Reading work documents efficiently is mostly a matter of selection. Decide what you need from a document, look at its structure, read the parts that matter at full attention and skim or skip the rest. Reading speed itself changes little; what changes is how many words you spend it on.

Start with the purpose

Before opening a document, answer one question: what will I do once I have read this? The usual answers at work are:

  • Decide — approve, reject, choose between options.
  • Act — carry out something the document asks for.
  • Respond — give feedback or an answer.
  • Stay informed — know roughly what is happening.
  • Check — confirm one fact, date or figure.

Each calls for a different kind of reading. Checking a figure means scanning. Staying informed means the summary is probably enough. Approving a budget means reading the numbers and assumptions closely and the background lightly. Without a stated purpose, the default is to read everything from the top at one pace, which is the slowest way to get what you need.

Preview before you read

Give a long document one or two minutes of structured looking before you commit to it:

  1. Title, author, date and who it was sent to.
  2. The executive summary or abstract, in full.
  3. The headings, to see how it is organised.
  4. The conclusion or recommendations, in full.
  5. Any tables or charts that carry the main results.

Often this is enough to act on. When it is not, you now know which sections hold what you need.

Previewing is close to what readers do naturally when time is short. Nielsen (1997), in usability research on web pages, reported that most users scan a page rather than read it word by word, picking out headings and highlighted text. Work documents invite the same behaviour. The difference between scanning by habit and previewing on purpose is that the second follows an order and ends in a decision about what to read next.

Triage: full read, skim or skip

After the preview, sort the document’s sections into three groups.

Treatment Use it for Examples
Read in full Anything you will be accountable for or must act on Recommendations, figures you are approving, commitments, deadlines, risks, anything with your name on it
Skim Material that gives context you partly know Background, methodology in a familiar area, status updates from other teams
Skip Material you can find again if needed Appendices, reference tables, boilerplate, long reply chains below the latest message

What skimming buys you, and what it costs, is reasonably well understood. Duggan & Payne (2009) studied readers working through text under time pressure, and their findings suggest that skimmers tend to pick up the more important ideas from the parts they sample, at the expense of detail. Rayner et al. (2016) come to a compatible conclusion in their review: skimming is a useful skill when the gist is enough, and it loses detail.

That gives a simple test for the triage. If being wrong about a detail in this section would cause a problem, it belongs in the first group. Contract terms, financial figures, safety instructions and anything you are signing should never be skimmed.

Read the key sections at full attention

For the sections you have chosen to read in full, do the opposite of rushing. Close email and chat, read at the pace the material needs, and go back over anything that did not make sense the first time. Rayner et al. (2016) are clear that speed and comprehension trade off, so pushing the pace on the parts that matter defeats the point of having selected them.

The time you saved by skimming the background is what pays for this. Efficient reading is uneven on purpose: quick over most of the document, slow over the part you will act on.

Budget time for a long report

Long documents get postponed because their size is vague. Putting a number on it helps. Paste the text into the reading time calculator, or enter the word count, and choose a pace.

Take a 6,000-word report.

  • At 238 WPM, the average for adult silent reading of English non-fiction (Brysbaert, 2019), it takes 25 min 13 sec.
  • At 180 WPM, the calculator’s careful preset for dense or unfamiliar material, it takes 33 min 20 sec.

So a full read needs a half-hour block, a little more if the content is demanding. That is a different planning problem from “a long report I’ll get to later”.

Now apply the triage. Suppose the preview shows that the sections you must read closely are the summary (400 words), the findings (1,500 words) and the recommendations (700 words). That is 2,600 words. At 180 WPM, careful reading of those sections takes 14 min 27 sec. Add a few minutes to skim the remaining 3,400 words of background and method, and a block of about 20 minutes covers the report at the depth each part needs.

Three cautions about these numbers:

  • They are estimates of reading time only. Thinking about the content, checking a figure against a spreadsheet and writing a response all take additional time.
  • Tables, charts and financial statements are not read at a words-per-minute rate. Allow extra for them.
  • 238 WPM is an average across many readers, not your personal speed. If you know your own pace, use that instead.

For quick reference figures, see reading time by word count. If your estimates keep coming out wrong, the guide to common reading time calculator mistakes lists the usual causes, and why reading speed changes between text types explains why a dense report takes longer than its word count suggests.

Handle email in batches

Email is a different reading problem from reports. The individual messages are short. The cost comes from the number of them and from the interruptions.

Reading time itself is rarely the issue. Forty messages averaging 120 words each come to 4,800 words, which is 20 min 10 sec of reading at 238 WPM. What takes the rest of the time is switching in and out of other work, re-reading messages you opened but did not deal with, and deciding what to do with each one.

A practical approach, offered as working practice rather than a research finding:

  • Read email at set times rather than as each message arrives. Two or three sessions a day suits many roles; adjust to what your job requires.
  • Triage from the subject line and first two lines. Sort each message into act now, act later or no action.
  • Read once and decide. If a reply takes a minute or two, send it. If it needs more, schedule it and move on.
  • Read long threads from the latest message. Go back through earlier messages only if the latest one does not make sense without them.
  • Check whether you are a recipient or copied. Messages you are copied on usually need the first paragraph only.

The same preview-then-triage logic applies to an inbox as to a report: look at the whole thing first, then spend attention where the consequences are.

Write down the decision

Reading is not finished until its result is recorded somewhere you will find it. Immediately after a document, note in two or three lines:

  • the main point, in your own words;
  • what you decided or what you need to do, with a date;
  • any open question and who can answer it.

This takes about a minute, and it prevents the most expensive kind of inefficiency in work reading, which is reading the same document a second time because you cannot remember what you concluded. It is also a check on comprehension. If you cannot state the main point, the skim was too thin and one section needs a proper read.

A short checklist

  1. State the purpose: decide, act, respond, stay informed or check.
  2. Preview: summary, headings, conclusion, key tables.
  3. Estimate the time with the calculator and block it out.
  4. Triage sections into read, skim and skip.
  5. Read the key sections without interruption.
  6. Record the main point, the decision and the next step.

None of this makes your eyes move faster, and it does not need to. If you also write documents for other people, the same reasoning runs in reverse: a clear summary, informative headings and a stated reading time make them easier to triage. Our guide on estimating article reading time covers that side.