AdPlexity spy tool: discover 150M+ winning ads. Up to 30% off, get the discount.
Wednesday, November 18, 2026Updated · 29 new stories this weekNewsletterAdvertise
Case fact-check

134% ROI on mVAS Pops in Kazakhstan: Fact-Checking a Case the Ad Network Published First

Independent and reader-supported. Some links here are affiliate links: if you sign up through them, we may earn a commission at no cost to you. It never changes our scores or verdicts. Affiliate disclosure

Illustrated desk: a finger taps a single round button on a smartphone that drops tiny gold coins into a glass jar, while a magnifying glass shows a second jar fuller than the first, next to a striped paper megaphone, stacked browser-window cards, a padlock with 2 keys, a $486 receipt, a calculator showing 134% and a fact-check badge showing 3 of 5 stars.
Contents 12

An affiliate buys popunder traffic in Kazakhstan at $1.51 per 1,000 impressions, sends it to 1-click mobile subscriptions on Tele2, and 41 days later reports $486 in, $1,143 back and 134% ROI. Cheap pops into carrier billing is an old money machine, so a fresh case with real dashboards gets copied.

This is our review of the case for media buyers who run pops, push or mVAS: the author's text and every version of it we found, our own math on top of his dashboards, Kazakhstan's new consent rule and what the case doesn't show. Editorial fact-check: 3/5.

Short version: the traffic-side numbers are real and add up to the cent. But the case reached the public through the ad network that sold the traffic, a network that pays for cases. The payer's dashboard shows about 30% more revenue than the network recorded, and 4 weeks after the run Kazakhstan required 2-step consent for exactly this kind of subscription.

Who wrote the mVAS case and who published it first

The author is Daniel Inyang. In the version he posted under his own forum account on July 22, 2026, he writes: "Today, I'm sharing a complete breakdown of a campaign where I turned $486 in ad spend into $1,143 in revenue within 40 days using OctoClick popunder traffic and Juddy.biz. OctoClick has consistently been one of my go-to sources for high-quality popunder inventory."

In OctoClick's version he introduces himself: "I've been running a blog since 2019 and started working in affiliate marketing in 2021. I mainly focus on MVAS and sweepstakes offers." The blog checks out. An archived copy from January 2025 calls him "a Nigerian Blogger, Evangelist, Media Buyer and Legal Scholar" who "has been blogging since 2019". Its only post was about faith, and the domain no longer resolves. We found no earlier affiliate track record under his name.

His forum account was registered in September 2025 and stayed silent until July 21, 2026, when it became a paid membership and posted its first message. The case went up the next day. The account follows 1 member: OctoClick. Logged-out readers see the overview, the results and the first setup steps; the optimization part is members-only, so we read it in the public versions below.

But his post isn't where the case first appeared. OctoClick's was:

  • July 15: OctoClick posts the case in its own sales thread on a forum, under "Over to Daniel," with a 10% top-up promo code and "a special mention to our partner, Juddy.biz."
  • July 16: Juddy.biz republishes it in Russian: "Colleagues from OctoClick posted a breakdown of a campaign by our common partner" (our translation).
  • July 22: Daniel posts his own version. OctoClick replies in the thread, "Thank you, Daniel, for your excellent work and for demonstrating how our traffic performs in practice!", and adds a promo code.
  • July 23: OctoClick's Telegram channel announces the case with a link to its bot.

The Telegram post calls Daniel "our advertiser," says the network rarely publishes cases because "not everyone is ready to share their setups and show the numbers," and thanks Juddy.biz "for a strong offer and the chance to show real results on live traffic." It ends with this line: "And also: we're always open to new cases. Ready to credit authorship and pay for the material." It links to OctoClick's post from August 8, 2025: "You have a successful case? Share it with us and get a reward... Send us the details, we evaluate the case and pay for it... Write to the manager, we'll review your case and discuss the price." All quotes from the 2 posts are our translation from Russian.

Our take: this case is OctoClick's marketing, co-signed by Juddy.biz. That doesn't make the numbers fake: the network has the data, and the dashboards below are real. It means the case was selected, packaged and distributed by the seller of the traffic, under a program that pays for cases. None of that is in Daniel's own post, and nobody says whether this one was paid. Read it as a vendor case with a customer's name on it.

