Building an Impact Index (and What Broke Along the Way)

Over the past 48 hours, we set out to answer a deceptively simple question so that we can better understand what resources might deliver impact to entrepreneurs — at least as a baseline — before we see what users of our site actually find impactful.

The burning question is this: how do you measure impact across entrepreneurial resources without relying on hype? What started as a technical experiment turned into something much more interesting.

The Initial Idea: Measure the Web

Over years of stumbling upon meaningful content, I’ve wondered whether there might be a way to identify and validate powerful resources across the internet. The instinct was straightforward: if something is impactful, it should leave traces. Mentions. Citations. Discussions. Appearances.

So we (me and my buddy Chat) tried to tap directly into Google’s Custom Search API to count global web presence. It looked elegant on paper. It broke almost immediately — after hours down technological rabbit holes and chasing things that didn’t work. Expired API keys. Permission errors. Quota ceilings. Deprecated behaviors.

We spent hours circling something that no longer behaves the way it once did. Ah, entrepreneurship.

At first, it felt like failure. But the friction forced a better question: are we trying to measure the whole web — or the right web?

The Pivot: Define the Ecosystem First

Instead of “the internet,” we defined a surface list aligned with small business operators:

• Entrepreneur.com
• Inc.com
• SCORE.org
• SmallBizTrends
• Shopify Blog
• Reddit
• Hacker News
• .edu syllabi

No prestige weighting. No venture capital bias. No hierarchy of authority. Just binary presence.

Does this resource show up in the ecosystem that small business owners actually inhabit?

That constraint changed the project from scraping to curation.

What We Built

We integrated Serper.dev into Apps Script and ran bundled surface queries for roughly 150 book resources.

Each resource now has:

• Binary presence flags (0/1 per surface)
• A total surface presence count
• Stored example URLs for auditability
• A timestamped snapshot

Nothing weighted yet. Nothing normalized yet. Just raw ecosystem footprint.

Then we sorted by presence — if a resource showed up in any reference across those surfaces. That’s when it got interesting.

What Books Rose to the Top (Presence = 6)

• Built to Sell
• Mindset
• Profit First
• The Lean Startup

Not billionaire memoirs. Not startup theater. Operator books. Sellability. Cash discipline. Psychology. Iterative testing.

It was reassuring. The surface list was aligned with my own thinking — although some of my absolute favorites, such as The Universal Traveler, received very low presence scores.

Recommended books from people we trust often scored lower than classics. At first glance, that looked like a flaw. But it wasn’t.

Presence measures longevity × penetration.

As a publisher, I know how much momentum successful books carry — and how hard it is for truly great books by lesser-known authors or publishers to break through. Classics accumulate density over decades. Newer or niche books often don’t.

Presence is not utility. It’s footprint. And that distinction matters.

The Real Hypothesis Emerging

We now have an external baseline layer. The next layer will be internal — once we begin seeing how people actually use the site:

• Shares
• Saves
• Click-throughs
• User feedback

Here’s the hypothesis (add yours in the comments): internal virality will correlate with external presence — but not perfectly. And the interesting discoveries will live in the divergence:

• High presence / low sharing
• Moderate presence / high sharing
  

Those are the signals that shape the canon.

What Looked Like Failure

The Google API detour wasn’t wasted time. It forced clarity. We stopped trying to measure everything and started measuring something intentional: small business–aligned ecosystems, binary presence, and frozen baseline snapshots.

Sometimes failure is just a forced refinement.

What’s Next

• Add Google Books ratingsCount
• Add YouTube discourse count
• Normalize within resource type (mean = 50)
• Freeze the external baseline
• Layer in transformation feedback

Then we watch patterns, not proclamations.

The Deeper Insight

Impact is never just fame. As R.W. Emerson says in Self-Reliance, “To be great is to be misunderstood.”

Impact is:

• Visibility
• Circulation
• Lived transformation
• Shared language

We’re beginning to measure the first two. The rest will emerge.