The Missing Middle

Innovation and scale are only the means to an end.

Yet it sometimes seems like they’ve become the goal. The pathways movement celebrates strategies that scale, applauds bold approaches, and races to generate new ideas. And for good reason: transforming education and workforce systems requires innovative thinking and scale, and the work is urgent. But conflating innovation with improvement, scale with success, or novelty with nuance won’t help. Instead, doing so risks undermining our progress toward the outcomes we want for young people.

A confession: after our last blog post suggesting that it was time to take a step back, look at the data, and adopt evidence-based pathways strategies, we got worried that we might be accused of being anti-innovation. It shouldn’t feel professionally dangerous to say, “Let’s gather evidence to make sure what we’re doing will work.” But we suspect we’re not the only ones who have hesitated to raise questions about whether the pathways movement might sometimes pursue innovation for its own sake. To be clear: there’s lots of thoughtful pathways work happening across the country. Yet there’s also pressure—subtle and not—to innovate and scale quickly. That pressure can perpetuate the problem of linear pathways and make it hard to pause and ask whether we’ve identified the right problem, whether that problem requires an innovative solution, and whether we’re ready to scale.

(And as long as we’re confessing things, we might as well admit that we’d hoped this post would be shorter than our first one, but it’s not. It turns out that our avowed commitment to nuance and complexity is at odds with our aspiration for brevity.)

Innovation Theater and the Consequences of Skipping Steps

“Innovation theater” is a term that refers to flashy activities, like hack-a-thons or innovation labs, that some businesses use to signal cutting-edge thinking—but that yield little by way of real outcomes. When done right, these activities can be valuable, but innovation theater reduces them to performances, more concerned with signaling than with implementation. Innovation theater is a problem associated with large companies, especially in the tech sector, that are seeking to compete with start-ups. With the national pathways movement now firmly established, it seems we may be running into some of the same challenges common in established companies.

Maybe that’s not surprising, since the pathways movement has adopted some ways of working characteristic of Silicon Valley, including rapid prototyping and a “move fast” mentality. These approaches can work in tech, where first-to-market wins, and companies can often pivot after failures. When companies fall into the trap of innovation theater, they risk falling behind competitors or launching products that fail. We run a more serious risk: failing young people. Young people aren’t beta testers, and our public education and workforce systems—which were out of necessity built to support scale, not experimentation—can’t just pivot when an innovation fails. The phrase “innovation at scale” doesn’t fill us with excitement; it makes us nervous. Resisting the temptation to scale unproven pathways strategies is a sign of responsibility, not failure or inaction, even in the face of pressing challenges.

Jumping from idea to press release to scale—while skipping steps like piloting, iterating, and building infrastructure—turns the process of innovation into a performance of innovation.

Figure 1. One way of thinking about how innovation and scale fit into a larger process

(Notice how many steps there are? That’s not an accident.)

Avoiding innovation theater requires us not to skip the middle steps between “interesting idea” and “ready to scale.” Innovation often starts with good intentions and meets a real need for novel solutions, but it falters and turns into theater due to a lack of leadership commitment and dedication to the process needed to turn an idea into outcomes. Jumping from idea to press release to scale—while skipping steps like piloting, iterating, and building infrastructure—turns the process of innovation into a performance of innovation.

The antidote is a deliberate process that recognizes that innovation and scale aren’t the goal—and aren’t inherently superior to carefully designed pilots that build evidence about what works. The graphic to the left shows our back-of-the-envelope thinking about what that might look like. Each step serves a purpose, and the process is cyclical because improvement efforts surface new problems requiring innovative solutions. Skip steps, and you risk innovation theater.

Recently, we’ve seen several innovations, each based on good intentions and smart ideas, rush toward scale—with results that highlight the risks of skipping the middle steps in the process.

