Settle the account with the audience and lower the temperature
What do you owe the audience now, and what do you change next time?
This chapter closes the cycle: reporting back to the people you asked, publishing your methods and funders, correcting errors that surface late, and doing the work of depolarisation after a contested campaign. It ends with the handover that the next planning cycle starts from.
All stages
What you owe the audience
The cycle ends where it began, with the audience. You asked people for their attention and, in some cases, their trust; the review stage settles the account. Tell them what happened, through the same channels and messengers you used to ask, in the same plain language: what the objective was, whether it was met, and what you learned. Thank the people who took part in research and testing and show them what their answers changed. Keep the promises the campaign made, especially about services, and say how long they will last. Delete the personal data on the date you set and tell people you have done it. Stay reachable: the contact line outlives the campaign.
Publish methods and funders
Publish a report that would let a stranger check your work or repeat it: the objective and baseline, how the audience was chosen and on what criteria, the messages and products, who the messengers were and what any of them received, the budget and every funder, the tests and thresholds, the approval record, the full results including nulls, the harm findings, and the corrections log. This is the publicity condition in practice. An influence campaign is legitimate when the people it was aimed at could see the whole of it and not feel tricked. If the work was election-related, statutory finance reports are a separate duty and do not replace this one.
Correct errors, including late ones
Errors discovered after the end still need correcting. Send the correction where the error went, lead with the fact, and remember that a retracted claim keeps influencing people unless the correction explains what is true instead. Where the campaign caused harm, say so, apologise to the people affected, and put it right as far as you can. Do not quietly delete what you published. Keep an archive, with corrections marked, so that the record of what was said stays complete.
Lower the temperature
A contested campaign leaves a community more divided than it found it unless someone works at the opposite. That work is yours whether or not you met your objective. Do not take a victory lap, and do not name or mock people who were not persuaded. Say in public which of your critics' concerns were fair.
People on each side of a dispute tend to overestimate how hostile the other side is, and correcting that misperception has been found to reduce negative attributions and support for partisan violence (Lees and Cikara; Mernyk and colleagues). A megastudy by Voelkel and colleagues found that the most effective ways of reducing partisan animosity presented relatable, sympathetic people with different views and stressed identities that both sides share. So meet opponents, listen, find a goal you hold in common, and work on it together where everyone can see.
Hand over to the next cycle
Finally, write the handover for the next planning cycle: what to keep, what to stop, what to test next time, and which limit came under most pressure and why. The next objective starts from this page.
Techniques at this stage, and where the line is
Read critics' objections in their strongest form and say publicly which of them were right.
The line: A show of fairness in public while the report misstates what critics said is charity used as cover.
Late corrections must reach the people who saw the error and must say what is true, not only what was wrong.
The line: Correcting a favourable error where nobody will see it, and relying on the original to keep working, exploits the effect.
There is no honest use. Keep an archive of everything published, with corrections marked.
The line: Deleting or silently editing what you put out removes the audience's ability to hold you to it.
There is no honest use. The honest form is full disclosure of methods, funders, results, and harm.
The line: Admitting a small fault in order to draw attention away from a larger one is a managed confession.
You owe the audience a return for their attention and trust: results, thanks, and kept promises, given as a debt paid and with no new request attached.
The line: Using thanks or gifts to create a sense of obligation ahead of the next ask turns gratitude into leverage.
Check it in your own team during review: people who were not persuaded are not thereby foolish or malicious.
The line: Telling supporters that only the ignorant or the bad could disagree builds contempt that outlasts the campaign.
Expect each side to read your report as biased against it, and invite critics to review the draft before it is published.
The line: Stoking a sense of grievance about unfair treatment, to keep supporters mobilised, feeds the division you are supposed to be reducing.
Worked example
Fictional example, concluded. The Riverside Health Partnership recorded 1,024 catch-up visits against an objective of 1,100, leaving 1,290 children still behind, with nine complaints and one correction on the log.
