Skip to content
Academic Writing

How to Write a Conclusion for an Essay or Research Paper

A conclusion is the last section of the paper, not a second abstract and not a pasted thesis statement. The UNC Writing Center says it should synthesize your thoughts, show why the ideas matter, and move the reader to a new view of the subject. This page is that writing how-to. It is not how to write an introduction, which opens the paper, and not how to write a research paper, which covers the whole draft.

By 26 min read

A vintage typewriter under a desk lamp with a page in the carriage
Photo: Alexa Mazzarello, Wikimedia Commons (CC0)
On this page

Quick Answer: How to Write a Conclusion

Harvard College Writing Center asks the same three questions in every discipline: what the paper established, so what (why those stakes still matter), and now what (a question, application, or action the body actually earned). University of Toronto Writing Advice adds the length rule: one well-developed paragraph for most essays; two or three only when the paper is long enough to need them. Do not invent a new claim, a new source, or a new statistic in the last paragraph.

Starting templates (fill from the paper you wrote; do not invent evidence)

Course essay / humanities paper:

Read this way, [the main claim restated after the evidence], not merely that [the topic]. That changes how we see [the text, event, or debate] because [the so-what the body earned]. The next question is [a question that would not have made sense before the analysis].

Empirical / IMRaD paper (Discussion, not a second Results list):

These results show that [finding in one sentence], which is consistent with [named prior work] and inconsistent with [the alternative the design could actually test]. The main limitation is [the one that changes interpretation]. A useful next study would [a design that closes that gap].

Policy or applied paper:

Because [the evidence already in the paper], [who] should [a specific action the findings support], rather than [the usual alternative]. Leaving the current practice unchanged would [a consequence the paper can actually defend].

Use the thesis-statement generator only to tighten the claim you are circling back to. Then cite the sources that actually support that claim with the CitationEasy generator.

What a Conclusion Is For — Official Definitions

Four writing centers describe the same job. UNC names the rhetorical work and calls the last paragraph a gift. Toronto names closure plus a wider frame. Hamilton College's Nesbitt-Johnston Writing Center names the justification: why the argument is worth having made. Purdue Global Academic Success Center (17 July 2026) names the last impression: a restated claim, the support the body already earned, and one final thought. None of them treat the last paragraph as a recap of headings.

UNC Writing Center, “Conclusions”

Your conclusion is your chance to have the last word on the subject. The conclusion allows you to have the final say on the issues you have raised in your paper, to synthesize your thoughts, to demonstrate the importance of your ideas, and to propel your reader to a new view of the subject.
Just as your introduction acts as a bridge that transports your readers from their own lives into the “place” of your analysis, your conclusion can provide a bridge to help your readers make the transition back to their daily lives.
Synthesize, don’t summarize. Include a brief summary of the paper’s main points, but don’t simply repeat things that were in your paper. Instead, show your reader how the points you made and the support and examples you used fit together. Pull it all together.
Your conclusion can go beyond the confines of the assignment. The conclusion pushes beyond the boundaries of the prompt and allows you to consider broader issues, make new connections, and elaborate on the significance of your findings.
Your conclusion should make your readers glad they read your paper. Your conclusion gives your reader something to take away that will help them see things differently or appreciate your topic in personally relevant ways. It can suggest broader implications that will not only interest your reader, but also enrich your reader’s life in some way. It is your gift to the reader.

University of Toronto Writing Advice, “Introductions and Conclusions” (Leora Freedman and Jerry Plotnick)

A strong conclusion will provide a sense of closure to the essay while again placing your concepts in a somewhat wider context. It will also, in some instances, add a stimulus to further thought.
A conclusion is not merely a summary of your points or a re-statement of your thesis. If you wish to summarize—and often you must—do so in fresh language. Remind the reader of how the evidence you’ve presented has contributed to your thesis.
A good last sentence leaves your reader with something to think about, a concept in some way illuminated by what you’ve written in the paper.