The setup: OctoClick pops, Juddy.biz and 1-click Tele2 Kazakhstan

From the versions combined:

  • affiliate network: Juddy.biz;
  • vertical: mVAS, WAP-click;
  • payout: not stated; the payer's screen below averages 6.3 rubles, about $0.08, per accepted subscription;
  • GEO: Kazakhstan, Android and iOS users on Tele2;
  • traffic: popunder from OctoClick, 1 campaign for Android and 1 for iOS;
  • dates: April 21 to May 31, 2026.

The payment model is "CPL" in OctoClick's and Juddy's versions and "CPA" in Daniel's. For a 1-click subscription it's the same thing: a fixed payout per subscriber.

How the flow works: on mobile data the carrier recognizes the number, and 1 tap on its page starts a paid subscription billed to the phone balance. As for the traffic, OctoClick told a prospect in its sales thread: "Our platform specializes in traffic from adult tube websites." Juddy describes its offers as running "from classic adult to mainstream (music, files, movies)" (our translation).

Not shown: the offer. No offer ID, no payout, no content type, no carrier page, no pre-lander, and no word on whether it was adult or mainstream.

Our take: Juddy.biz is run by InternetWeb LLC of Voronezh, Russia, and counts money in rubles. Outside the CIS, ask how you'll be paid before you send the first click.

The test: $5 per campaign and a cap that changes between versions

In OctoClick's version, Daniel starts with 2 popunder campaigns, Android and iOS. He "allowed each campaign to spend up to $5 before pausing it for optimization" and set "an hourly budget cap of $1 for each campaign", letting them "run for several hours to collect enough data for analysis." Juddy's Russian version has the same $1 cap.

Daniel's own version sets the cap at $3 an hour, "to stretch the spend across different hours of the day and capture a well-rounded dataset."

Our math: $5 at $3 an hour runs out in 1 hour 40 minutes, which isn't "different hours of the day". At $1 an hour it lasts 5 hours, which matches "several hours". The test itself suits this offer: at the case's average $1.51 CPM, $5 buys about 3,300 pops. If they convert like the payer's average, 3.8 subscriptions per 100 unique visitors, that's roughly 125 subscriptions. That's campaign-level data, fast. Spread across dozens of site IDs, it's noise.

Optimization: 3 payouts to prove yourself, then a lower bid

"After the initial spend, I filtered out the site IDs that had spent more than [3] times the offer payout without generating profitable results and added them to the blacklist... I also noticed that some site IDs were generating conversions but were still running at a negative ROI. Instead of blacklisting them immediately, I lowered the bid and monitored their performance over time. This optimization process can take 2–3 days."

His tip, in Juddy's version: "Time of day and day of the week can significantly affect conversion. A site that shows weak results today can become profitable tomorrow, or even at a different time of the same day." And: "Don't make optimization decisions too quickly" (our translation).

Our math: the 3-payout rule is stronger than it looks for this offer. The posts don't give the payout; at the payer's average of about $0.08, 3 payouts is about $0.24, or about 160 pops at $1.51 CPM. At 3.8% the average site brings 6 subscriptions in that window, and the chance of 0 is about 0.2%. A site with 0 after 3 payouts is almost certainly dead.

Lowering the bid on sites that convert at a loss is the right call too. In pops you bid per site at EPM divided by (1 + target ROI); our max bid calculator does the arithmetic. A site that earns the case's average $3.55 per 1,000 pops can be bid up to $2.37 CPM and still return 50%. The catch: a lower bid wins fewer auctions, and the case shows no site-level numbers, so you can't see how much volume the cut sites lost.

The money: $486.48 in, $1,143.10 back

OctoClick advertiser report, April 21 to May 31, 2026: total spent $486.48 at a $1.51 CPM, EPM $3.55, ROI 134.97%, leads earned $1,143.10; campaign 246720 Kz Top $181.53 spent, ROI 122.73%; 203497 Kazakhstan juddy $153.80, ROI 239.8%; 586090 Kazakhstan juddy copy $61.76, ROI 25.49%; 635849 top kz $39.39, ROI -12.03%; further rows cut off.

Daniel's report has the totals to the cent, with revenue counted from postbacks. Juddy's July 16 version carries the same screen.

