When a clean energy campaign ends, which of its materials do people actually remember: the explanation they watched, or the release they were sent? We ran a campaign post test for a clean energy group using a pre and post design with 1,600 respondents plus a geographic holdout area that received no campaign. Creator explainers drove 52 percent recall, against 23 percent for press release distribution, a gap wide enough to move money.
I Discrimen
A person explaining beat a document announcing, by twenty nine points
The format gap in this study is not a margin; it is a different league of memory.
Creator explainers drove 52 percent recall in our post test, while press release distribution drove 23 percent. The release is the format communications teams control most tightly and distribute most widely, and it was remembered by fewer than one in four people. The explainer, a person talking an audience through an idea in their own idiom, was remembered by more than half.
Independent evidence has pointed this way for years. Nielsen reported in 2016 that branded content generated an average of 86 percent brand recall among viewers, compared with 65 percent for pre roll advertising, across more than 100 pieces of content it analysed.
The lesson generalises beyond energy, and teams assessing how to choose a PR agency for consumer technology will recognise the same pattern in product launches. People remember people. A document asks to be filed; a face asks to be followed, questioned, and remembered.
Memory attaches to a voice, not to a letterhead.
II Machina
A pre and post design with a holdout separates the campaign from the noise
Recall claimed without a comparison area is weather reporting, not measurement.
Our post test surveyed 1,600 respondents before and after the campaign, and held back a geographic area, a geo holdout, where the campaign did not run. That structure matters because recall moves for many reasons: news coverage, weather, energy prices, and the general conversation. The holdout gives the study a baseline for how much recall would have drifted upward anyway.
The instruments were kept deliberately plain. Respondents were asked what they recalled of the campaign and its messages, and answers were recorded against the formats the campaign had used. We measured recall of the campaign content itself, not approval of it.
This is the same discipline that makes the analysis habits in choosing a PR agency for data analytics worth importing into campaign work. Define the measure before the campaign runs, protect a comparison, and let the unflattering numbers survive to the report. The 23 percent in this study is exactly the kind of figure that disappears from decks without a holdout to defend the method.
Measure against somewhere the campaign never went, or measure nothing.
III Exemplum
The study in full: creator recall above the top quartile, release recall below the median
Every figure below is from the post test record, benchmarks included.
The record sets a Benchmark Creator Led Recall median of 31 percent and a top quartile of 44 percent. Creator explainers in this campaign reached 52 percent, above that top quartile. Press release distribution reached 23 percent, below even the median benchmark for creator led work. The comparison is not perfectly matched, since the benchmark belongs to the creator format, but it shows the release performing under the midpoint of the format that beat it.
| Measure | Post test | Benchmark median | Benchmark top quartile |
|---|---|---|---|
| Creator led recall (percent) | 52 | 31 | 44 |
| Press release distribution recall (percent) | 23 | Not benchmarked in the record | Not benchmarked in the record |
| Treatment area recall lift (points) | 13 | Not benchmarked in the record | Not benchmarked in the record |
| Geo holdout recall lift (points) | 4 | Not benchmarked in the record | Not benchmarked in the record |
| Respondents, pre and post combined design | 1,600 | Not benchmarked in the record | Not benchmarked in the record |
fons: Aerovector research record, post test waves pre and post, for every figure and benchmark shown.
Notice the holdout's behaviour in that table. Recall in the untreated area still rose by 4 points, which is the background drift a naive post test would have credited to the campaign. The treatment area rose by 13 points. Reporting both numbers is what makes the 13 believable.
A result with its control group attached is a result you can defend.
IV Damnum
Defaulting to the release spends the budget on the format remembered least
The cost of habit is paid in the impressions that leave nothing behind.
A press release still has legitimate work to do: it creates the formal record, serves journalists who need verifiable detail, and anchors search results. The damage begins when the release becomes the campaign, the main vehicle expected to carry public memory. At 23 percent recall against a benchmark median of 31 percent for creator led recall, that expectation fails on this study's own numbers.
The wider information environment keeps moving against document led habits. The Reuters Institute Digital News Report 2025 documents the continuing shift of news attention toward social and video networks, where individual presenters and creators shape what audiences encounter. A campaign that publishes text into that environment and stops there is whispering in a room where everyone else is on camera.
Campaign windows in energy are tied to consultations and funding rounds, and a window spent on a low recall format does not reopen.
Habit is the most expensive line in any media plan.
V Genus
Clean energy audiences divide by belief and by bill, and creators reach the divisions differently
One national message lands as six different arguments in six different living rooms.
Climate audiences are famously segmented. The Yale Program on Climate Change Communication has for years divided Americans into six audiences, from the alarmed to the dismissive, based on their beliefs and concern about global warming.
Creators also sit inside the news habits of younger audiences in particular. Pew Research Center reported in 2024 that 21 percent of United States adults regularly get news from news influencers on social media, and that the share is far higher among younger adults. For a clean energy group trying to reach householders under forty, the explainer format is not an experiment at the edge of the plan. It is where the audience already lives.
