How to Build Innovation That Actually Scales
Nicolas Landry, MBA, PMP | North Ridge Advisory Group
Taking a great idea and scaling it across an organization is about proving you can deliver real value over and over again, even when conditions change. A proof of concept shows that an idea can work in theory. But scaling is where you find out if people will actually embrace it, whether your operations can support it, and if the economics still make sense as you grow.
For project leaders and executive teams, this phase brings some of the most critical choices you'll face: Who really owns long-term adoption? What capabilities do we need to build out next? And what kind of real-world evidence do we need before making our next big commitment?
If research across corporate innovation, agile transformations, and major capital projects teaches us anything, it’s that you have to set up the conditions for scaling while you're still actively testing and fine-tuning the innovation itself.
Plan for the business beyond the pilot
In their work on innovation scaling, Binns and Griffin (2023) frame scaling as a dedicated discipline right alongside ideation and incubation. They urge leaders to start with the ultimate end goal in mind and map out the exact resources, channels, and capabilities needed to reach customers, deliver seamless experiences, and handle real scale. Those assets might come from internal teams, strategic acquisitions, existing operations, or external partnerships.
This fundamentally changes what should go into your business case. A lean prototype budget can prove technical feasibility, but it rarely covers the training, rollout, integration, and ongoing support needed for full-scale adoption. Incorporating a stage-gate funding approach lets you evaluate clear results at every step before unlocking the next wave of capital. When these downstream requirements are ignored early on, a pilot that looks like a wild success can quickly turn into a costly request for extra budget.
Here’s a quick example: imagine an automated reporting tool that works well in one unit simply because a dedicated analyst spends hours cleaning data behind the scenes every week. Before rolling that tool out to twenty more teams, leaders need to ask: Can we automate that data cleanup? How do data structures vary across teams? Who is going to maintain this tool over time? Without answering those questions, expansion just multiplies the hidden effort that made the initial test look so smooth.
A great starting habit is defining success in clear operational terms right out of the gate. Be clear about who you're serving, the exact metrics you expect to improve, your target unit economics, and the specific triggers that would prompt you to pause or adjust course. Your path will naturally evolve as you gather more data, but clear goals keep everyone aligned.
Make delivery repeatable and adaptation deliberate
As Flyvbjerg (2021) points out, using modular, repeatable components and fast iteration cycles dramatically reduces the risk of large-scale initiatives. Repeating proven building blocks lets you leverage learnings from one phase and apply them immediately to the next. And when modular pieces can deliver value on their own, you start seeing returns long before the entire project is finished.
In large-scale systems integration programs, I put this principle into action by establishing reusable data migration playbooks, core business process templates, and clear acceptance criteria; refining them with every successive deployment phase. Simply slicing a timeline into phases doesn't help much if every single phase still requires starting from scratch with custom designs and heavy rework.
At the same time, repeatability needs to leave room for genuine local needs. Parvarandeh et al. (2023) highlight how great local innovations often get stuck in silos because individual business units overestimate how unique they are, lack reasons to share, or simply don't know what peer teams have already created. The flip side is just as common: central corporate teams can easily underestimate local market nuances and the specific workflows needed to maintain a competitive edge on the ground.
Smart leadership means agreeing on what must stay consistent across the company versus what can be tailored locally. Standard data definitions might be essential for executive reporting, whereas local training materials and minor workflow tweaks might benefit from flexibility. Every deployment should test whether those choices protect the core benefit. Sometimes, you'll find an innovation is best kept right where it started.
Give people the authority and capacity to scale
Rigby et al. (2018) emphasize that scaling agile practices requires adapting the broader organization surrounding those teams. A project team can move at lightning speed, only to get stalled by untimely decision making and supporting daily operations. They also caution against shielding early pilot teams with developing shortcuts that can't scale across the rest of the business.
The key takeaway is to look closely at the entire path from idea to day-to-day execution. If a team builds a feature in two weeks but waits six weeks for corporate approval, adding more development teams just builds a bigger bottleneck. Executive sponsors need to focus on removing those systemic friction points and empowering teams to make operational decisions on their own.
Drawing on extensive case studies at Procter & Gamble, Koen et al. (2024) highlight the vital role of active executive support, committed resources, and dedicated "integration leaders" who bridge project teams with senior leadership and key stakeholders. They also stress bringing commercial and go-to-market expertise into the fold early, while customer insights can still shape the product.
Hill et al. (2026) build on this idea with the concept of "bridgers"; leaders who excel at uniting diverse partners, translating different team priorities, and creating strong collaborative trust. For project managers, this is a powerful opportunity to step up. Often, what looks like resistance to change is really just a reasonable concern about system stability or local performance targets.