Our math: $1,143.10 minus $486.48 is $656.62, and $656.62 / $486.48 is 134.97%, which rounds to 135%. The headline's 134% drops the decimals, and Daniel's own post says 134.7%, a typo of the dashboard figure. The report URL encodes a window of 3,542,398 seconds, 41 days to within 2 seconds, so the dates match. It's 41 calendar days, not 40.

The text says 2 campaigns. The screenshot shows 4 named campaigns, including a copy, and more rows cut off below; we derived those from the totals.

Campaign

Spend

Revenue

Profit

203497 Kazakhstan juddy

Spent $153.80 at a $1.31 CPM

Revenue $522.61

Profit $368.81, ROI 239.8%

246720 Kz Top

Spent $181.53 at a $1.36 CPM

Revenue $404.31

Profit $222.78, ROI 122.7%

586090 Kazakhstan juddy (copy)

Spent $61.76 at a $2.15 CPM

Revenue $77.50

Profit $15.74, ROI 25.5%

635849 top kz

Spent $39.39 at a $2.27 CPM

Revenue $34.65

Loss $4.74, ROI -12.0%

Rows cut off, derived from the totals

Spent $50.00

Revenue $104.03

Profit $54.03, ROI 108%

Total

Spent $486.48 at a $1.51 CPM

Revenue $1,143.10

Profit $656.62, ROI 134.97%

Our math: 2 campaigns made 90% of the profit on 69% of the spend, and 203497 alone made 56%. They bought at $1.31 and $1.36 CPM and earned $4.46 and $3.03 per 1,000 pops. The copy and "top kz" paid $2.15 and $2.27 and earned $2.70 and $2.00. Copying a winner at a higher price didn't copy its results, and the case doesn't say why.

Over 41 days that's $11.87 of spend and $16 of profit a day: a test, not a scaled campaign.

The payer's screen shows 30% more revenue

Juddy.biz statistics by OS, April 21 to May 31, 2026, filtered to Kazakhstan: Android 243,924 hits, 235,850 hosts, 9,731 accepted conversions, 60,366.63 rubles, CR 4.126%; iOS 327,202 hits, 250,509 hosts, 8,619 conversions, 54,951.21 rubles, CR 3.441%; small rows for Windows, Mac and Linux.

Juddy's own screen for Kazakhstan over the same dates: 574,083 hits, 488,625 unique hosts, 18,435 accepted conversions and 115,824.84 rubles. It's in Juddy's July 16 version of the case.

Our math: $486.48 at a $1.51 CPM is about 322,000 impressions sold by OctoClick, and Juddy counted 1.8 times as many hits. Juddy's eCPM column implies 78.04 rubles to the dollar on both the Android and iOS rows, which puts the screen at $1,484: $341, or 30%, above the case's revenue. The screen is filtered by country only, not by operator, source or sub ID, and it includes 85 conversions on Windows, Mac and Linux, which the 2 mobile campaigns in the text shouldn't bring.

That leaves 2 explanations, and the case offers neither:

  • the Juddy account carried other Kazakh traffic in the same weeks, and this screenshot proves less than it seems;
  • about 23% of the revenue, $341 of $1,484, never came back to OctoClick by postback. Then the real ROI was about 205%, and the case undercounts its own result.

Our take: this is what "no tracker needed" costs. The case closes on it: "I did not use an external tracker for this campaign. All I needed was some help from my account manager to set up the postback." With postback-only stats inside the traffic source, you can't tell a postback gap from mixed traffic. OctoClick's own Telegram, in the post right after the case: "We always recommend launching campaigns together with a tracker" (our translation). That post is right: run a tracker.

Kazakhstan's Ministry of AI and Digital Development signed an order on June 18, 2026, in force from June 28: additional paid mobile services are connected only after 2-step consent, confirmed through the operator's app, SMS, a call or USSD. On July 21, vice minister Doszhan Musaliyev explained why: "Previously there were frequent cases when users accidentally clicked on an ad on the internet, after which a paid subscription was set up automatically" (our translation). He added that a new telecom law was due to take effect at the end of August, with more than 20 bylaws planned by the end of September.

That sentence describes the flow in this case. The run ended May 31, 4 weeks before the rule. Every version of the case came out after June 28, and none of those we could read mentions it.

How 1-click on Tele2 works now isn't public. Juddy's news channel announced new Tele2 Kazakhstan offers, adult and mainstream, on July 15. On September 11 it warned of "problems with 1-click offers, with subscriptions forming on the operator's side" (our translation). It posted similar warnings on April 15, 6 days before this run started, and on June 11.

