Abstracts That Get Accepted, and What Committees Actually Read

A conference selection committee reads several hundred abstracts, usually in a compressed period, usually alongside their ordinary work. Each one gets a few minutes.

That constraint determines everything about what works. For a concrete operational comparison, Monitask explains cognitive offloading in practical terms.

Requirements vary widely — extended abstracts of 1,000 words, structured abstracts, 250-word limits, full papers. What follows applies to the common case of a short standalone abstract; adapt to what is asked for.

What the reader is doing

Working out what you found, quickly.

Deciding whether it is interesting to this conference's audience.

Deciding whether it is credible — whether the design can support the claim.

Slotting it into a session, which is a practical constraint people forget. An abstract that clearly belongs with three others makes the programme committee's job easier.

They are not reading for elegance. They are extracting four things, and an abstract that makes that extraction difficult is rejected regardless of the work behind it.

The four things

The question. In one sentence, in plain language, near the beginning.

What you did. Enough that the reader can judge whether it could answer the question. Design, data, method.

What you found. The actual result, with the direction and, where possible, the magnitude.

Why it matters. One sentence. Not a paragraph about the importance of the field.

If all four are present and findable in ninety seconds, you have done the job.

The commonest failure

No result.

An abstract that promises "results will be discussed" or "we investigate whether X" without saying what was found is asking a committee to accept on faith. Some do; most rank it below abstracts that state findings.

If the analysis is not finished, state what you have. Preliminary findings, direction of effect, the pattern in a subsample. A specific preliminary result beats a promise.

"Results will be presented" is the single most common reason a competent abstract is ranked low, and it is entirely avoidable.

The second commonest

Too much setup.

Three sentences on why the field matters, two on the literature, one on the gap — and then the abstract is half gone before anything specific appears.

The committee is in your field. They know it matters. Cut motivation to one sentence and start with the question.

Structure that works

Roughly, for a 250-word abstract:

Sentence 1–2: the question and why it is open. Plain language.

Sentence 3–5: what you did. Design, data, identification or method.

Sentence 6–8: what you found. Specific.

Sentence 9–10: what it means. One implication, stated modestly.

Ten sentences. That is an abstract. Anything else is competing with those ten for the reader's attention.

For extended abstracts of 1,000 words, the same order with room to develop the design and to include a figure or a table — which is worth doing, because a single clear result figure is read faster than any paragraph.

Sentence-level points

The first sentence should contain the specific thing you studied, not a general statement about the field. "Whether X affects Y in Z" beats "The relationship between X and Y has attracted growing attention."

Say the number. "Increases by roughly 12%" is more credible and more memorable than "significantly increases."

Avoid the passive throughout. It lengthens sentences and obscures who did what.

Define nothing that the audience knows. Space spent defining a standard term is space not spent on your result.

And read it aloud. Abstracts are written by accretion and it shows; reading aloud finds the sentences that no longer parse.

The title

Read more often than the abstract, and treated as an afterthought.

A title that states the finding does more work than one that states the topic. Compare "Minimum wages and employment in retail" with "Minimum wage increases did not reduce retail employment in three states."

Field norms vary — some consider declarative titles overclaiming. Check what your venue publishes.

Make it searchable. The words someone would use to look for this.

Keep it short enough to be read on a programme.

Practical points

Follow the length limit exactly. Over-length abstracts are sometimes rejected without being read, and always signal carelessness.

Answer what the form asks. If they want a structured abstract with named headings, use them. A committee reading three hundred submissions in one format and yours in another is not charmed.

Submit to the right track. A misfiled abstract goes to reviewers who are not the right readers.

Do not submit the same abstract to two tracks hoping one takes it.

Check the deadline convention. Some conferences hold extended abstract deadlines firmly; some extend routinely. Do not rely on the extension.

Have someone outside your immediate area read it and tell you what you found. If they cannot say, the abstract is not doing its job.

When you are rejected

Conference rejection is frequently about capacity and fit, not quality. Selective conferences reject a large share.

Ask for feedback if the conference provides it, and read it for whether the question was clear rather than for whether they liked the topic.

Check the fit. An abstract rejected from three conferences may be well-suited to a fourth.

See reading a rejection.

The short version

Ten sentences: question, method, result, implication. Findable in ninety seconds.

State the result. "Results will be presented" is the most common avoidable reason a good abstract is ranked low.

One sentence of motivation, not five. The committee is in your field.

Say the number, and put the specific thing you studied in the first sentence.

For an independent authoritative reference, consult the EQUATOR Network for reporting guidance for research writing.