Writing a strong abstract
The abstract is very likely the most-read part of your entire document, and often the only part a lot of people will ever read. A supervisor skimming submissions, a researcher deciding whether to open your full paper, a marker forming their first impression — all of them start here. It's a small amount of text carrying a disproportionate amount of weight.
What a good abstract contains
Most strong abstracts, across fields, cover roughly the same ground in a tight paragraph or two:
- Context/motivation — one or two sentences on the problem area and why it matters.
- The specific gap or question — what exactly you're addressing that wasn't already solved.
- What you did — your approach or method, briefly, without full methodological detail.
- What you found — your key result(s), ideally with a concrete number or specific finding rather than a vague claim like "the results were promising."
- Why it matters — the implication or significance of the finding, in one closing sentence.
This maps closely onto the IMRaD structure compressed into a handful of sentences — which is exactly why writing the abstract last, once the full paper is settled, tends to work best.
Why is this important?
A vague or poorly written abstract can cause a genuinely strong piece of work to be undersold or overlooked entirely — readers decide whether to invest time in your full document based almost entirely on this paragraph. It's also usually the hardest part to write well precisely because it has to say the most in the least space, which is exactly why it deserves dedicated editing time of its own, not just a rushed summary tacked on at the end.
Common mistakes
- Being vague about results. "We evaluate the model's performance" tells the reader nothing; "our model improves accuracy by 8% over the baseline" tells them something concrete and citable.
- Overloading with background. An abstract isn't the place for a mini literature review — a sentence or two of context is enough before moving to your specific contribution.
- Using citations or undefined jargon. Abstracts are usually meant to be readable in isolation, without needing the rest of the paper (or a reference list) to make sense of them.
- Writing it first and never revising it. An abstract drafted before the paper is finished almost always drifts out of sync with what the paper actually ended up saying. Always revisit it as a final step.
Tips
- Write the abstract last, but budget real time for it — treat it as its own piece of writing, not an afterthought squeezed in before a deadline.
- Check your target word/character limit early (journals, conferences, and university templates often enforce a hard limit) and write within it from the start, rather than writing freely and cutting drastically at the end.
- Read it completely on its own, pretending you know nothing about the rest of the paper — does it stand alone and make sense? If not, revise until it does.
- Compare it against a few strong abstracts from papers in your field to calibrate tone and length — abstract conventions vary noticeably between disciplines.
References
George Mason University Writing Center — Writing an IMRaD Report (covers the abstract's role alongside the rest of the structure)