Lack of clarity about the problem: There’s been a rush to scale artificial intelligence (AI) tools across schools and businesses alike, often without a clear consensus on the problem they’re meant to solve or whether AI is the right tool to solve it. (Turns out “because AI” isn’t a sufficient answer to “why are we doing this?”) Now, schools struggle with growing digital equity gaps, educator overwhelm, lack of infrastructure, and data privacy, and companies are pulling back on AI. AI adoption rates are declining for large firms, and 42% of companies report that they’re abandoning the majority of their AI initiatives due to implementation challenges.

Lack of evidence: Short-term credential programs aim to solve a clear problem: many learners and workers seek to quickly gain the specific skills they need to access better job opportunities and adapt to the shifting labor market. That promise has led states to invest over $5.6 billion in short-term credentials, with hundreds of millions of federal dollars on the horizon via Workforce Pell. But the scant data we have on these programs suggests these scaling efforts are premature. With just 12% of certificates leading to real wage gains, over half of credentials misaligned to labor market demand, and employers confused by the sheer number of credentials and how they map to talent needs, the need for more careful design and strategic implementation is clear.

Lack of infrastructure to support scale: Skills-based hiring is a promising way to address the inherent inequity in a labor market that excludes capable workers because it prioritizes degrees over—or uses them as a sloppy proxy for—skills. In the last few years, large employers such as Google, Bank of America, and IBM, as well as 26 state governments, have publicly committed to skills-based practices. But the speed at which the movement has accelerated has left little opportunity to vet best practices and build infrastructure to support employers or engage the education field. We’re left with results that lag behind the rhetoric. Fewer than one in every 700 hires in the U.S. is actually made using skills-based practices, representing a change of only 0.14 percent in incremental hiring of candidates without degrees.

The urge to innovate and scale is usually rooted in admirable ambitions to achieve broader reach, transformation, and equity. But if we don’t test our ideas, refine them, and learn from failure, we could undermine our own efforts. Without evidence, we risk misunderstanding what works, scaling the wrong elements of our strategies, or failing to scale at all.

The pathways movement’s struggle to scale work-based learning over the past decade offers one example of the challenge. While apprenticeships and internships have growing evidence bases, we still need to know more about what works—and why—across the full spectrum of work-based learning models. It may be that scaling work-based learning has been so difficult because we aren’t focused on the right things. With more information, we can better codify what works and build the infrastructure—including intermediaries—needed to support scale.

The uncomfortable truth is that, while many pathways strategies may make intuitive sense, they lack strong evidence. We need more local and regional pilots to test and refine ideas and improve our knowledge of what works. And we need policies, incentives, and funding that support pilots and the time they take, as well as an honest recognition that some will fail, which is a far better outcome than a failure at scale with lasting effects on young people. Insisting on evidence isn’t stodgy; it’s a vital safeguard to protect young people and our limited resources.

Dual enrollment’s story offers an alternate vision for a way to steadily improve an evidence-based intervention and bring it to true national scale. Over the past quarter century, supporters of dual enrollment have designed, tested, and improved multiple models, efforts accompanied from the beginning by dedicated research that has built a strong evidence base. These results spurred federal and state policies and incentives to support dual enrollment. It’s been a long process—a glacial timeline by Silicon Valley standards, though that’s kind of the point—but dual enrollment has achieved genuine scale. All 50 states now have policies to support it, and in two states, dually enrolled students comprise over half of community college enrollment. Nationally, one in five community college students is dually enrolled, and the numbers are still growing rapidly.

We need policies, incentives, and funding that support pilots and the time they take, as well as an honest recognition that some will fail, which is a far better outcome than a failure at scale with lasting effects on young people.

For several years now, we’ve been on our soapbox about the need to build “systems, not programs.” Time to climb down and add some nuance. We fear that our emphasis on systems inadvertently reinforced a bias toward scale, as if programs were just steppingstones to the “real work” of building systems. But dual enrollment shows how programs and systems play different and complementary roles in the innovation-to-scale process. We still believe in the core ideas behind our “systems, not programs” mantra. To succeed, pathways can’t be a handful of cobbled-together or add-on programs; they require systems change and alignment. But pathways do and must include many programs that have a critical role in helping to design the new strategies we need. Programs can:

Create multiple ways of getting to the destination. Scaling pathways doesn’t have to mean that all young people have exactly the same experience. Multiple models and strategies can lead to the same destination—the urban planning mindset we described in our last post. We need high-quality programs that meet students where they are, not just homogenous systems that only work for some. Systems can play a role in building the infrastructure needed to bring together and navigate the multiplicity of options created by high-quality programs.

