzed its hiring data and discovered that academic performance had little correlation with job success. Laszlo Bock, Google’s former HR chief, noted that “GPAs are worthless as a criteria for hiring” and that many of Google’s best employees did not come from elite universities twentyonetoys.com. By 2013, Google had teams where 14% of members had never even attended college twentyonetoys.com, thanks to its shift toward skills-based evaluation. Google started focusing on work sample tests and general cognitive ability rather than pedigree, a practice that other tech firms followed. This opens the door to gifted programmers, designers, or data scientists who may be self-taught or come from unconventional backgrounds but can outperform those with fancier credentials.

Lean teams also tend to hire for cultural fit and passion for the mission. Every member of a 10-person team hugely impacts team dynamics, so factors like curiosity, grit, and collaborative attitude are critical. Companies like ElevenLabs and Anysphere look for “builders” – people who have demonstrable side projects, open-source contributions, hackathon wins, or other evidence that they love to create and solve problems. In fact, Anysphere’s recruiting process, as shown earlier, asks applicants to share a project they are proud of and emphasizes that the team is looking for passionate, truth-seeking, creative individuals who thrive on challenging problems anysphere.incanysphere.inc. Notice what’s missing: any mention of needing X years of experience or a specific credential. The implicit belief is that an outstanding new graduate or self-taught coder who has built impressive things can be more valuable than a 15-year veteran who checked the traditional boxes. Startups often cannot afford hefty salaries for “brand-name” hires in any case, so they focus on raw talent and willingness to grow. This scrappy approach frequently yields a team with diverse backgrounds but a shared dedication to the product’s success. It also creates a culture where contribution is king – respect is earned by what you deliver, not your job title. Members of such teams are more likely to challenge ideas regardless of hierarchy, which is healthy for innovation.

Established firms can learn from this by broadening their hiring criteria for digital talent. Rather than filtering out, say, an excellent coder because he or she doesn’t have a Master’s degree, consider giving them a chance via a technical challenge or project audition. Some corporations have started initiatives like coding bootcamp hiring programs or partnerships with hackathons to spot non-traditional talent. The goal should be to find people who demonstrate the capability and drive to build great products. An additional benefit of focusing on skills over pedigree is increasing talent diversity – you’ll tap into candidates from different industries, age groups, or educational paths who bring fresh perspectives.

Finally, once on the team, continuously developing talent is important. Lean teams invest in learning: members wear multiple hats, picking up new skills in design, DevOps, or AI as needed. Companies that allow employees to spend part of their time on self-driven research or creative projects (like Google’s 20% time or hack weeks at many startups) find that those experiences can directly spark new product ideas. In short, to staff a high-impact innovation team, seek out people with a maker mindset and give them room to shine – rather than strictly those with the “perfect” resume. The next section provides practical steps for organizations to put all these principles into action.

Recommendations for Engineering & Manufacturing Companies

For high-tech engineering and manufacturing firms aiming to build competitive digital products, the above findings point to a clear mandate: foster small, empowered teams and equip them to innovate from within. This requires cultural and structural shifts, but the payoff is accelerated innovation and a fighting chance against more agile competitors. Here are specific recommendations to implement lean innovation in practice:

