Case data: three campaigns traced from Ahrefs, with method notes
Popularity campaigns for SEO

The other half of ranking.

Links and content move rankings in months. Popularity moves them in days. We supply it: real people from paid media, sent straight to your pages.

Book a call See the data
24-hour test on one URL
Case Finance · Nordics 4.2K → 13.9K est. monthly organic visits · 12 Jul – 6 Oct 2025 +233% Case iGaming · Japan 19.6K → 36.4K est. monthly organic visits · first 7 days after launch, Oct 2025 +86% Case News publisher 15.8K → 30.5K est. monthly organic visits · 14 Jul – 7 Oct 2025 +93%

Ahrefs traffic estimates, traced from the screenshots in each case. Baseline: 5-day pre-launch average or launch-day value, whichever is higher.

01Mechanism

Two ranking layers. We work on the faster one.

The core layer scores pages on content and links. It moves slowly: a new link can take weeks to be crawled and counted.

A second layer re-ranks the results on popularity: how much real use each page gets. It’s a separate system that refreshes continuously, which is why it can move a page in days.

FIG.2 · Two-layer ranking, schematic. Signals as listed in court exhibit PXR0356.1

  1. 1.1

    Links and content set the baseline.

    Most SEO budgets go here, and they should. A page has to be in contention before popularity can lift it.

  2. 1.2

    Popularity re-ranks.

    Google rarely talks about it, but a Google engineer’s account in court lists a popularity signal right after PageRank,1 and Google’s leaked API documentation records site-level views.2

  3. 1.3

    We supply the popularity.

    Real people from paid media, sent straight to your pages. That’s the input the popularity layer reads.

From the court record

“PageRank. This is a single signal relating to distance from a known good source, and it is used as an input to the Quality score.

(popularity) signal that uses Chrome data.”

Notes of a call with Google engineer Hyung-Jin Kim, “Other Signals”. U.S. v. Google LLC, exhibit PXR0356, 2025.1 · justice.gov PDF ↗

Links and popularity, one after the other in a Google engineer’s list of ranking signals. Most SEO budgets only pay for the first.

Real people, not bots.

Every visit is a real person, reached through paid media.

source
paid media
google_search
untouched
site_access
none
rankings_in
your Search Console
02Evidence

Three campaigns, traced from source. Dips included.

Each line is the Ahrefs organic-traffic estimate for the client’s domain, traced day by day from the screenshot published with the case. The marker is the day the campaign started. Hover to read any day.

Ahrefs figures are modelled estimates, not measured traffic. A campaign can’t be fully separated from seasonality or the client’s other SEO work; the case notes say where that matters.

Read the case notes
03Pricing

Start with one page. Scale only on evidence.

3.1

Call

We review your pages and pick one that’s already in contention: positions 4–20 for a keyword worth moving.

3.2

Test

We agree the URL, the keywords and the starting positions in writing, then send 24 hours of traffic to the page. You see the rank checks as they come in.

3.3

Monthly

If the test moves, we extend to more URLs on a monthly plan. If it doesn’t, you’ve lost a day, not a retainer.

Test
24 h
$100 · one-off
  • One URL, its main keyword and long tail
  • Rank checks before, during and after
  • Swap to another URL if the first doesn’t respond
Start with a test
Monthly
Custom
per month
  • Several URLs and keyword sets
  • Named account manager
  • Reporting on your schedule
  • Priced on URLs, traffic volume and market
Discuss a plan

Terms on every plan

Unlimited URL rotation

Change target URLs whenever you like.

Category exclusivity

We don’t work with your direct competitors in the same market segment.

Regular reporting

Rank and traffic movement on a fixed schedule.

Algorithm updates

When Google changes, we adjust the campaign. It’s part of the plan.

04Risk

What this can’t do, and what can go wrong.

Works best when

  • The page already ranks 4–20 for the target keyword.
  • It’s worth the visit: fast, relevant and usable.
  • The keyword has real search demand.

Not a fit when

  • The page isn’t indexed or sits past page two. Fix the core layer first.
  • The page is thin, slow or broken. More visitors won’t fix that.
  • You need a guarantee. Nobody outside Google can honestly give one.
  1. 4.1

    Nothing moves.

    Some keywords are held in place by brand intent or by very strong incumbents. That’s what the test is for.

  2. 4.2

    Gains fade after you stop.

    Popularity is measured over a rolling window. Without new visits, positions drift back toward where content and links alone would put them, usually gradually.

  3. 4.3

    Google discounts the signal.

    Google doesn’t endorse attempts to influence its ranking signals, ours included. We size each campaign to the page’s real demand, but we won’t promise zero risk.

Questions

Is this bot traffic?
No. Every visit is a real person, reached through paid media.
Do you send searches or clicks through Google?
No. Nothing we do touches Google Search: no queries, no result clicks. Visitors go straight to your pages.
How do I check the results?
In your own Search Console: positions and clicks for the keywords we agreed on. You don’t have to take our reports on trust.
How fast does it work?
Movement usually starts within 24–72 hours of launch, because the popularity layer refreshes far faster than the link graph. In the three published cases, traffic was up 25% or more within 3–6 days.
Do you need access to my site?
No. We need the URLs and the keywords you want to track. No code, no logins, no content changes.
Do you work with my competitors?
Not in your market segment while you’re a client.
What does it cost after the test?
A monthly fee per campaign, based on the number of URLs, traffic volume and market. We quote after the test, when both sides have data.
Book a call

Bring one page. We’ll tell you if it’s a candidate.

A short call to look at your pages and their positions. If one is in range, we set up a test on it. If none are, we’ll say so.

Prefer email? hello@serpsoar.com

Sources

  1. U.S. v. Google LLC, No. 1:20-cv-03010 (D.D.C.), exhibit PXR0356: notes of an 18 Feb 2025 call with Google engineer HJ Kim, section III, “Other Signals”. → https://www.justice.gov/atr/media/1398871/dl
  2. Google Content Warehouse API documentation (published in the May 2024 leak): QualityNsrNsrData — chromeInTotal, “Site-level Chrome views.” → https://hexdocs.pm/google_api_content_warehouse/0.4.0/GoogleApi.ContentWarehouse.V1.Model.QualityNsrNsrData.html
  3. SISTRIX, “Why (almost) everything you knew about Google CTR is no longer valid”, 2020. → https://www.sistrix.com/blog/why-almost-everything-you-knew-about-google-ctr-is-no-longer-valid/