Why Good Ideas Get Trapped and How to Advance Them in the Age of AI

There is a very quiet type of failure in the content creation profession: you produce something decent, thoroughly researched, and then… no one uses it. No public outcry, no debate, just silence. A good idea usually doesn't die because it was rejected. It dies because it failed to reach the right people.

Junjie Ren — a senior communications manager at a major public policy research institute in Washington and editor-in-chief of a publication on development economics — calls this phenomenon by a very concise formulation: crafting a good idea and making that idea go far are two entirely different problems. He was promoted to a management role at the age of 23 and has worked with Economist Impact and the United Nations Capital Development Fund. But what stands out most about him is a principle that is very difficult to maintain: content can change its shape, but it must never dilute its substance.

Policy Communications Is Not Corporate PR

Most communication advice in the world is written for a business trying to make sales. A policy research institute faces a tougher challenge: it must reach a handful of people capable of taking action, and it must arrive with the content completely intact.

Ren describes his work as not starting with the question "what to write," but with "understanding what": understanding the political and policy landscape, understanding available resources, and understanding the tightening constraints. Corporate PR is optimized for reach, emotion, and conversion. Policy communications cannot rely on those metrics. The central question is always: who specifically needs this discovery? A congressional aide, a multilateral official, or a journalist specializing in that exact field?

• Businesses typically measure success by reach, engagement, and conversion.
• Research institutes measure success by whether an idea enters the room where decisions are made.

The Real Metric Is Not Reach

This is the most easily overlooked point when content teams copy each other's communication formulas. One million views can be an achievement — or it can be a sign that you have just spoken eloquently to a room where no one has the authority to change anything.

For Ren, the metric is an uncomfortable yet useful question: did this idea make it into the room where decisions are made? That may sound distant for commercial content creators, but upon closer inspection, it applies almost entirely to a B2B blog post, a whitepaper, or a brand campaign. If the right reader never opens it, you are running very fast in the wrong direction.

A simple practical exercise: before writing, picture the exact single person who will read it. Not a generic "small business owner," but "Ms. Lan, owner of a twenty-person garment workshop, reading on her phone at six in the morning with a cup of coffee in hand." That detail dictates the format, length, and even your choice of words.

Why Good Ideas Evaporate Upon Leaving the Door

Ren states bluntly: for policy organizations, failure usually does not lie with the general public. The public shapes public opinion on a large scale, but the people who can truly take action are policymakers — and that is the target audience of the research. Research packed with evidence will evaporate in a competitive marketplace of ideas if it lands in the wrong place, takes the wrong shape, or fails to resonate with how people consume information.

He echoes Marshall McLuhan's philosophy — the medium is the message — but adds a crucial condition: the message must still remain intact. Forms change constantly, and you are forced to adapt, but do not let that adaptation dilute the core you are carrying.

You are allowed to change the container. You are not allowed to change what is inside.

In reality, most corporate content fails due to one of three mistakes: reaching the wrong person, arriving in the wrong format, or reaching the right person but failing to trigger the motivation that makes them want to act.

Choosing the Right Format for the Right Idea

Ren does not believe in a fixed formula. Podcasts are conversational, well-suited for human stories and setting the context, while also capable of humanizing the person behind the idea in a way other formats struggle to do. But they also consume real time and require a narrative arc strong enough to carry listeners all the way through.

Video interviews are similar to podcasts in format: a skilled interviewer guides viewers along the exact path of reasoning taken by the interlocutor, helping them understand why a conclusion exists rather than just hearing the conclusion itself. Explanatory articles serve an entirely different group: people seeking answers who need something durable, a record they can point others to later.

• Podcasts: for those who want to listen and need context and emotion.
• Video interviews: for those who need to understand how a conclusion was formed.
• Explanatory articles: for those searching and needing something referenceable.

The right question is not "which channel is trending," but "how does this person consume information, and what decision are they trying to make?"

The Most Common Mistake: Believing People Will Read

Ren puts it briefly: experts and organizations often assume someone will read their work. In reality, the first audience is usually a search engine or an AI crawler, not a human. Without inherent reputation and search rankings, genuinely good content can still be buried.

Another mistake is experts writing for other experts. That builds intra-industry prestige, but does little to reach the broader public that policy work aims for. At the opposite extreme, chasing sensational formats to force attention diminishes brand credibility.

The balance lies here: understand where readers are looking, and rely on what determines whether something is worth their time.

AI Fills Every Channel — Reputation Becomes a Scarce Asset

Currently, readers can still partially distinguish human-written content from "AI junk" — sometimes just by flat tones and characteristic long dashes. But Ren notes that window is narrowing. As boundaries blur, the differentiators will be provenance and a proven track record before an increasingly skeptical public.

The lesson for businesses building content operations: research institutes build reputation very slowly, through consistent standards and a body of work that withstands the test of time year after year — not through a single viral post.

Invest in editorial discipline and a track record, rather than chasing output.

That kind of accountability is very difficult to maintain. And precisely because it is hard to maintain, it is hard to fake at scale — so as AI-generated content grows, its value only increases.

Advice for Beginners: Understand Deeply Before Simplifying

Ren advises: get close to the core before trying to translate it for others. You cannot simplify what you do not yet understand, and readers can tell the difference between clarity and oversimplification. Spend real time with the research and the people who created it before writing for the external public.

After that, study your readers with the same level of seriousness. Who are they, what do they already believe, how do they consume information, and what decision are you helping them make? Many communications professionals prematurely default to choosing a trending format instead of stopping to ask what readers actually need.

If I had to take a single sentence from this conversation back to my office, it would be: do not measure by reach, measure by whether your idea reaches the room that needs it. It sounds small, but it changes almost every subsequent decision — who you write for, how long you write, which channels you choose, and what you let go of.

What about you? In your work, where is your "decision-making room" — and have you ever written something great but sent it to the wrong room? If you have an idea stuck in a drawer, try telling me who it needs to survive. Who knows, your next reader might be right down there in the comments section.