Most of that viewing happens on phones, in short sessions, which makes the mobile discipline in how to choose a PR agency for mobile applications a planning input rather than a design afterthought. An explainer that cannot survive a small screen and a muted autoplay has failed before its argument begins.
Speak to the segment in front of you, in the format it already trusts.
VI Regio
The holdout showed that place itself changes what a campaign can claim
Geography is not a targeting detail in energy; it is half the story.
Clean energy is experienced locally: a wind project on one horizon, a retrofit scheme on one street, a grid upgrade in one county. Our geo holdout existed precisely because regions differ, and it returned a 4 point recall rise with no campaign at all, reflecting whatever local conversation was already running there.
The global backdrop explains why local messages carry such charge. Ember reported in its Global Electricity Review 2025 that low carbon sources supplied about 41 percent of global electricity in 2024, a transition audiences can now see in their own landscapes and bills. The International Energy Agency likewise tracks renewables as a central growth story of the energy system, which means local campaigns argue against a moving background, not a static one.
Treat each region as its own test cell where you can. The holdout discipline that protected this study's conclusions is available to any campaign big enough to have a map, and it converts regional variation from a reporting nuisance into evidence about where the work is actually landing.
Place is a variable, so measure it like one.
VII Occasio
Reallocating thirty percent to creators lifted treatment recall thirteen points
The action was a budget decision, and the budget is where the result came from.
The PR action our post test evaluated was to reallocate 30 percent to creators. Not the whole budget, and not a garnish on the release plan: a third of the spend moved toward creator explainers, with the release retained for the record and for journalists. The format mix, not a new slogan, was the intervention.
The recorded outcome was treatment recall up 13 points, against up 4 points in the holdout. Counted plainly, that is a nine point difference between the area that received the rebalanced campaign and the area that received none, after the background drift in the holdout is taken into account. We note that recall is a memory measure, not a behaviour change measure.
The modesty of the reallocation is part of the finding. A team does not need to abandon its existing machinery to test whether creators carry memory better. It needs to move enough spend to generate a readable result, in this study 30 percent, and to protect a comparison area so the result can be believed.
Move a third, measure honestly, then argue about the rest.
VIII Structura
Creator work still needs a brief, a disclosure, and a message worth repeating
Freedom inside a frame is what makes an explainer both credible and useful.
The structure that sits behind effective creator content is unglamorous: a Message House that fixes the claims every explainer must carry, a brief that leaves the delivery in the creator's own voice, and clear disclosure that the content is part of a campaign.
The stakes of clarity are high in this sector. The IPCC, in its Sixth Assessment Report work on mitigation published in 2022, assessed that demand side measures could reduce emissions in end use sectors by 40 to 70 percent by 2050, which makes household understanding a material part of the transition rather than a communications garnish. UNEP's Emissions Gap Report 2024 recorded global greenhouse gas emissions rising to a record 57.1 gigatonnes in 2023, so the explaining cannot wait for perfect channels.
Connected home technology gives creators concrete things to explain, from smart meters to heat pumps, and the device literacy behind how to choose a PR agency for Internet of Things (IoT) is a useful cousin discipline here. An explainer holding the actual device, in an actual kitchen, is the format working at full strength.
Brief the claim tightly and the delivery loosely.
IX Consilium
Post test every campaign, protect a holdout, and let recall move the budget
The prescription is a measurement habit before it is a media preference.
Build the post test into the campaign plan before launch, with the pre measure taken early enough to be a true baseline. Then set the reallocation for the next round from the recall evidence, as this study did when it moved 30 percent toward creators.
Publish the campaign's core explainers where search engines and AI answer tools can find, transcribe, and cite them, because recall now compounds through machine summaries as well as memory, a visibility question explored in how to choose a PR agency for artificial intelligence (AI). The full programme scope, from Media Map planning to Coverage Reporting, is set out in our public relations services.
| What your post test shows | Budget response | Measure to watch |
|---|---|---|
| Creator recall above the 44 percent benchmark top quartile | Reallocate a further share toward creator explainers | Recall in the next post test, treatment against holdout |
| Creator recall near the 31 percent benchmark median | Keep the creator share and improve briefs and creator fit | Recall gap between creator and release formats |
| Release recall leading the formats | Keep the release central and test creators at a small share | Whether the gap closes as creator share rises |
fons: Aerovector research record for the benchmark figures; decision rows are this post's prescription built on the post test findings.
Repeat the design each round and the budget conversation changes character. Instead of defending formats by loyalty, the team argues from a table of recall results with a control attached. That is a shorter argument, and a better use of a campaign meeting.
Evidence ends format arguments faster than taste does.
X Quaesita
Questions
What did the campaign post test measure?
We surveyed 1,600 respondents before and after a clean energy campaign and kept a geographic holdout area where the campaign did not run. The post test measured recall of the campaign and compared the formats it used, principally creator explainers and press release distribution.
Why did creator explainers outperform press releases so heavily?
