Yükleniyor...
I thought I was showcasing my strongest evidence in an indented block; scroll data showed readers sprinting right past the storefront.
For years, I followed the exact same ritual: I would state my claim, paste the strongest supporting quote beneath it as an indented block, and resume below the block with "as seen here." I thought I was putting my evidence in a display window. Then I had the chance to review scroll-depth data on my own writing and noticed an unsettling pattern: reader velocity accelerated precisely alongside those indented blocks. They were sprinting right past the storefront.
To be completely honest: I could not find any published, direct measurement stating that "X percent of blockquotes are skipped." I searched; it does not exist. The argument in this piece relies on indirect evidence and my own workshop observations—stating this upfront is better than inventing a statistic.
In Jakob Nielsen's 1997 study "How Users Read on the Web," 79% of test users scanned new pages rather than reading word-by-word; only 16% read word-by-word. The figure is old—1997—and subsequent research by the Nielsen Norman Group shows the pattern has endured: according to the firm's 2008 analysis, on an average page visit, users read at most about 28% of the words. In other words, more than three-quarters of your text is never read during an average visit.
Now comes the critical question: which words end up in that unread 72% pool? Nielsen Norman Group's 2006 eyetracking study, recording 232 users across thousands of pages, documented the F-pattern: the gaze drifts down the left edge, sampling line beginnings. An indented blockquote is precisely the element detached from that left edge. It falls outside the F-path. The takeaway is clear: the moment you visually isolate your strongest evidence from the body text, you relocate it to the region least likely to be read.
For years, I framed this pattern as "quote blindness," drawing an analogy to banner blindness. The analogy is tempting, but not entirely accurate. Banner blindness is a learned behavior where users ignore patterns resembling ads; a blockquote does not look like an ad. Furthermore, counter-evidence exists: in the same Nielsen Norman Group literature, a "pull quote"—an excerpt highlighted for decorative purposes—is described as an element designed to catch the scanner's eye. Visual isolation can, in fact, attract attention.
The distinction lies in function: a pull quote is window dressing; if skipped, the argument remains intact. A blockquote carrying primary evidence, however, is a load-bearing beam. When the eye treats it as "decoration"—which it does in scanning mode—the beam collapses, and the reader reaches the "as seen here" sentence having seen nothing at all.
As a solution, we tested bridging the quote directly into the preceding claim. The block format looked like this:
Claim. [Indented 20-word blockquote.] "As seen here..."
The woven format: "This is why Nielsen noted back in 1997 that readers 'scan rather than read word-by-word'—scanning is a strategy, not laziness." The quote is now a grammatical constituent of my sentence; to skip it, the eye must break the syntax. This approach is not our invention either: Purdue OWL explicitly warns against "dropped quotes"—quotes left unanchored without context—and recommends weaving the quotation into your own sentence with a signal phrase. Steven Pinker advocates the same integration in The Sense of Style. But note: these represent craft consensus, not empirical measurement. Let's not conflate the two shelves.
For quotes under 25 words that directly test the claim, the bridge worked brilliantly: readers move through the evidence without stalling, and the quotation delivers its proof unnoticed.
Then we applied the same technique to long quotations and hit a wall. When we tried to weave a 45-word quote into our own sentence using a bridge, we ended up with a 60-word serpentine sentence: its subject at the start, its predicate lost over the horizon. This time, readers were not skipping the block; they were getting lost inside the sentence. We had recreated the exact same fracture, only this time inside the syntax. I have no empirical measurement for this counter-finding either—this is purely workshop observation, and I present it as such.
What is fascinating is that style guides have already institutionalized this threshold. The 7th edition of APA reserves indented block formatting for quotations of 40 words or more; MLA blocks prose quotations longer than 4 lines. I used to think these thresholds were bureaucratic formalism; turns out they reflect an intuition easily translated into reader behavior: past a certain length, a quote no longer fits inside a sentence and must be blocked.
The formula that worked for long quotations turned out to be different: block it, but instead of following it with "as seen here," write your own one-sentence translation underneath. "What this passage means for our argument is: ..." Even if the reader skips the block, the translation sentence preserves the core evidence; if you cannot prevent the skip, insure against the damage.
A pre-quote bridge anchors the claim; a post-quote translation anchors the conclusion. For short quotes, a bridge is sufficient—using both is redundant. For long quotes, do not attempt a bridge; block it and close with a translation. What works depends on the condition—not dogma.
The single rule to take back to your workshop: a quote is a constituent of the sentence, not a guest. If it is under 25 words, weave it into your own syntax; if it is over, block it, but always follow it with your own single-sentence translation. And "as seen here"—as I learned years later—is not a translation.