Hamilton College Writing Center, “Conclusions”

You might begin a conclusion by reminding the reader of your thesis and main points, but a conclusion is not a summary. A successful conclusion brings your argument to a logical close; it is where you justify your argument to the reader.
Use your conclusion to describe the consequences of your argument, to answer the question “why is my argument important?”
Do not begin with the unnecessary and over-worked phrase “in conclusion.”
Make a reasonable final claim; don't overstate your findings.
Take special care with your final sentence; make sure it is carefully constructed and leaves a strong impression.

Essay Conclusion vs Research-Article Discussion

The heading on the last section is a genre choice, not a synonym. A course essay usually ends with a section you can call the conclusion. An empirical article in APA or Vancouver style usually ends with a Discussion that interprets results. Do not paste an essay closer onto a lab report, and do not turn a close-reading paper into an IMRaD discussion with a fake "limitations" paragraph you did not earn.

Course essay / humanities paper

  • Returns to the thesis in new words after the evidence
  • Synthesizes the readings or cases; does not list body headings
  • May close the circle on an image or question from the introduction
  • May end on a sharper question or a rereading of a key quotation
  • Does not dump unused archival finds (UNC's "Grab Bag")

Empirical / medical / APA paper

  • Interprets results against the hypothesis or purpose
  • Places findings in the published evidence, not only in your sample
  • States the limitations that change how the result can be used
  • Names a next study or a practice implication the data support
  • Does not repeat the Results section or invent economic claims

APA Style's Discussion Phrases Guide (7th ed.) treats the Discussion as the paper's concluding section. Use present tense there. Quantitative and qualitative reporting detail live in the Publication Manual (7th ed.) §§3.8 and 3.16. Student papers still follow the instructor: some courses want a heading titled Conclusion; many psychology reports want Discussion and no separate closer.

APA Style, Discussion Phrases Guide (7th ed.)

Papers usually end with a concluding section, often called the “Discussion.” The Discussion is your opportunity to evaluate and interpret the results of your study or paper, draw inferences and conclusions from it, and communicate its contributions to science and/or society. Use the present tense when writing the Discussion section.
In the Discussion section of a research paper, you should evaluate and interpret the implications of study results with respect to your original hypotheses. It is also where you can discuss your study’s importance, present its strengths and limitations, and propose new directions for future research.

Official phrase for linking findings (fill the gap; do not invent a study):

This pattern of results is consistent with the previous literature … (citation).

Official phrase for a limitation:

One limitation of this study is that … It is possible that ... Although the present research cannot rule out these explanations, it seems useful to point out issues that may conflict with these results.

Medical journals follow the ICMJE Recommendations (updated January 2026). The Discussion is not a second Results dump, and the closing claims have to match the data. The January 2026 text also asks authors to discuss sex and/or gender where those variables matter.

ICMJE Recommendations, “Discussion” (January 2026)

It is useful to begin the discussion by briefly summarizing the main findings, and explore possible mechanisms or explanations for these findings. Emphasize the new and important aspects of your study and put your findings in the context of the totality of the relevant evidence. State the limitations of your study, and explore the implications of your findings for future research and for clinical practice or policy.
Discuss the influence or association of variables, such as sex and/or gender, on your findings, where appropriate, and the limitations of the data. Do not repeat in detail data or other information given in other parts of the manuscript, such as in the Introduction or the Results section.
Link the conclusions with the goals of the study but avoid unqualified statements and conclusions not adequately supported by the data. In particular, distinguish between clinical and statistical significance, and avoid making statements on economic benefits and costs unless the manuscript includes the appropriate economic data and analyses. Avoid claiming priority or alluding to work that has not been completed. State new hypotheses when warranted, but label them clearly.

Harvard’s Three Moves: What, So What, Now What

Harvard's page is the shortest working method. The first move is a transition, not an inventory. The second move returns to the stakes you named in the introduction, then widens them. The third move is the sentence a reader can carry out of the room. Harvard is explicit that STEM papers often end on further-study questions, while a literature paper may reread a line from the central text in light of the argument.