Become proving grounds for innovation. Programs are more nimble and flexible than systems. Their responsiveness to young people and to specific contexts makes them vital sites for testing and iterating on innovations that can then be scaled through systems.

Demonstrate what works. Some programs succeed exactly because they’ve chosen to maintain their original focus instead of scaling. These programs often go deep, building detailed knowledge about what works and why that can inform implementation elsewhere. Pressuring them to scale prematurely could mean that we lose both an excellent local program and the knowledge it could generate. There are other ways that programs can contribute to ensuring more young people benefit from proven practices. Instead of scaling, programs could instead adopt an “open source” approach to sharing their model—which can then be adapted by other programs and scaled through systems.

The Rubber Band Principle

Throughout this post, we’ve argued against binary thinking and false choices between innovation and evidence, pilots and scale, programs and systems, risk and responsibility, moving forward and pausing to learn. Like a rubber band, pathways strategies need tension to hold ideas together. A rubber band works because—not in spite—of tension. It’s useless if it’s too loose, but will snap if it’s too tight. The right amount of tension is what makes a rubber band functional. We need the same kind of calibration in our approach to innovation, evidence, and scale.

How might we approach things differently?

Practitioners: resist pressure to scale prematurely; “open source” your models and lessons instead.

Policymakers and funders: create incentives that prioritize rigorous pilots and evidence-building alongside innovation and scale.

Advocates and TA providers: support and document local pilots without pushing for premature scale.

Innovation and scale should solve problems, not create them. We are pro-innovation, which is essential for finding new strategies that meet the challenges of the pathways movement. And we are pro-scale, without which we’ll fall short of our ambition to support all young people, especially those furthest from opportunity. We are also pro-nuance, pro-evidence, and pro-responsibility. That means piloting rigorously, iterating honestly, building evidence systematically, and scaling strategically. We’re also advocating for the creation of conditions where saying “we need more time to learn” is seen as responsible, not timid, and where we celebrate programs that are building knowledge, not only increasing numbers.

To get there, we’re committing to asking uncomfortable questions (no, really, we’re fun at parties) that help distinguish between genuine innovation and innovation theater:

About problem definition: Are we solving a problem young people actually have, or one that reflects our own interests?

About process: Have we given innovations enough time to fail in small, contained ways, and are we learning from our failures? Are we scaling because we have evidence of effectiveness or because scaling strengthens the case for investment in our work?

About incentives: Are we innovating and scaling because it’s needed or because we think it will give us an edge in the national conversation about pathways? Are we designing this work to create pathways to thriving for young people, or are we responding to other pressures in our ecosystem—the need to show innovation, to compete for funding, to differentiate our organization?

We realize these questions surface tensions. That’s intentional, and it’s exactly what the rubber band principle is about. We need enough tension to maintain the shape and purpose of pathways strategies, but not so much that they snap. That means holding innovation and evidence together, staying curious about what works and honest about what doesn’t, and always keeping young people at the center of our work. The stakes are high: when we scale innovation without evidence, learners furthest from opportunity bear the consequences.

Young people deserve better than our best guesses about what will work. They deserve our discipline, our patience, and our willingness to be wrong—and admit it—in small ways before we risk being wrong at scale.

This post is the second in All4Ed’s Normal Gets Us Nowhere series, which seeks to ask hard questions, spotlight fresh data and thinking, challenge longstanding assumptions, and offer new approaches that go beyond tinkering in order to contribute to the development of the next generation of pathways strategies. Read the first installment, “Course Correction,” here. If you’re working to build better pathways systems, we’d love to learn more about what you’re up to and how we can work together, so please get in touch!