Creator explainers drove 52 percent recall against 23 percent for press release distribution. The study records the size of the gap rather than a single cause, but a person explaining an idea in their own voice gives memory more to hold than a formal document designed for the record.
How did creator recall compare with the benchmarks in the record?
The record sets a Benchmark Creator Led Recall median of 31 percent and a top quartile of 44 percent. Creator recall of 52 percent in this campaign sat above that top quartile, while release recall of 23 percent sat below the creator benchmark median.
What was the geo holdout for?
The holdout area received no campaign, so its recall movement shows background drift from news and general conversation. Recall there rose 4 points, which lets the treatment area's 13 point rise be read against a baseline instead of being credited entirely to the campaign.
What PR action did the study evaluate?
The action was to reallocate 30 percent to creators, moving a third of the spend toward creator explainers while keeping the press release for the formal record. The outcome was treatment recall up 13 points against 4 points in the holdout.
Does this mean press releases are useless?
No, because the release still creates the verifiable record journalists and search tools rely on. The study argues against expecting the release to carry public memory on its own, at 23 percent recall, when an explainer format in the same campaign reached 52 percent.
How large a creator reallocation should another campaign test?
Our study tested 30 percent, which was enough to produce a readable difference against the holdout. A smaller test may show direction without confidence, so we would start near that share and protect a comparison area so the result can be trusted.
How this guide was built
This post reports Aerovector's own research record for a campaign post test of a clean energy group: a pre and post survey design with 1,600 respondents, a geographic holdout, the recall measures and benchmark figures stated in the text, a reallocation of 30 percent to creators as the PR action, and the treatment and holdout outcomes recorded afterwards. The study figures are Aerovector's own research record, and identities are withheld, so the organisation is described only as a clean energy group. The study covers one campaign, recall measures memory rather than behaviour change, and its benchmarks come from the same research record. Third party sources were used only to set context on content recall, news habits, audience segmentation, and the energy transition, and each is named with its year where it appears.
Sources
- Aerovector, Campaign post-test research, 2026: the first-party study reported in this post, figures from the practice research record, identities withheld.
- Nielsen, Quality Branded Content Outperforms Pre-Roll Advertising, reporting 86 percent brand recall for branded content against 65 percent for pre roll, 2016. https://www.nielsen.com/insights/2016/quality-branded-content-outperforms-pre-roll-advertising/
- Reuters Institute, Digital News Report 2025, documenting the shift of news attention toward social and video networks, 2025. https://reutersinstitute.politics.ox.ac.uk/digital-news-report/2025
- Pew Research Center, America's News Influencers, reporting 21 percent of adults regularly getting news from news influencers, 2024. https://www.pewresearch.org/journalism/2024/11/18/americas-news-influencers/
- Yale Program on Climate Change Communication, Climate Change in the American Mind, the six audiences segmentation of climate beliefs and concern, 2024. https://climatecommunication.yale.edu/publications/climate-change-in-the-american-mind-beliefs-attitudes-spring-2024/
- Ember, Global Electricity Review 2025, reporting low carbon sources supplying about 41 percent of global electricity in 2024, 2025. https://ember-energy.org/latest-updates/global-electricity-review-2025/
- International Energy Agency, Renewables energy system overview, tracking renewable deployment across the energy system, 2025. https://www.iea.org/energy-system/renewables
- IPCC, Sixth Assessment Report, Mitigation of Climate Change, assessing demand side measures at 40 to 70 percent of end use emissions by 2050, 2022. https://www.ipcc.ch/report/ar6/wg3/
- UNEP, Emissions Gap Report 2024, recording global emissions rising to a record 57.1 gigatonnes in 2023, 2024. https://www.unep.org/resources/emissions-gap-report-2024
- United States Energy Information Administration, Renewable energy explained, an overview of renewable sources and their role in the energy mix, 2025. https://www.eia.gov/energyexplained/renewable-sources/
XI Incipe
When the release is forgotten and the explainer is remembered, Aerovector plans for the explainer
Our Media Map work identifies the creators your audiences already watch and trust, before a Message House fixes the claims each explainer must carry and a Newsroom Calendar sequences releases and creator content as one campaign. Coverage Reporting then measures recall and coverage together, with a holdout where the campaign allows one, so the next budget round argues from evidence. That is the programme this post test was measuring a piece of.
With Aerovector, clients get support ranking in AI search results alongside media coverage that carries a guarantee, and those two working together are what produce leads. For a clean energy group, that means your best explainer is published, transcribed, and cited where journalists and AI answer tools look, and the coverage it earns is coverage you can point to. If your last campaign was remembered at anything like 23 percent, send us its format mix through the form on our contact page, with the regions it ran in and the recall you recorded, and we will show you how a creator reallocation would be structured and measured.
Aerovector builds campaigns around the formats people actually remember.
Additional Details
Description: Our Research: Campaign post-test for a Clean-energy group: what the study measured, what it found, and what the finding changes for teams facing the same.
Published: 2026-10-25