Harvard College Writing Center, “Conclusions”

How you conclude your paper will depend on where you started—and where you traveled.
In a short paper—even a research paper—you don’t need to provide an exhaustive summary as part of your conclusion.
After the “what” and the “so what,” you should leave your reader with some final thoughts.
What can my readers now understand, see in a new light, or grapple with that they would not have understood in the same way before reading my paper?
What questions can I now raise that would not have made sense at the beginning of my paper? Questions for further research? Other ways that this topic could be approached?
fake transitions like “in conclusion” that are followed by sentences that aren’t actually conclusions. (“In conclusion, I have now demonstrated that my thesis is correct.”)

Harvard's worked example is Vanessa Roser's paper on ADHD and rejection sensitivity (RS). Those sentences are hers, as the Writing Center published them in the Strategies for Essay Writing packet (March 2024). They are not a template to copy into an unrelated topic.

Harvard’s Vanessa Roser example (ADHD and rejection sensitivity)

Thesis (what the paper set out to show):

While socialization may indeed be an important factor in RS, I argue that individuals with ADHD may also possess a neurological predisposition to RS that is exacerbated by the differing executive and emotional regulation characteristic of ADHD.

Conclusion echo (what, restated after the literature):

This literature demonstrates that, as with many other conditions, ADHD and RS share a delicately intertwined pattern of neurological similarities that is rooted in the innate biology of an individual’s mind, a connection that cannot be explained in full by the behavioral mediation hypothesis.

Now-what (fault and treatment):

seems to place a degree of fault on the individual. It implies that individuals with ADHD must have elicited such frequent or intense rejection by virtue of their inadequate social skills, erasing the possibility that they may simply possess a natural sensitivity to emotion.

Now-what (who should care):

has profound implications for understanding how individuals with ADHD might best be treated in educational settings, by counselors, family, peers, or even society as a whole.

Notice what Roser does not do. She does not list every study again. She does not announce "in conclusion." She does not apologize for the papers she did not read. She restates the claim in light of the literature, then says what a wrong theory costs real people. That is the so-what and the now-what in two steps.

UNC Strategies That Actually Move the Last Paragraph

UNC's handout is the one to keep open while you revise. The strategies below are theirs. Use one or two. Stacking all of them produces the grab-bag ending UNC warns against.

Play the “So What” game

UNC's diagnostic is a conversation, not a formula. After each concluding sentence, ask "So what?" and then "Why should anybody care?" Their published example uses Frederick Douglass's Narrative of the Life of Frederick Douglass, an American Slave, Written by Himself — the book listed in the handout's works consulted, not an invented case.

UNC’s Douglass “So What” exchange

You:

Basically, I’m just saying that education was important to Douglass.

Friend:

So what?

You:

Well, it was important because it was a key to him feeling like a free and equal citizen.

Friend:

Why should anybody care?

You:

That’s important because plantation owners tried to keep slaves from being educated so that they could maintain control. When Douglass obtained an education, he undermined that control personally.

The first sentence is a topic. The last sentence is a claim about power. If your draft still sounds like the first line, you do not yet have a conclusion. You have a restatement.

Return to the introduction — then change it

UNC and Toronto both recommend closing the circle: return to a scenario, anecdote, or quotation from the opening, but only after the body has earned a new reading. Toronto's wording is the useful constraint: add further insight that comes from the essay, do not replay the hook. A paper that opens on the Greensboro sit-ins, UNC notes, can end by placing that event in the Civil Rights Movement as a whole. A paper on Virginia Woolf's style can end on later writers or later feminists. Those are UNC's examples, not prompts to invent sit-in statistics.

Propose an action, a solution, or a next question

This is Harvard's now what and UNC's "course of action" in one move. It only works when the body already argued for that action. A literature paper rarely needs a policy recommendation. A lab report rarely needs a meditation. Hamilton adds a matching caution: make a reasonable final claim; do not overstate the findings. Hamilton also lists a move the other handouts leave implicit: raising qualifications — limiting or opposing viewpoints — when the assignment allows it.

