I check the New York Times bestseller list almost every week. And every week I close the tab thinking the same thing.

This doesn’t tell me what people are actually reading.

It tells me what they bought.

(That’s the entire problem, but it took me a long time to articulate it that way.)

A book in a cash register is not a book on a nightstand

Bestseller lists measure sales. The NYT panel counts units reported by a network of stores. Amazon ranks units shipped from one store. The lists are real and useful for what they measure, but they don’t measure reading.

A novel that sold like crazy six weeks ago and is now in everyone’s “currently reading” stack? Invisible. A backlist title that’s quietly being passed between coworkers because the show came out? Invisible. A nonfiction book that hit #4 in February and is genuinely shaping how a million people think this month? Invisible.

The cash register and the nightstand are two different surfaces. Bestseller lists are a measurement of the first. Almost nobody is measuring the second.

So we built something to measure the second.

What we mean by “reading culture”

Books move through a culture in a way that doesn’t track 1:1 with what’s at the register in a given week. The book your friend started this weekend. The audiobook three people you trust mentioned in the last month. The title that’s suddenly all over a corner of the internet you happen to spend time in.

That movement is the thing. The book that’s in cultural circulation right now is the book worth knowing about, whether it sold yesterday or two years ago.

We’ve started calling that signal reading culture. It’s distinct from retail sales (NYT’s frame), member ratings (Goodreads’ frame), or unit velocity (Amazon’s frame). Reading culture is the answer to a different question: not “what is being bought,” but “what is being read, talked about, and passed along.”

That question deserves its own measurement. We didn’t see one that existed. So we’re building it.

Introducing The Siftivo Reading Index

The Siftivo Reading Index is a weekly chart of the books in heaviest cultural circulation. Two sibling charts ship every Friday morning ET: Fiction Top 25 and Nonfiction Top 25.

The Index runs on real signal from real sources, aggregated with three things layered on top: position weighting (a #1 ranking carries more weight than #15), per-source weighting (some sources are more representative of reading culture than others), and momentum tracking (a book climbing matters differently than a book holding). Every entry on the chart carries weeks-on-chart, peak rank, and debut detection.

The math is versioned. Every chart we publish is tagged with the formula version that produced it.

What this is and what it isn’t

I want to be honest about where we are.

This is a starting baseline, not a finished product. The source mix and the weighting formula will refine over time as we layer in more signal and tune the math against what’s actually working. Each refinement gets published as a new methodology version, and the full methodology will become public as it firms up.

We’re building this in the open. The chart you’ll see this Friday is the visible artifact of a data set we’d otherwise be sitting on. We’d rather publish a Week 1 that says “this is where we’re starting” than wait six months and publish something polished that nobody saw us build.

Why “Index” and not “list”

We picked the word deliberately.

A list is a curated set. An index is a measurement.

“Ben’s favorite books of 2026” is a list. The S&P 500 is an index. The first is taste. The second is a measurement system that happens to produce a ranked output.

The Reading Index is meant to be the second thing. The chart you’ll see Friday is a measurement (with editorial framing layered on top), not editorial with data backing. That distinction matters because of what comes later.

The Index is the front of a data product. The per-book weekly signal score sitting underneath the chart is something publishers, AI assistants, and adjacent products can build on. The chart is the part you see. The signal is the part we’re really making.

What to expect on Fridays

The first issue drops this Friday morning ET. Then every Friday after that, the Index publishes:

  • Reading Index, Fiction (Top 25)
  • Reading Index, Nonfiction (Top 25)
  • Movement (up, down, hold), weeks on chart, peak rank
  • Debuts called out
  • A short editorial pull from us: the biggest mover, the surprise debut, the streak that won’t quit

Over time the chart will shift as the methodology refines. We’ll write up what changed and why each time. If a book that should obviously be in cultural circulation isn’t on the Index, tell us. The whole thing only gets better with more signal.

One last thing

We didn’t build the Reading Index to compete with the New York Times. We built it because the thing we actually wanted to read didn’t exist.

If that resonates, join us on Siftivo. The Index is free, weekly, and always will be.

First issue drops Friday morning ET. (We’re a little nervous about it, which feels right.)