- 1Report back: in September each of the 1,720 households receives a one-page letter from the school nurse, in the same three languages, saying what the campaign aimed for, what happened, and that evening clinics continue every Tuesday and on the first Saturday of each month until December.
- 2Publication: the full report goes on the county website with the objective and baseline, the targeting criteria (zip code and language), every product, the names of the messengers, the voucher scheme (94 vouchers, 2,350 dollars), the 48,000 dollar budget with its two funders (30,000 dollars from the county general fund and 18,000 from a state immunisation grant), the test results, the approval sheets, the null result on advertising, the harm findings, and the corrections log.
- 3Apology: the six complaints about the registry-based letter lead to a written apology for using registry records for outreach without having told families beforehand that this could happen. The department adds a notice to its registry leaflet.
- 4Data: research notes are deleted on 30 September as promised, and the report records the deletion.
- 5Lowering the temperature: the Partnership invites the blogger who criticised it and three parents who decided against vaccinating to a session hosted by the parish, with no persuasion on the agenda. It publishes which of their concerns it accepts. The group agrees on one shared task: a plain-language guide to school health requirements, including how exemptions work under state law, reviewed by both sides.
- 6Handover: keep the evening clinics and the nurse letters, drop paid social advertising, test a letter from the family's own doctor. The limit that came under most pressure was the cap on text reminders when numbers lagged in week six; it held.
Outcome. A published report, an apology, a deleted dataset, a continuing clinic timetable, and a first piece of joint work with former critics. The next cycle's objective is drafted from the handover and concerns the 1,290 children still behind.
Review and close-out
Complete this with all partners present. Nothing here is optional because the campaign went well. Saved in this browser only.
What you will tell them, in which languages, through which channels and messengers.
Every commitment the campaign made, and its status.
Tick each item that is in the public report.
Errors or harm found since the end, who was affected, and what you have done.
The date personal data was deleted and who confirmed it.
Which critics' concerns were fair, who you will meet, and the shared goal you will work on.
Agree this with every partner before any public statement.
Keep, stop, test next time, and the limit that came under most pressure.
Ethics gate: closing the cycle
0/8If any of these is not true: Do not close the campaign or begin the next one. Finish the report, issue the correction or apology, delete the data, or withdraw the statement, and then return to this gate.
Common mistakes
- ×Reporting to the funders and never to the audience.
- ×Taking a victory lap that turns people who were not persuaded into the losers of a contest.
- ×Publishing results without the methods and funders that would let anyone check them.
- ×Deleting embarrassing material in place of correcting it.
- ×Leaving the lessons in one person's head, so that the next cycle repeats the mistakes.
What do I do about it: ignore, verify, report, rebut, or prebunk?
Related
References
- Department of the Army (2013). FM 3-53: Military Information Support OperationsEvaluation feeds the next planning cycle; the process is iterative and does not end with dissemination.
- Sunstein, C. R. (2016). The Ethics of Influence: Government in the Age of Behavioral Science. Cambridge University PressTransparency and accountability as conditions of legitimate influence: the influencer should be able to disclose what was done.
- Lewandowsky, S., Cook, J., Ecker, U. K. H., et al. (2020). The Debunking Handbook 2020Misinformation continues to influence people after retraction unless the correction supplies the accurate account.
- Voelkel, J. G., Stagnaro, M. N., Chu, J. Y., et al. (2024). Megastudy testing 25 treatments to reduce antidemocratic attitudes and partisan animosity. ScienceThe most effective treatments for reducing animosity highlighted sympathetic people with different beliefs and shared identities.
- Lees, J., & Cikara, M. (2020). Inaccurate group meta-perceptions drive negative out-group attributions in competitive contexts. Nature Human BehaviourPeople overestimate how negatively the other side sees them, and correcting this reduces negative attributions.
- Mernyk, J. S., Pink, S. L., Druckman, J. N., & Willer, R. (2022). Correcting inaccurate metaperceptions reduces Americans' support for partisan violence. Proceedings of the National Academy of SciencesCorrecting exaggerated beliefs about the other side's hostility reduced support for partisan violence.