UNC Writing Center, strategies to use

Return to the theme or themes in the introduction. This strategy brings the reader full circle. For example, if you begin by describing a scenario, you can end with the same scenario as proof that your essay is helpful in creating a new understanding.
Propose a course of action, a solution to an issue, or questions for further study. This can redirect your reader’s thought process and help them to apply your info and ideas to their own life or to see the broader implications.
Point to broader implications. For example, if your paper examines the Greensboro sit-ins or another event in the Civil Rights Movement, you could point out its impact on the Civil Rights Movement as a whole. A paper about the style of writer Virginia Woolf could point to her influence on other writers or on later feminists.

Toronto’s Seven Moves Beyond Summary

Toronto's list is the one to use when the last paragraph is accurate but still only a recap. These seven moves are Freedman and Plotnick's, not a house style. Pick the one your genre already earned. Toronto also warns that genre changes the ending: an annotated bibliography typically has no conclusion, a book review may end with an overall assessment, and a policy briefing may end with recommendations.

University of Toronto, “How do I write an interesting, effective conclusion?”

If your essay deals with a contemporary problem, warn readers of the possible consequences of not attending to the problem.
Recommend a specific course of action.
Use an apt quotation or expert opinion to lend authority to the conclusion you have reached.
Give a startling statistic, fact, or visual image to drive home the ultimate point of your paper.
If your discipline encourages personal reflection, illustrate your concluding point with a relevant narrative drawn from your own life experiences.
Return to an anecdote, example, or quotation that you introduced in your introduction, but add further insight that derives from the body of your essay.
In a science or social science paper, mention worthwhile avenues for future research on your topic.
A policy briefing usually includes an introduction but may conclude with a series of recommendations.

What Not to Do — UNC’s Avoid List and Four Failed Endings

UNC's avoid list is more precise than a generic "don't be boring." Harvard's avoid list overlaps and adds two failures UNC leaves implicit: a new counterargument you cannot refute, and an apology for the paper you did not write. Keep both lists; they catch different failures.

UNC Writing Center, strategies to avoid

Beginning with an unnecessary, overused phrase such as “in conclusion,” “in summary,” or “in closing.” Although these phrases can work in speeches, they come across as wooden and trite in writing.
Stating the thesis for the very first time in the conclusion.
Introducing a new idea or subtopic in your conclusion.
Ending with a rephrased thesis statement without any substantive changes.
Making sentimental, emotional appeals that are out of character with the rest of an analytical paper.
Including evidence (quotations, statistics, etc.) that should be in the body of the paper.

Harvard College Writing Center, what to avoid

a complete restatement of all that you have said in your paper.
a substantial counterargument that you do not have space to refute; you should introduce counterarguments before your conclusion.
an apology for what you have not said. If you need to explain the scope of your paper, you should do this sooner—but don’t apologize for what you have not discussed in your paper.

UNC then names four ineffective types. The examples are theirs, written about Douglass. They are what not to write. The first is a restatement with a speech transition. The second hides the thesis until the last line. The third swaps analysis for tribute. The fourth smuggles unused research into the closer.

UNC: four kinds of ineffective conclusions (do not imitate)

“That’s My Story and I’m Sticking to It”:

In conclusion, Frederick Douglass was, as we have seen, a pioneer in American education, proving that education was a major force for social change with regard to slavery.

“Sherlock Holmes” (thesis withheld until the end):

So, as the evidence above demonstrates, Douglass saw education as a way to undermine the slaveholders’ power and also an important step toward freedom.

“America the Beautiful” / “I Am Woman” / “We Shall Overcome”:

Because of the efforts of fine Americans like Frederick Douglass, countless others have seen the shining beacon of light that is education. His example was a torch that lit the way for others. Frederick Douglass was truly an American hero.

“Grab Bag”:

In addition to being an educational pioneer, Frederick Douglass provides an interesting case study for masculinity in the American South. He also offers historians an interesting glimpse into slave resistance when he confronts Covey, the overseer. His relationships with female relatives reveal the importance of family in the slave community.

How Long Should a Conclusion Be?

Do not use a percentage you found on a blog. Toronto is the rule you can defend to an instructor:

University of Toronto, length

For most essays, one well-developed paragraph is sufficient for a conclusion. In some cases, a two-or-three paragraph conclusion may be appropriate. As with introductions, the length of the conclusion should reflect the length of the essay.