1. Identify and Empower the Internal Innovators: Look within your organization for entrepreneurial talent. Often, there are engineers or scientists who have creative ideas for new software tools or services but are buried in day-to-day operations. Create channels for these individuals to surface their ideas – for example, host internal hackathons or innovation challenges. (Facebook famously generated features like the “Like” button and Facebook Chat from internal hackathons where any engineer could prototype ideas wired.com.) Employees who show passion and viable concepts should be spun off into dedicated small teams to pursue them. Give these teams air cover from senior leadership so they can focus exclusively on the new project rather than being pulled back into routine work. Bosch’s “Grow” incubator is a model: it provides a sandbox for internal startups to develop ideas with funding and freedom bosch.us. Even if you can’t set up a full incubator, you can designate “tiger teams” and grant them special status to operate outside normal process on high-potential innovations.
2. Form Small, Cross-Functional Teams: When kicking off a new digital product initiative, deliberately staff it with a small, cross-disciplinary team (roughly 5–10 people to start). Ensure all key skills are represented (e.g. software engineering, UX design, domain expertise, product management), but avoid staffing more people than absolutely needed. It’s better to have a compact team of star players than a large team with diluted talent. Give this team end-to-end responsibility for the project – they should collectively handle everything from initial design to prototype development, user testing, and iterations. This mirrors the “two-pizza team” approach at Amazon where one small team owns a service entirely aws.amazon.com. Keep internal communication paths short: ideally the team sits together (physically or virtually) and has direct access to decision-makers. By containing the project within a tight unit, you minimize the coordination tax and maximize speed. As the project grows, resist the urge to add bodies; instead, let the team scale the product via technology. (Remember Brooks’s Law: adding manpower to a late software project makes it later.) If the scope truly expands beyond the team’s capacity, consider splitting into two focused teams rather than a bloated one.
3. Minimize Bureaucracy and Enable Autonomy: Audit your development process and strip away barriers that slow down small teams. This might mean relaxing some corporate rules for these innovation teams – for instance, fast-tracking procurement of software tools or cloud services they need, even if it bypasses the usual approval chain. Cut down on status meetings, reports, and multi-level approvals. Ideally, a lean team should have a direct line to an executive sponsor who can make quick calls on major decisions and shield the team from needless administrative tasks. Adopt Agile methodologies (scrum or Kanban) to let the team self-organize their work in sprints, rather than following a rigid long-term plan. The team’s mandate should be: build a minimum viable product (MVP) as quickly as possible, get it in front of users, and iterate. Tolerate some chaos in service of creativity. Big companies often smother new ventures with their regular processes — avoid this at all costs. Consider physically separating the team from headquarters or allowing a skunk-works style location, which symbolically and practically frees them to work differently. The Skunk Works example showed how powerful complete autonomy can be latimes.com. Modern companies can emulate this by carving out internal “innovation zones” where normal rules are suspended to let experimentation thrive.
4. Redirect Budget from Consultants to In-House Experimentation: It’s common for firms to spend large budgets on consultants, market research firms, or lengthy strategy projects when trying to go digital. Instead, reallocate those funds directly to your internal innovation efforts. For the cost of a typical multi-month consultant study, you could likely fund a 10-person development team for a similar period to build and test a working prototype. For example, rather than commissioning a consultant to write a 100-page report on “digital trends in smart manufacturing,” empower a small internal team to develop a beta version of a data dashboard or AI solution for your equipment. They will learn far more by doing, and you’ll have an actual product to show for it. This is the essence of lean startup methodology: use experimentation and real user feedback to guide decisions, not just upfront analysis. By putting budget into engineering prototypes and pilot programs, you signal that making is valued over meeting. This doesn’t mean never seeking outside advice, but consultants should play a supporting role (e.g., providing specific technical expertise or market connections) under the direction of the internal product team, rather than driving the vision. Trim down spending on generic “innovation workshops” and instead give that time and money to your own people to create something tangible.
5. Foster a Research-Driven, Experimental Culture: Perhaps most importantly, leaders must cultivate a culture where innovation is driven by curiosity and continuous learning. Encourage teams to stay up-to-date with the latest technologies (AI techniques, cloud services, open-source tools) and experiment with them. This could involve sponsoring employees to attend industry conferences, invest time in training, or collaborate with startups and academia. Make it acceptable to try bold ideas – and equally acceptable if some of those ideas fail. As Amazon’s Jeff Bezos put it, “failure and invention are inseparable twins.” He has emphasized that to innovate, a company must be willing to be wrong and keep experimenting, and Amazon strives to be “the best place in the world to fail” versionone.vc because each failure is seen as a learning step towards a big win. Engineering firms should take this to heart: shift the mindset from avoiding failure at all costs to failing fast and smart. This can be done by running lots of low-cost experiments (A/B test that new IoT app with a subset of customers, or pilot a machine learning model on one production line) so that even if one doesn’t pan out, the stakes are low and lessons are learned quickly. Celebrate teams that iterate and improve, not just those that deliver a perfect result on the first try. Moreover, integrate your R&D staff with product teams – let researchers and practical engineers work side by side in these small teams so that new ideas flow freely into development. Over time, a research-driven culture makes the organization more adept at spotting and seizing new opportunities, rather than clinging to “business as usual.”