Project leaders are in a prime position to surface those underlying tensions and work out realistic solutions. But executives must back them up with the real authority and resources needed to deliver on those agreements. Expecting a project leader to manage a major rollout while leaving staffing, metrics, and operational ownership floating in uncertainty creates a major hurdle to value delivery.
Learn from Operation Warp Speed
Operation Warp Speed offers a masterclass in scaling under pressure: start preparing your delivery channels while you're still testing the solution. During the pandemic, vaccine manufacturers began scaling up production during clinical trials, long before knowing if their vaccines would be approved. By leveraging proven tech platforms and running developmental phases in parallel, they drastically compressed normal timelines (U.S. Government Accountability Office [GAO], 2021).
That speed came with a calculated risk. Investing early meant risking significant resources on candidates that might never cross the finish line. At the same time, public and private sector partners collaborated to expand manufacturing facilities, secure supply chains, and address staffing bottlenecks. By late January 2021, over 63 million doses were released, even as manufacturing hurdles continued to be solved in real time (GAO, 2021).
For business leaders, the big takeaway is to decide which preparations are worth kickstarting before the final sign-off. While your pilot is running, you can start building training programs, vetting vendors, or mapping out system integrations. Doing this work early cuts down lag time when greenlit, as long as you thoughtfully weigh the time advantage against the financial risk of unused work.
Most organizations won't operate with the sheer urgency or funding of a global crisis. Still, the underlying principle holds true: plan for rollout early, and make clear, deliberate decisions about how much to invest while uncertainty is still in the room.
Measure whether the benefit survives expansion
In a study tracking an AI assistant rolled out to 5,172 customer support reps, issue resolution speed increased 15% on average, though results varied widely depending on worker experience. That study offers great proof of value in a specific context, but it doesn't guarantee an automatic return on every AI tool everywhere (Brynjolfsson et al., 2025).
When evaluating your own programs, dig into results by specific user groups and operational settings. High-level averages can hide where a tool shines, where it falls flat, or where extra coaching is needed. Keep an eye on user adoption, service quality, true operating costs, and speed, then double-check if those positive results hold up as new teams onboard.
Putting it all together, here are five core questions to guide any scaling decision:
What real evidence proves that our benefits will hold up beyond the initial pilot team?
Which parts of the solution can we standardize and repeat, and which need local customization?
Who truly owns adoption and long-term performance, and do they have the bandwidth to deliver?
What bottleneck will limit our next expansion phase, and who has the authority to solve it?
What metrics will tell us whether to keep scaling, pivot our approach, or stop?
My main takeaway is that project leadership is central to making innovation sustainable. We are in a unique position to help organizations uncover hidden dependencies early, forge meaningful commitments, and carry valuable lessons from one deployment to the next. The next time a pilot gets the green light to expand, making sure those leadership responsibilities are funded and assigned is just as crucial as investing in the technology itself.
References
Binns, A., & Griffin, C. (2023). The missing discipline behind failure to scale. MIT Sloan Management Review, 64(4), 40–45. https://shop.sloanreview.mit.edu/store/the-missing-discipline-behind-failure-to-scale
Brynjolfsson, E., Li, D., & Raymond, L. (2025). Generative AI at work. The Quarterly Journal of Economics, 140(2), 889–942. https://doi.org/10.1093/qje/qjae044
Flyvbjerg, B. (2021, November–December). Make megaprojects more modular. Harvard Business Review, 58–63. https://hbr.org/2021/11/make-megaprojects-more-modular
Hill, L. A., Tedards, E., & Wild, J. (2026, March–April). Why great innovations fail to scale. Harvard Business Review. https://hbr.org/2026/03/why-great-innovations-fail-to-scale
Koen, P., Sheth, A., DiPaola, M., & Hill, L. A. (2024, November–December). Scaling up transformational innovations. Harvard Business Review. https://hbr.org/2024/11/scaling-up-transformational-innovations
Parvarandeh, S., Colella, K., & Duncan, D. S. (2023, June 16). How to scale local innovations in big companies. Harvard Business Review. https://hbr.org/2023/06/how-to-scale-local-innovations-in-big-companies
Rigby, D., Sutherland, J., & Noble, A. (2018, May–June). Agile at scale. Harvard Business Review. https://hbr.org/2018/05/agile-at-scale
U.S. Government Accountability Office. (2021, February 11). Operation Warp Speed: Accelerated COVID-19 vaccine development status and efforts to address manufacturing challenges (GAO-21-319). https://www.gao.gov/products/gao-21-319