A five-page close reading that needs two pages of closer is padded. A thesis chapter that ends in five sentences is unfinished. Journal articles vary by venue: some science letters fold interpretation into a short final paragraph; ICMJE original-research articles use a full Discussion. Ask the assignment, the journal, or the supervisor before you copy a classmate's word count.

Working length ranges, not rules. Confirm with the assignment.
GenreTypical closerWhat "done" looks like
Short course essay (about 5–8 pages)One paragraph (Toronto's default)Thesis in new words, synthesis, one so-what sentence
Long essay or chapterTwo or three paragraphs (Toronto)Same jobs, plus a return to the opening or a next question
APA / IMRaD articleDiscussion section (APA 7 §§3.8, 3.16)Interpretation, evidence context, limitations, next study
Medical original researchICMJE Discussion (January 2026)Mechanisms, totality of evidence, sex/gender where relevant, limits, practice or policy

Published Student Conclusions — Hamilton College

Hamilton publishes sample conclusions from real courses, with permission from Sarah Gould '05 and Matthew Marks '05. Keep the wording; do not file these under a fake citation. The Philosophy 110 sample does the so-what work: it does not retell Aquinas's proofs, and it does not announce that God is a big topic. It says what the argument's method costs.

Hamilton College, sample conclusion for Philosophy 110

It is a natural, human inclination to wonder if God exists. Aquinas, like many other philosophers, attempted to prove God’s existence through logical reasoning. His arguments, however, seem to be based on the presumption that there is in fact a God. He went on from this supposition to question human concepts, such as the way that we determine what is good, true, and noble. If no obvious, tangible cause for a concept or behavior could be found, he concluded that God must be the cause and then claimed proof of God’s existence. Aquinas was clearly biased in his study of God; he seemed to have a firm belief that God exists before he began the process of logical reasoning. This fault would be a problem for anyone trying to prove God’s existence through concrete logic. There is something uncertain and enigmatic about the concepts of God and religious belief that logic will never be able to penetrate.

The last sentence is the now-what. Everything before it earns that sentence. If you deleted the Aquinas-specific middle, the closer would collapse into a greeting-card line about mystery — UNC's sentimental failure mode. Hamilton's other two published samples show the same job in other genres.

Hamilton College, sample conclusion for English 150 (Coetzee)

The physical disabilities of the girl serve as symbolic functions that affect how the reader interprets her relationship with the magistrate. These injuries lend themselves to symbolism because they are reflections of the weaknesses in all people—the insecurities, the compassion, and the love for one another that may manifest itself in various forms. Coetzee applies this symbolism to the complex relationship between two very different characters. The effect is a complicated and often contradictory representation of love and the emotions closely related to it.

The English 150 closer does not retell the plot and does not announce that love is important. It says what Coetzee's injuries do to the reading. That is synthesis. The Government 112 sample is useful for a different reason: it cites Samuel Huntington's three paths, then uses them to refuse a bigger claim the paper did not prove. The Hu Jintao setting is the course paper's own moment, published with permission around 2005; quote it as Hamilton's sample, not as current China analysis.

Hamilton College, sample conclusion for Government 112

The coming of capitalism and modernization to China does not indicate that democracy will follow. Strong forces both within and without the CCP ensure that the status quo of one-party politics will remain. According to political theorist Samuel Huntington, there are three ways a country can democratize. There can be transformation of the power elites; however, the likelihood of this occurring under Hu Jintao, with supporters in the Standing Committee, is slim. Democratic transition may occur through replacement, but the support the CCP receives from wealthy entrepreneurs and the middle class precludes the likelihood of collapse or overthrow. Transformation, the third path to democratization, is a mixture of the first two forms and thus unlikely. The verdict is in; despite prospects for democracy, the hallmark of the present era will be increased market reform without the democratic baggage.

A Teaching Restatement — Purdue Global (17 July 2026)

Purdue Global Academic Success Center published a classroom restatement on 17 July 2026. Travis Murphy's post is a teaching pair, not a study: a policy thesis about a homeowners' association, then a closer that restates the claim after the body. Use it to see the restatement move. Do not cite Wrightwood as if it were a municipal report.

