Goodreads only remembers today. Here's how I map another author's journey.

The question

The very first piece of advice I found when I started looking into writing romance was this: find authors in your tropes and niche, a few years ahead of you, and learn from their arcs.

Great. So I opened Goodreads. Every book page shows a rating count, a review count and a want-to-read number. For today. There is no history tab, so there's no arc. A book with 40,000 ratings could have earned them in one wild month or across eight slow years, and the page won't say which.

What I tried first

Reading the reviews by date. Reviews carry dates, so in theory I could sample them and see when people were reading. On a small book, maybe. On a book with hundreds of thousands of ratings, it's hopeless by hand, and reviews are only a slice of ratings anyway.

Looking under the hood. I opened the page source in my browser, hoping the book's history would be sitting in the raw page. It wasn't. The current rating count is in there, in the page's structured data, but it's still only today's number.

What worked: the Internet Archive

I love the Internet Archive. Its Wayback Machine saves copies of public web pages, Goodreads included, each one frozen on the day it was saved and given its own permanent address. Two saved copies of the same page, a year apart, give me two real points on a book's arc.

One big catch: the archive doesn't save every page every day. Its own automated crawlers move around the web on their own schedule, and the archive says plainly that some sites are missing because its crawlers didn't know they existed. Anyone can also save a single page by hand with the archive's "Save Page Now" button. A popular book's Goodreads page might be saved hundreds of times. A quiet one might be saved twice. And a site can block the archive, so some saves come back as a blocked page. So my records are always spotty, and I write that down rather than pretend otherwise.

How I find the snapshots

1. Start from the author's own link. Many books have more than one Goodreads page: different editions, translations, a re-release under a new number. I use the Goodreads link the author gives on her own website or her Goodreads profile. That main page pools ratings from all the editions, so it carries the whole count. Two edition pages of Heated Rivalry, with different covers and different addresses, show the exact same number. The archive, though, files every version of an address separately, with or without the title on the end, so I check those too.
2. See what the archive has. I go to the Wayback Machine at web.archive.org and paste the Goodreads address into its box. (I don't use the search on the main archive.org page. It searches the archive's books and media, and it threw an error for me.) It shows a calendar with a dot on every day someone saved that page. Clicking a dot opens the page exactly as it looked that day, ratings count and all.
3. I write down dates before opening anything, because the next step decides which ones are worth opening.

How I build the record: halve it until the growth shows up

Opening every snapshot is slow, and most of them won't tell me anything new. So I use a halving method. It's a simple way to find where the growth is hiding with as few snapshots as possible.

1. I anchor both ends and the big releases. The earliest capture, today's live number, and a capture just before each of the author's most popular books came out. Big releases are where arcs usually bend, so I don't wait for halving to stumble onto them.
2. I weigh each window. Between each pair of anchors I can see what share of the total growth landed there. That share is the weight. Windows are rarely the same length, though, so I also check growth per day. A long window can win on share just by being long.
3. I halve the heaviest window. I open the capture closest to its midpoint, and weigh the two halves. Then I repeat.
4. I stop when the window is small, around three months, or when there's no capture close enough to split it. If the nearest capture is more than a month or so off the date I wanted, I write down how far off it is.
5. Then I go find out why. Once I've narrowed the growth to one window, I look at what the author did then: a sequel, a sale, a newsletter, a moment on TikTok.

Here's a real one. I picked Abby Jimenez. Her most-rated book on Goodreads is Just for the Summer, with 1,244,366 ratings on October 16, 2026, so that's the book I tracked, with her first, The Friend Zone, alongside it.

I started with two captures: two days after it came out, and March 24, 2026. Just for the Summer went from 5,374 ratings on April 4, 2024 to 1,111,040 on March 24, 2026. Two years is too long to match to anything, so I halved it. A capture from March 30, 2025 read 722,958: about 1,993 ratings a day in the first year, and about 1,081 a day in the second. That first year begins two days after it was the Good Morning America Book Club pick for April.

Then I checked her first book in that same year. The Friend Zone gained about 67 ratings a day before it, from July 2019 to December 2021, and about 462 a day during it: almost seven times faster.

So her first book woke up when her biggest one took off. That fits new readers going back for more. It doesn't prove it. Ratings aren't sales, and Goodreads itself and readers' habits keep changing.

I used archived captures, more roughly, for the fan works in CS-001: four readings of one AO3 tag showed slow growth for years and then a flood after a show premiered.

What I write down for every snapshot

The date of the snapshot, the date I wanted, the number, and the snapshot's address. Blanks stay blank, never zero: a page that loaded without a number is missing data, not a zero. And I keep the address, so anyone who asks can open the same page and check me.

What this can't tell me

Ratings aren't sales, and nothing in the archive shows Kindle Unlimited reads, ad spend or an author's mailing list. The halving method works best when growth is lumpy. When every split comes out close to fifty-fifty, that suggests no one moment did it. And it can't prove cause. A release that lines up with growth is worth a look, nothing more. I also haven't spoken with Abby Jimenez, and she may remember these years differently than the page counts suggest.

What I'm taking back to my desk

I started this because the advice said to study authors ahead of me. Mostly it taught me to ask when before I ask how, and now I know where to go looking.

Corrections to this piece are logged on the Corrections page. Want the next one? Get updates.