By implementing these recommendations, high-tech companies can transform their approach to digital product development. The transition won’t happen overnight – it requires commitment from leadership to empower people and perhaps some retraining of managers used to more hierarchical control. But the case for change is compelling. When you cultivate internal startups and give them a clear mission, resources, and freedom, you create the conditions for the kind of breakthrough products that no amount of external consulting can buy. The final section concludes with the outlook for companies that embrace lean teams versus those that do not.

Conclusion

In an era where software and AI capabilities are becoming key differentiators even for traditional engineering companies, the ability to innovate quickly is paramount. This paper has shown that size and bureaucracy are the enemies of speed. The most successful digital innovators — from scrappy startups to enlightened large enterprises — have demonstrated the power of keeping teams small, hungry, and free from undue constraints. A mere dozen skilled individuals, given a supportive environment, can achieve in months what many firms struggle to do in years. By contrast, companies clinging to old ways of big committees and top-down planning often find themselves lagging, if not completely disrupted, when technology cycles accelerate.

The implications are clear: competing in the digital era means breaking from old molds. It requires trusting small teams to pursue ideas—even when their methods stray from standard protocols. It means hiring brilliant tinkerers and giving them room to explore, rather than demanding long résumés or bogging them down in rigid processes. Leadership must also embrace a new definition of success—one that values learning, customer impact, and agility over strict adherence to plans. The good news? These shifts are within reach. As the case studies in this paper show, established companies have deep resources and industry expertise. When paired with the lean team model, that foundation can unlock powerful creative energy from within.

In closing, building successful digital products is less a function of  “mass”—massive teams, mass budgets, mass consultants—and more a function of “mission”.  A small, mission-driven team that is empowered can move mountains. Instead of asking “How many people do we need for this project?” forward-looking companies ask, “Who are the right people, and how do we remove obstacles for them?”. The future belongs to organizations willing to act boldly on this principle. By doing so, even century-old engineering firms can rejuvenate themselves and remain formidable competitors in the digital product arena. Ultimately, the formula is simple: small teams, big innovations. Embracing it may well separate the industry leaders of tomorrow from the rest.

 Sources:

1. Entrepreneur – $2.6B AI Startup Didn’t Market AI, Gained a Million Users entrepreneur.comentrepreneur.com
2. The VC Corner – The Billion-Dollar Startup Formula: Why AI-Driven Small Teams Are Beating Giants thevccorner.com
3. Sacra – ElevenLabs Revenue, Valuation & Growth Rate (Oct 2024) sacra.comsacra.com
4. UpStarts – AI Startup Doss Raises $18M To Take On NetSuite's ERP upstartsmedia.comupstartsmedia.com
5. AIM Research – Lace AI Raises $14M after 1000% ARR Growth in 2024 aimresearch.coaimresearch.co
6. AWS Executive Insights – Amazon’s Two-Pizza Teams (2021)aws.amazon.comaws.amazon.com
7. Los Angeles Times – Lockheed's 'Skunk Works' Turns 50latimes.com
8. Business Insider – Elon Musk: You Don’t Need a College Degree to Work in Techbusinessinsider.combusinessinsider.com
9. TwentyOneToys Blog – Google Says Ignore GPA in Hiringtwentyonetoys.com
10. Anysphere (company site) – Careers – Research Engineer Posting anysphere.incanysphere.inc
11. Version One Ventures – Jeff Bezos on Experimentation and Failureversionone.vc
12. Bosch US – Bosch Named “Best Workplace for Innovators” (Internal Incubator)bosch.us
13. TechGig – Facebook’s Hackathons and Product Innovations wired.com
14. Standish Group via PM Essentials – Small vs Large Project Success Statistics pmessentials.uspmessentials.us
15. Forbes – Why Smaller Teams Are Better Than Larger Ones - https://www.forbes.com/sites/jacobmorgan/2015/04/15/why-smaller-teams-are-better-than-larger-ones/