Purdue Global Academic Success Center, “Writing The Conclusion Paragraph” (17 July 2026)

Original teaching thesis:

The Wrightwood homeowners’ association should put street cleaning services in the budget because doing so will remove trash from the street gutter, deter pests, and improve the overall appearance of the neighborhood.

Restated claim after the body:

Providing street cleaning services to the Wrightwood neighborhood will help improve the neighborhood in three important ways.

Full teaching closer (no new evidence):

Providing street cleaning services to the Wrightwood neighborhood will help improve the neighborhood in three important ways. The services will keep the gutters clear, allowing rainwater to drain and avoid flooding. It will also deter pests by removing easy food sources. Improving the look and safety of the neighborhood will result in a better appearance, adding value to all the homes in the neighborhood. Having street cleaning services will improve the entire neighborhood and benefit all of the residents equally.
You may notice that the example uses no new evidence. This is because there should be no new ideas, information, or claims to support in your concluding paragraph.
Avoid starting the concluding paragraph with phrases like “In conclusion” or “In summary”. Phrases like these are informal because they are directed toward the reader and create a sudden shift in tone.

Purdue Global's closer is a clean restatement. UNC and Harvard would still ask for a sharper so-what than "benefit all of the residents equally." Use the pair to learn the restatement move, then play UNC's "So What" game on the last sentence.

Discipline Differences You Should Not Ignore

Harvard's first paragraph is the assignment check: ask the instructor what the discipline wants. The split below is that page plus ICMJE and APA, not a new house style.

  • STEM and lab reports. End on what the result changes and what a next experiment would have to measure. Further-study questions belong here (Harvard). Do not add a humanities-style flourish or a policy demand the data cannot support (ICMJE).
  • Social and behavioral sciences (APA). The Discussion interprets the result against the hypothesis, names strengths and limits, and proposes directions (Publication Manual §§3.8, 3.16). Present tense. Do not reprint the Results.
  • Medicine (ICMJE / Vancouver). Open with the main finding and a mechanism, place it in the totality of evidence, discuss sex and/or gender where those variables matter, state limits, and separate clinical from statistical significance. Do not claim priority or discuss unfinished work.
  • Humanities. A reread of a central quotation is often better than a call to action (Harvard). UNC's Woolf and Douglass examples are the model: implication inside the scholarly conversation, not a public-service announcement.
  • Policy briefs. Toronto notes that a briefing may end with recommendations rather than a reflective closer. Follow the genre, not this essay template.

Limitations Without an Apology

A limitation sentence is part of the interpretation, not a confession that the paper was a waste of time. APA and ICMJE both require limits in empirical Discussion sections. UNC's essay handout does not tell you to undermine a literary argument with "this is only my opinion." Harvard is blunt: if you need to explain scope, do it earlier — and do not apologize for the paper you did not write.

Limitation sentences that do the job (templates, not studies)

Use when the sample or archive is bounded:

This reading is limited to [the texts or years you actually analyzed]; it does not settle how later editions or other archives would answer the same question.

Use in an empirical Discussion:

The design cannot separate [variable A] from [variable B], so the association should not be read as a causal claim. A useful next study would [the test that could separate them].

Do not write:

This paper has many limitations and may be wrong, but I still hope it is useful. (That is Harvard's apology, not a limitation.)

Do / Don’t at a Glance

Do

  • Restate the claim in light of the evidence (Harvard's what)
  • Answer UNC's "So what?" until the sentence is about stakes, not the topic
  • Leave a now-what: question, application, rereading, or next study
  • Match the genre: essay closer vs APA/ICMJE Discussion
  • Keep the last sentence specific (Hamilton; Toronto)

Don’t

  • Open with "in conclusion," "in summary," or "in closing" (UNC, Harvard, Hamilton, Purdue Global)
  • State the thesis for the first time here (UNC's Sherlock Holmes)
  • Introduce a new case, quotation, or statistic that belongs in the body
  • Swap analysis for a patriotic or sentimental tribute (UNC)
  • Claim clinical, economic, or priority results the data do not support (ICMJE)

