I have worked with NSF-funded researchers for nearly twenty years. More than 14,000 academics across every scientific discipline. I have read more broader impacts statements than I can count, and almost all of them follow the same pattern. The intellectual merit section gets weeks of careful thought. The broader impacts section gets a paragraph the night before submission. Something about presenting at conferences. Something about peer-reviewed journals. Maybe a vague mention of a website. Occasionally, a promise to visit a local school.
Then the grant is awarded. The research begins. The broader impacts plan sits untouched until the annual report is due.
I get why this happens. Researchers are hired and trained to do research. The broader impacts requirement can feel like an administrative add-on to a process that already demands everything you have. But NSF does not treat it as an add-on. It treats it as one of only two criteria that determine whether your proposal gets funded. Equal weight to intellectual merit. That has been the case for over two decades, and the way most researchers approach the requirement has barely changed since it was introduced.
I have spent my career in the space between research and the public. I have watched billions of dollars of publicly funded science produce findings that almost nobody outside the field ever hears about. Not because the research was unimportant, but because nobody invested in making it visible. That is the gap Scrolly Science was built to close. And NSF's own rules make the case for it better than I ever could.
One of my earliest experiences with NSF-funded work stays with me. Back in 2009, I worked with a research team at Louisiana State University on getting their findings in front of a wider audience. This was before social media had become a serious distribution channel for science. Gaining real visibility for research was slower and harder. You pitched outlets individually, built relationships with editors, followed up repeatedly, and hoped the timing worked. We pushed their work out to multiple global outlets over several weeks, and eventually a handful picked it up. The piece ran across several science websites and research forums, and over the following months it generated somewhere in the region of 40,000 reads and a string of inbound enquiries from researchers at other institutions who wanted to collaborate. For an academic team working quietly out of Baton Rouge, that kind of reach was genuinely new. It was modest by today's standards, but it proved something that has driven everything I have done since: the research was always good enough to find an audience. It just needed someone to carry it there.
What NSF Actually Requires
Every NSF proposal is evaluated on two criteria: Intellectual Merit and Broader Impacts. Equal weight. Reviewers are asked to evaluate your plan against five questions, and the first one is blunt: what is the potential for the proposed activity to benefit society or advance desired societal outcomes?
Two of NSF's desired outcomes are particularly relevant to how research gets communicated.
The first is increasing public scientific literacy and engagement with STEM. Not publishing another journal article that will be read by the same fifty people who already work in the field. Not presenting at a conference attended exclusively by specialists. Reaching people outside academia. The public. Policymakers. Journalists. Funders. People who need to understand what publicly funded science has produced, in language they can actually follow.
The second is broad dissemination to enhance scientific and technological understanding. NSF's own guidance on this is remarkably specific. It lists examples including publishing in diverse media to reach broad audiences, creating websites enhanced by engaging animations and films, presenting research in formats useful to policymakers and the general public, and developing new formats for communicating research to wider audiences.
Films. Animations. Websites. Diverse media formats. This is not buried in a footnote. It is in NSF's own broader impacts documentation. The Foundation is explicitly asking you to communicate your work in formats that go well beyond traditional academic publishing.
Most researchers either do not know this, or know it and treat it as aspirational. In my experience, the latter is more common. They know the expectation exists. They just have no idea how to deliver on it, and no partner to help them.
The Reporting Problem
NSF does not just ask you to promise broader impacts.
It asks you to prove you delivered them.
Every year, you file an annual project report describing your broader impacts activities for that period. NSF's guidance says this should include websites, social media, written materials, conference presentations, and public engagement activities. If you promised to enhance public understanding of your research and you have nothing to show for it, that gap is visible to your programme officer. It is not a gap you want.
When your grant ends, you submit a Project Outcomes Report. This is a public document, posted verbatim on Research.gov for anyone to read. Written for a lay audience. No jargon. Plain language. It must describe what your project achieved in terms of both intellectual merit and broader impacts. It stays online indefinitely. It is the permanent public record of what your NSF funding produced.
And when you apply for your next grant, you report on results from prior NSF support, including broader impacts. Reviewers look at what you actually did, not just what you promised. A PI who invested in genuine public-facing communication has demonstrable evidence. A PI who listed conference presentations and left it at that has a weaker case. That is not my opinion. That is how the review process works.
The cycle is clear: promise broader impacts to get funded, deliver them to report successfully, demonstrate them to get funded again. Most researchers are strong on step one and weak on steps two and three. That weakness compounds.
The Budget Line Most PIs Forget Exists
The most common thing I hear from researchers is that they do not have the budget for proper science communication. No money for a film. No money for a professional website. No money for content that reaches beyond the university press office.
This is almost always wrong.
The NSF Proposal and Award Policies and Procedures Guide, Chapter II, includes a dedicated budget line for dissemination. Line G2 on the proposal budget explicitly allows funds for documenting, preparing, publishing or otherwise making available to others the findings and products of the work. The guidance specifies that this covers illustrations, websites, data visualisation, and making research accessible in formats beyond traditional publication.
You can budget for science communication directly in your NSF proposal, and the guidance explicitly supports it as a stated purpose of that budget category. If you are not including dissemination costs in your proposals, you are leaving money on the table and simultaneously weakening your broader impacts plan. If your current award included broader impacts activities involving public communication, the resources to deliver should already be in your budget. The question is not whether the money is there. The question is whether you are spending it.