As of November 2026 the offers are still listed: Juddy announced new mainstream Tele2 Kazakhstan offers on October 13. On November 2 it reported problems accepting traffic on KZ Tele2 and Kcell, and on November 4 it said the problems on the content provider's side were continuing and asked partners to adjust traffic until service was restored. By November 18 we found no all-clear for Kazakhstan in the channel.

Our math: at the case's averages, pops earned $3.55 per 1,000 against $1.51 of cost, 2.35 times break-even. If a 2nd confirmation step cuts conversions by more than 57%, this setup loses money at the same bids.

Ask before you test. Get the network to confirm in writing whether the Tele2 Kazakhstan offer is still 1-click or now carries a 2nd confirmation, and what it has converted at since June 28, 2026. A case from April and May can't answer that.

Fact-check: what the author showed and what they hid

We rate every case on 10 points: what an outsider needs to judge the result and repeat it. We scored the versions together. Like most case studies, this one is selectively true rather than fake.

Point

Status

What we found

Setup: period, GEO, offer, model

Partial

Dates, Kazakhstan, Tele2, Android and iOS and Juddy.biz are given. The offer and payout aren't, and the model is "CPL" in 2 versions and "CPA" in the author's.

Author and independence

Partial

Daniel Inyang is a real, findable blogger, but the case went public through OctoClick and Juddy.biz first. OctoClick pays for cases and says so under this one; whether this case was paid isn't disclosed.

Finances

Showed

Spend $486.48, revenue $1,143.10 and ROI 134.97% reconcile to the cent in OctoClick's report, with 4 campaigns itemized.

Proof

Partial

Real dashboards from both sides, but the payer's screen shows 30% more revenue than the traffic source, is filtered by country only and comes with no explanation.

Source and targeting

Partial

Pops on Tele2, separate Android and iOS campaigns, a $5 test, an hourly cap, the 3-payout blacklist rule and lower bids for losing sites. No bids or site-level numbers, and the cap is $1 or $3 depending on the version.

Tools

Showed

OctoClick's built-in stats and a postback set up by the account manager. No tracker, which is why the 2 dashboards can't be reconciled.

Creatives

Hid

Nothing shown.

Landing pages and flow

Hid

No carrier page, no pre-lander, no offer ID, not even whether the content was adult or mainstream.

Consumables

Showed

None needed: pops run from a funded account with a $20 minimum deposit, per Juddy's version, without ad accounts, proxies or cloaks.

Repeatability today

Partial

The bidding method works on any pop source. The flow doesn't come with a guarantee: Kazakhstan has required 2-step consent for paid mobile services since June 28, 2026, and Juddy.biz flagged problems on Tele2 Kazakhstan in September and November.

Score: 3/5. The traffic-side money is shown in a real dashboard that reconciles to the cent, with campaign IDs and CPMs, and the payer's dashboard sits next to it, which is more than most cases give. It doesn't go higher because the offer, payout and flow are hidden, the payer's screen shows 30% more revenue than the traffic source without a word of explanation, the details change between versions, and the case comes from the seller of the traffic, which pays for cases. Solid data, vendor packaging: a 3, not a 4.

Checklist: what to do

Questions and answers

What is mVAS in affiliate marketing?

Mobile value-added services: content subscriptions such as games, video or music billed to the phone balance. In a WAP-click flow the carrier recognizes the number on mobile data and 1 tap subscribes the user, with no card. Affiliates are paid per subscription or per charge.

Do popunders still work for mVAS in 2026?

They can. In this case, pops on Tele2 Kazakhstan returned 135% on $486 at an average $1.51 CPM, earning $3.55 per 1,000 pops. Kazakhstan's 2-step consent rule since June 28, 2026 may change the conversion rate.

What changed for mobile subscriptions in Kazakhstan in 2026?

From June 28, 2026, operators may connect additional paid services only after 2-step consent through the operator's app, SMS, a call or USSD. The vice minister cited users who clicked an ad and got a paid subscription automatically.

How do you optimize popunder campaigns by site ID?

Track EPM per site, bid each site below its EPM and blacklist sites that spend several payouts without converting. In this case, sites that spent 3 payouts without profit went to the blacklist, and sites that converted at a loss got a lower bid instead.

