Skip to content

Latest US poll: 63% say gas prices are affecting their household budget a great deal or a moderate amount

See the poll

Social feed testing

Will your ad stop the scroll?

Place your social ad in a mock feed among ordinary posts and let people scroll as they normally would. Afterwards, find out whether they noticed it, remember whose it was and want to see more.

5audience groups compared side by side, at most
309ready-made question templates, ad questions among them
20question types, each with its own chart

Why test in the feed

Your ad competes with everything else in the feed

On social platforms an ad sits between friends' photos, news and other brands. A test that plays the ad on its own tells you whether it is clear, but not whether anyone would stop for it. A feed test puts the ad back where it will actually run.

What can go on the bench

Vertical videoSquare imagesStatic social postsShort video adsA mock feed of ordinary posts

Feed versus close up

In the feed first, then close up

Each row is one measure, named the way the report names it.

Base: US adults aged 18-34 who use social media every day; the same people, first in the feed, then close up.

Shares of people. In the feed = before they saw the ad again; close up = after a second, focused look.

Social feed testing · scorecardBrand A · feed test report
Measure
FeedIn the feed
CloseClose up
NoticedDid they spot your post among the others?
64%spotted the post
not asked here
Brand recallCould they say whose ad it was?
41%
not asked here
MessageWhat did they think it was about?
23%named the new flavor
72%got the message
InterestWould they tap to find out more?
29%
47%
Platform fitDoes it feel like it belongs in their feed?
not asked here
68%feels at home in the feed

Scroll, then look

Watch an ad try to stop the scroll

  1. 01The ad goes into the feed
  2. 02People scroll, then remember
  3. 03Then a close look
  4. 04Did it stop the scroll?

Step 1 / 4

The ad goes into the feed

Your post sits in a phone-style feed among other posts, in the format you plan to run.

Study set-up

Social feed testing

  • StimulusA 15-second vertical video for a new snack flavor
  • FeedA mock feed of ordinary posts, the ad placed among them
  • WhoUS adults aged 18-34 who use social media every day
  • FirstScroll as they normally would, with no hint
  • ThenWhat they remember, then the ad again, close up

Set up with our team, or yourself in CatalyseAI

Step 2 / 4

People scroll, then remember

No hint about which post is tested. Afterwards: which posts and brands do they recall?

After the feed

Which of these brands do you remember seeing?

  • Brand A snacks
  • A coffee shop
  • A phone plan
  • A sneaker brand
Next

"I saw the chips but not the flavor."

"It looked like a normal post, so I kept going."

"The crunch sound made me stop."

Step 3 / 4

Then a close look

They see your ad again on its own and rate its appeal, its message and its fit with the platform.

The same people, twice

In the feedClose up

  • 64%Noticednot asked
  • 41%Brand recallnot asked
  • 23%Message72%
  • 29%Interest47%
  • not askedPlatform fit68%

In the feed: before they saw the ad again. Close up: after a second, focused look.

Step 4 / 4

Did it stop the scroll?

In-feed results next to close-up results show whether the ad gets noticed, and what got lost.

Verdict

Noticed, but the flavor got lost

Put the name in the first second and test the feed again.

  • The 15-second videoName it sooner
  • Close upThe ad works

The set-up

How the feed test works

From the feed to the close look.

  1. 01

    Build the feed

    Your post goes into a phone-style feed with other posts around it, in the vertical or square format you plan to run.

  2. 02

    Let people scroll

    Each person scrolls at their own pace, with no hint about which post is being tested.

  3. 03

    Ask what they remember

    Which posts and brands do they recall? Which one was yours, and what was it about?

  4. 04

    Show it again, close up

    Then they see your ad on its own and rate its appeal, its message and how well it fits the platform.

  5. Then: decide

    Put the name in the first second and test the feed again.

    Plan your study

What got lost

Example: a new flavor in the feed

A snack brand has a 15-second vertical video for a new flavor. US adults aged 18-34 who use social media every day scroll a mock feed of ordinary posts with the ad placed among them, then answer a few questions. The brand learns whether people who scrolled past still remembered the flavor, and whether the ad felt at home next to the other posts.

  • The 15-second video

    Name it sooner

    Noticed by 64% in the feed, but only 23% could name the new flavor before seeing it again.

    Put the flavor name in the first second.

  • Close up

    The ad works

    Once people look, 72% get the message and it feels at home in the feed.

    The problem is the scroll, not the ad.

Ask the studyAnswers from your own data

Did people who scrolled past still remember the flavor?

Partly. 41% named the brand from the feed, but only 23% could name the new flavor before they saw the ad again.

It checks its numbers against your data and draws the chart.

What you get

What you get

  • Output 01

    In-feed results

    How many people noticed the ad and named the brand before they were shown it again.

  • Output 02

    Close-up results

    Appeal, message and interest after a second, focused look.

  • Output 03

    Comments in their words

    Open answers on what made people stop or scroll on, grouped into themes.

FAQ

Questions, answered

Is my ad posted on a real social app?

No. People see a mock feed inside the survey that looks like a social app. Your ad is never published anywhere.

Will you measure how long people stop on my ad?

We have not confirmed this yet. The survey already records time spent on each question, but timing a single post inside a feed needs more work.

Planning a social campaign?

Tell us the platform, the format and the audience. We will suggest how to set up the feed test.

Nearest office: Orlando, FL (corporate office)