I saw this firsthand in 2018. We started working with an engineering research group out in California who had no idea their NSF budget could cover video production. They had been sitting on dissemination funds for two years, unspent, while filing annual reports with thin broader impacts sections. We produced a 90-second explainer film that broke down their work for a general audience. Honestly, the subject matter was dense, complex structural engineering, but that is exactly the kind of research that benefits most from visual storytelling. The film entered a science communication competition and won. Within a few weeks of release it passed 100,000 views across platforms, which for an engineering research team was unheard of. But what really mattered was the programme officer's response. Their annual review came back with an outstanding rating for outreach and public engagement, specifically citing the video and the measurable reach it generated. That rating followed them into their next proposal. One film, produced from budget they already had, changed the trajectory of their broader impacts reporting for years.
CAREER Grants Set an
Even Higher Bar
If you hold an NSF CAREER award, broader impacts are not an add-on. They are supposed to be woven into the entire programme. CAREER awards are designed to integrate outstanding research with education and outreach, and reviewers evaluate both as a unified whole. The outreach cannot quietly underperform.
I have worked with a lot of CAREER award holders, and the pattern is almost always the same: genuine desire to do meaningful outreach, paired with no realistic plan for executing it at the standard NSF expects. Early-career faculty are simultaneously building a lab, supervising students, teaching, publishing, chasing tenure, and somehow also running a public engagement programme. Nobody can do all of that and also become an expert science communicator on the side. It is not a reasonable expectation.
What is reasonable is bringing in a partner who has spent twenty years doing exactly that. A professionally produced explainer film, a plain-English summary with social distribution, a web presence that actually appears in search results: these are concrete, demonstrable broader impacts deliverables. They strengthen annual reports, Project Outcomes Reports, and future proposals. For an early-career researcher building a tenure case, they provide evidence that is difficult to argue with.
What Good Broader Impacts Dissemination Looks Like in Practice
Short films that explain your research to non-specialists. Not a recorded conference talk. Not a talking head in front of a whiteboard. A structured, professionally produced piece that someone outside your field would actually choose to watch. Sixty to ninety seconds. Clear narrative. Visual clarity. Something a programme officer, a funder, or a congressional staffer can watch before a meeting and understand what your research does. I have been producing these since 2006, and the difference between a professional explainer and a homemade video is not cosmetic. It is the difference between something that gets watched and something that gets skipped.
Written content in plain English. Not your abstract. Not your paper. A professionally written summary that a journalist can reference, a policymaker can scan, or a collaborator at another institution can use to understand your work without specialist training.
A digital presence that reflects the quality of your work. Not a page buried three levels deep on your university website. Not a WordPress site last updated in 2019. A structured, searchable, properly optimised web presence that appears when someone actually searches for your research area.
Social media content designed for the platforms your audience uses. An institutional YouTube channel with 400 subscribers is not a dissemination strategy. Content formatted for the platforms where your audiences spend their time, designed from the start for those platforms, promoted strategically and sustained over time: that is a dissemination strategy.
And measurable reach. NSF asks you to describe outcomes. "We put a video on our website" is not an outcome. "Our explainer film reached 50,000 people across three platforms, generating 2,400 engagements and coverage in two outlets" is an outcome. The difference matters when you are writing your annual report, your Project Outcomes Report, or your next proposal.
The Distribution Problem
Nobody Has Solved
This is what I care most about after nearly twenty years in this space, and it is where most science communication providers fall short.
The sector has invested almost everything in creation and almost nothing in making sure anyone sees the work. I have watched it happen thousands of times. An institution commissions a beautiful animation. It gets uploaded to an institutional channel. A handful of people see it. The broader impacts evidence it was supposed to generate never materialises, because the content was never actually distributed to the audiences who were supposed to see it.
A beautiful animation that nobody watches is not broader impacts. It is expensive decoration.
I built Scrolly Science to address this directly. Every project we deliver includes distribution as standard. We do not hand over a video and wish you luck. We handle the strategy, the scheduling, the platform-specific formatting, the captioning, the promotion, and the performance reporting. When we produce content for a researcher, we run the sustained campaign that turns a single piece of content into measurable public engagement. That is how we have generated over half a billion impressions. Not through hope. Through infrastructure.
This matters for NSF reporting because it gives you real numbers to cite. Not vanity metrics. Real engagement data showing who saw the work, where they saw it, and how they responded. The kind of evidence that satisfies a programme officer, strengthens a renewal application, and demonstrates that NSF's investment in your research produced genuine public benefit.
Your Grant Is Ending and Your Broader Impacts Evidence Is Thin
If your NSF grant ends in 2026 and you have not yet invested in broader impacts communication, the window is closing. But it is not closed.
A Scrolly Article, a professionally written plain-English summary of your research with social-ready versions for LinkedIn and X, takes four working days and starts from $500. A Scrolly Short, a 15 to 30-second vertical film designed for social platforms, takes five working days. A Scrolly Explainer, a structured 60 to 90-second narrative film, takes two to three weeks. All produce concrete, citable broader impacts deliverables you can reference in your Project Outcomes Report, your annual report, and your next proposal. All include distribution. All come with measurable reach data.
The NSF broader impacts requirement is not going away. The researchers who invest in genuine, professional, distributed science communication will have stronger reports, stronger renewal applications, and stronger public profiles. The ones who keep treating broader impacts as a paragraph they write at 2am will not.
If your research deserves to be understood by people beyond your field, and NSF requires you to demonstrate that it was, we should talk.