Skimming vs Scanning: Differences and When to Use Each

Skimming and scanning are both ways of not reading every word, and they answer different questions. Skimming asks "what is this text about?" Scanning asks "where is the thing I am looking for?" You skim when you do not yet know what a text contains, and you scan when you know exactly what you want from it.

Mixing them up wastes time in both directions: scanning a report you have never seen for its "main point" finds nothing, and skimming a timetable for one departure reads far more than you need.

What is skimming?

Skimming is reading selectively to get the gist. You sample the places where writers put their most important information, such as the title, headings, opening paragraph, first sentences and conclusion, and pass quickly over the rest. You finish knowing the topic, the main claims and the structure, but not the details, the evidence or the exact wording.

What is scanning?

Scanning is searching a text for one specific item: a name, a date, a figure, a keyword. You hold the target in mind and sweep your eyes over the page without trying to understand what you pass. Most words are rejected on shape alone. When something matches, you stop and read the sentence around it properly. You finish with one answer and little idea of what the rest of the text said.

Skimming vs scanning: the differences at a glance

Skim to decide what a text says and whether it deserves a full read. Scan to pull out one known item. Neither replaces careful reading when you need to understand or remember the details.

When should you skim, and when should you scan?

Skim when you are choosing which of several sources to read, previewing a chapter before studying it, catching up on a long thread or report where the outline is enough, or refreshing material you have already read carefully.

Scan when you have a concrete question: the deadline in an email, a definition in a textbook, a clause in a contract, a statistic you remember seeing. If you can phrase what you want as a word or number you would recognise on sight, scan.

Do neither when the text is the thing you are responsible for: instructions you must follow, an argument you have to evaluate, a contract you are about to sign, or anything you are reading for the pleasure of the writing. Read those at your normal pace.

The two also work well in sequence. Skim a document to learn its layout, scan to find the section that answers your question, then read that section closely.

How to skim a text, step by step

1. Set a purpose and a time limit. Decide what you want to know and give yourself roughly a quarter of the time a full read would take.
2. Read the title, the introduction and the conclusion in full. Writers state their point at the edges of a text.
3. Read every heading and subheading. Together they are the outline.
4. Read the first sentence of each paragraph. If it tells you the paragraph is an example or a detail, move on; if it introduces a new idea, read a little further.
5. Let your eye catch signals:numbers, names, bold or italic terms, lists, tables, figure captions, and words such as "however", "because" and "therefore" that mark a turn in the argument.
6. Move on when a section stops paying. Once a passage is no longer giving you anything new, jump to the next one.
7. Summarise in two or three sentences. If you cannot, you skimmed too fast or the text needs a proper read.

How to scan a text, step by step

1. Fix the target. Know the exact word, name or kind of number you want, and think of two or three alternatives the author might have used instead.
2. Predict where it will be. Use the table of contents, headings or index to go to the likely section before you start sweeping.
3. Use the shape of the target. Capital letters, digits, symbols and unusual word lengths stand out from running text.
4. Sweep in a steady pattern, for example straight down the middle of a narrow column or in a zigzag across a wide page, and resist the urge to read.
5. Stop and read around each hit to check that it is the answer and not just the same word in another context.
6. On a screen, search first. The find command does the scanning for you; your job becomes choosing good search terms and judging the hits.

Is skimming the same as speed reading?

In practice, largely yes, and that is not a criticism of skimming. The major research review of speed reading, by Rayner and colleagues (2016), concludes that reading every word at two or three times your normal rate with full understanding is not realistic, because speed and comprehension trade off against each other. What the review does support is effective skimming: giving priority to the more informative parts of a text and glossing over the rest when the gist is all you need. The same review observes that the people who appear to speed read successfully are usually skilled skimmers who already know a good deal about the subject.

Experiments on skimming point the same way. Duggan and Payne (2009) gave readers only enough time to read about half of each document. Those who skimmed the whole text remembered its important ideas better, but they did no better on minor details or on questions that required drawing inferences. Their readers seemed to work by staying with a passage while it was informative and leaving once the return dropped, which is a fair description of step 6 above.

So skimming is a real skill with a real cost. You gain coverage and lose depth, and the trade is best when you know the field well enough to recognise what matters. More on the evidence is in does speed reading work?, and the full set of habits worth changing is in how to read faster.

How to practise skimming and scanning

Scanning drills: two minutes

Scanning is the easier of the two to drill because it has a clear right answer. Tachi has two short exercises for it.

In Text Scan, a passage appears with a target word of at least four letters taken from it. You tap every occurrence before a countdown of 20, 30, 45 or 60 seconds ends, and the result shows how many you found and how many wrong taps you made. Start with 30 seconds and aim for zero wrong taps before you try the 20-second setting.

In Word Search, a target word is shown above a grid of words and you tap the match. A session is ten rounds on a 3×3, 4×4 or 5×5 grid, and it reports your correct answers, mistakes and total time. Start on 4×4; move to 5×5 when you can finish ten rounds without a mistake.

Both drills train you to hold a word shape in mind and reject everything else. Getting quicker at them shows you are better at the drill; whether that carries over to scanning real documents has not been established, so use them as a short, focused warm-up and do the rest of your practice on real text.

Skimming practice: ten minutes on real articles

1. Pick three articles of roughly 1,500 to 2,000 words on subjects you know something about. At an average non-fiction pace of about 238 words per minute, each would take six to eight minutes to read in full.
2. Give yourself two minutes per article and follow the skimming steps above.
3. Close the article and write a three-sentence summary: the topic, the main claim and the most important support for it.
4. Reopen the article and check the summary against the introduction and conclusion. Note what you missed and where in the text it was.

Repeat a few times a week. The aim is not a faster sweep but a better summary in the same two minutes. One caution: a word-at-a-time RSVP reader is the opposite of skimming, because it shows you every word in order and you cannot jump ahead. Skim the original document first, then decide whether it deserves a paced, complete read. Short definitions of both terms are in the glossary, and typical full-reading rates are covered in average reading speed.