Revision Checklist

Before you submit

  • A reader who skipped the body would still see a claim, not a table of contents
  • The thesis is restated, not pasted from the introduction
  • At least one sentence answers "why should anybody care?" (UNC)
  • No new source, figure, or subtopic appears for the first time
  • The closer matches the assignment: essay, Discussion, or recommendations
  • Limitations, if required, change the interpretation instead of apologizing
  • The last sentence is a thought, not a throat-clearing transition
  • Every source the closer relies on is already cited in the body; format those records in APA, MLA, or Chicago

Finish the paper, then format the list

A conclusion cannot repair a missing claim. Tighten the thesis first, then write the closer. When the argument is stable, generate the reference list your course actually requires.

Conclusion FAQs

How do I write a conclusion without repeating myself?
UNC's test is synthesis, not a shorter table of contents. Restate the claim in words the evidence made possible, then say how the body paragraphs fit together. If you can delete the last paragraph and lose no idea, you wrote a summary. Play the "So What" game until a sentence is about stakes.
Can I start a conclusion with "In conclusion"?
Not in academic prose. UNC, Harvard, Hamilton, and Purdue Global (17 July 2026) all flag "in conclusion," "in summary," and "in closing" as wooden. Those phrases can work in a speech. In a paper they usually introduce a sentence that is not yet a conclusion (Harvard's example: "In conclusion, I have now demonstrated that my thesis is correct.").
How long should a conclusion be?
Toronto: one well-developed paragraph for most essays; two or three when the paper is long enough to need them. Match the closer to the essay, not to a percentage. Empirical articles use a Discussion whose length follows APA 7 or the journal, not a five-sentence closer copied from a five-page paper.
Should I introduce a new quote or statistic at the end?
Almost never as unused evidence. UNC lists leftover quotations and statistics as a failure mode (the Grab Bag). Toronto allows an apt quotation or startling statistic only when it drives home a point the body already made — not when you are sneaking in a source you forgot. If the number needs explaining, it belongs earlier.
Is a Discussion the same thing as a conclusion?
In APA and ICMJE original-research articles, the Discussion is the concluding section: interpretation, context, limitations, and implications. APA's Discussion Phrases Guide says to use present tense and not reprint Results. It is not the reflective last paragraph of a course essay. Some student papers still use a Conclusion heading because the instructor asked for one. Follow that prompt.
Where does the thesis go if I have been told not to repeat it?
Restate it, do not paste it. UNC says ending on an unchanged paraphrase is a failed closer, and stating the thesis for the first time in the last paragraph is the Sherlock Holmes error. Purdue Global's Wrightwood pair shows the restatement move: the claim returns after the body, in new words. The claim should already appear in the introduction.
Do I have to mention limitations?
In an empirical Discussion, yes — APA 7 and ICMJE expect the limits that change how the result can be used. ICMJE January 2026 also asks you to discuss sex and/or gender where those variables matter. In a humanities essay, a vague "this is only my interpretation" is an apology, not a limitation. Name the archive, edition, or comparison you actually used if that bound changes the claim.
What is the difference between a conclusion and an abstract?
The abstract is a standalone summary and often reports results so a reader can decide whether to open the paper. The conclusion assumes the reader has just finished the body and is ready for significance, not a second synopsis. Do not paste the abstract into the last section.
How do I end a literature paper?
Harvard's example is a quotation from the central text that can now be read differently. UNC's Woolf example points to later writers or later feminists. Hamilton's English 150 Coetzee sample ends on what the injuries do to the reading, not on a plot recap. Do not bolt on a public-policy recommendation the close reading did not argue.
Should I cite sources in the conclusion?
Cite when you name a source, as Harvard does with the behavioral mediation hypothesis in Roser's closer and as Hamilton's Government 112 sample does with Huntington. Do not introduce a source that never appeared in the body. Format every record the paper already uses in the style your course assigned — APA, MLA, or Chicago — and generate the list with the citation generator.