Should you trust a case study published by a traffic source?

Trust the dashboards, check the framing. A network picks its best result, may pay for the write-up, and never shows the losing tests. Compare the network's numbers with the payer's for the same dates, and ask whether the case was paid.

Author’s conclusion

This is a real campaign, told by the people who sold the traffic. The numbers on OctoClick's side are as clean as cases get: campaign IDs, CPMs and totals that reconcile to the cent. The rest is the usual vendor-case shape: the best result picked and polished, an author who doesn't mention that the network published it first and pays for cases, and a payer's dashboard with 30% more revenue and no explanation. The bidding lesson holds. The flow now needs a 2nd confirmation in Kazakhstan.

My advice: take the method and leave the conclusion. Bid per site against EPM, kill sites that burn 3 payouts with 0 conversions, split by OS and give sites time across hours and days before you cut them. Run a tracker from day 1, so you can line up the traffic source's numbers against the payer's. And before you put money into 1-click mVAS in Kazakhstan, get the network to confirm in writing whether the flow is still 1-click and what it converts at after June 28.

Lu Discover, Editor-in-chief

Our verdict on this case

3/5

The traffic-side money is shown in a real dashboard that reconciles to the cent, with campaign IDs and CPMs, and the payer's dashboard sits next to it, which is more than most cases give. It doesn't go higher because the offer, payout and flow are hidden, the payer's screen shows 30% more revenue than the traffic source without a word of explanation, the details change between versions, and the case comes from the seller of the traffic, which pays for cases. Solid data, vendor packaging: a 3, not a 4.

Our fact-check of the public case: what the numbers show, what is missing. Scores are never for sale.

Sources for this article

  1. Daniel Inyang — Case Study: 134% ROI on an MVAS offer in Kazakhstan using Popunder traffic from Octoclick, his own post, partly members-only (July 22, 2026)
  2. OctoClick — the case in its own sales thread, "Over to Daniel," with a promo code (July 15, 2026)
  3. telegra.ph — KZ Tele2: breaking down a campaign with ROI 134%, with the OctoClick and Juddy.biz screens (July 16, 2026)
  4. OctoClick — CASE STUDY: ROI 134% on MVAS in Kazakhstan, Telegram post, "ready to credit authorship and pay for the material" (July 23, 2026)
  5. OctoClick — paid case program: "we evaluate the case and pay for it" (August 8, 2025)
  6. Ministry of AI and Digital Development of Kazakhstan — order on 2-step confirmation of paid services, signed June 18, in force June 28, 2026
  7. Doszhan Musaliyev, vice minister of AI and digital development of Kazakhstan — briefing on 2-step consent for paid mobile services (July 21, 2026)
  8. Daniel Inyang — forum profile: registered September 30, 2025, paid membership and first message on July 21, 2026, follows only OctoClick
  9. Daniel Inyang — activity log with OctoClick's replies in the thread (July 21–23, 2026)
  10. Daniel Inyang — blog About text, archived January 21, 2025
  11. OctoClick — "Our platform specializes in traffic from adult tube websites" (July 15, 2026)
  12. OctoClick — "we always recommend launching campaigns together with a tracker," with a PeerClick promo (July 27, 2026)
  13. t.me — announcement of the case in its news channel (July 17, 2026)
  14. t.me — new Tele2 Kazakhstan offers, adult and mainstream (July 15, 2026)
  15. t.me — KZ Tele2 1-click notice on subscriber number detection (April 15, 2026)
  16. t.me — KZ Tele2 1-click technical problems notice (June 11, 2026)
  17. t.me — KZ Tele2 notice on 1-click problems (September 11, 2026)
  18. t.me — new mainstream Tele2 Kazakhstan offers (October 13, 2026)
  19. t.me — problems on the content provider's side for KZ Tele2 and Kcell continue (November 4, 2026)
  20. juddy.biz — program information: organizer InternetWeb LLC, Voronezh
  21. juddy.biz — user agreement: partner reward and prohibited traffic
Would you recommend this article?
We try to stay objective, but we can make mistakes too.

Report a mistake

We try to stay objective, but we can make mistakes too. Tell us what is wrong and we will check it.

Page
Facts, numbers, a source if you have one.
Only if you want an answer.
Share

Comments

0 comments

No comments yet.

Add a comment

Links are not allowed.
Or sign in to comment under your profile.