Messages That Stick: How Memory, Budget, and Your Deadline Should Shape an Awareness Campaign

By Arnaud Blanchard, Vice President of Product

A teacher and an advertiser share the same goal: get a group of people to remember a message, and make sure they still remember it on the day it counts. For a teacher, the group is a class and the day is the exam. For an election, it’s the vote.

The midterms are three weeks away. AdImpact expects $10.8 billion in political ad spend this cycle, $2.5 billion of it on streaming TV. And we just shipped a release at Vibe that changes how our Awareness campaigns decide who sees an ad, and when.

Elections just make the deadline impossible to miss. Every awareness campaign has an exam of its own: a product launch, a store opening, a premiere, Black Friday. What follows applies to all of them. Politics happens to be the loudest example this month.

A man, a notebook and 2,300 nonsense syllables

In the 1880s, Hermann Ebbinghaus decided to study memory using the only test subject he fully trusted: himself. He wrote out about 2,300 nonsense syllables, things like “WID” and “ZOF”, so that nothing he already knew could help him. He learned lists of them, then checked how much was left after an hour, a day, a week.

What came out of it is the forgetting curve. Memory drops sharply right after learning, then the drop slows down. He also noticed that going back over a list brought it back, and that reviews spread out over several days held much longer than the same number of reviews crammed together.

Any teacher knows this without having read Ebbinghaus. Nobody teaches a chapter once in September and expects it to survive until June. You introduce it, come back to it a few days later, then a bit later again, and you do one last review the week before the exam.

What the classroom tells a campaign

Three ideas carry over almost as is.

The law of first impressions. The first time an audience sees your message, it goes from unknown to familiar. That’s the biggest jump any single impression will give you. The ones after still help, a bit less each time.

Scheduling the syllabus. A reminder is worth the most when the message is starting to fade. Run the same ad on the same screen twice in one evening and the second one adds very little. Four days later, it does real work.

Having the last word. The last review before an exam is the one a class remembers best. Advertising works the same way: the most recent message is usually the one an audience remembers. Except that in a lot of places, the vote isn’t one day anymore. Almost half of 2022 midterm voters voted early or by mail. For them, the exam happened weeks before Election Day.

There is no optimal frequency

The question we hear most from advertisers is some version of “what’s the right frequency?” Three? Five? Eight?

There isn’t one. There is an optimal exposure strategy, and it depends on three things: the budget you have, the size of the audience you’re going after, and when the exam is.

Take a state house race in a compact district with a decent budget. You can reach most of your audience early, then spend the rest reinforcing, spaced out, building up as ballots go out. Now take a statewide race with the same budget. You’ll never cover everyone more than a couple of times, so a first impression on a new household is worth more than a fifth one on a household that has already seen the ad several times. Same money, close to opposite plans. Move the vote two weeks closer and both plans change again.

Brands face the same math. A DTC brand going national for the first time on a modest budget should favor new households over repeats. A regional retailer with a big holiday budget can afford to reinforce, and should tighten the spacing as Black Friday gets closer.

We had this exact debate internally about reach versus awareness. In the end everyone was right. It’s a question of time horizon. A fixed frequency cap can’t see any of that: it treats a third impression the same way whether it lands an hour after the second one or a week after.

Reach is good. It’s just not the whole picture

Reach is still the first thing to get right. An audience that never saw your ad can’t remember it. But a reach number only tells you how many households saw the ad at least once. It doesn’t tell you when, how often, or how much of that audience still remembers the message today.

So we added the two missing variables to how we deliver ads: exposure and time. Vibe’s Memory Score estimates how much of your audience is likely to still remember your message, using only how many times and how recently your ad ran. It goes up with each exposure, fades as days pass, and fades more slowly when exposures are well spaced. It comes from ad delivery, nothing else.

In the Insights tab of any Awareness campaign, you can see how many households you reached next to how many are still likely to remember the ad, and how the two move over the flight.

Laurence: more reach, by not overdoing it

Laurence is the bidder behind our Awareness campaigns since our latest release, and it runs on the same logic.

Most Awareness buying spreads budget as wide and as cheap as it can, and some households end up seeing the same ad again and again while others never see it. Laurence checks your campaign’s delivery history before it bids. Where your ad ran an hour ago, there is little left to gain from showing it again, so Laurence pays less. As days pass, the value comes back up. The money it stops spending on overexposure goes where there is more remembrance to create across your audience: households that haven’t seen the ad yet, and households that saw it a while ago.

In live A/B tests at equal budget, Awareness campaigns reached 37% more unique households. Same spend, just less of it wasted on repeats.

We test every version of the bidder the same way, against the previous one with real budget on both sides, and we’ll keep improving it.

The exam is closer than it looks

If you’re on a race right now, the useful date probably isn’t November 3. It’s when mail ballots land and early voting opens in your state. If you’re a brand, it’s the launch date, the first day of the sale, the opening weekend. Either way, that’s when you want the most people remembering your message, and the flight should be planned backward from it.

No contracts and no minimums also means you can change creative or move budget the day things shift.

Teachers figured out the forgetting curve a long time ago. TV is finally able to buy against it.

Sources