I Moved From AdSense Toward Google AdX Through RevBros–LKZ: My Live Publisher Diary

Fast answer: what is live right now?

My original Google AdSense setup is still serving ads. I accepted a genuine Google Ad Manager Manage Account relationship with RevBros–LKZ, the child AdX account was enabled, and a second Google seller line was added to ads.txt. But no Google Publisher Tag (GPT), new AdX runtime, or LKZ code was live at this Day-0 check. Therefore I cannot yet say that AdX increased—or reduced—my income.

This is a live investigation, not a recommendation

I am publishing the setup stage because I could find little substantive independent publisher feedback about LKZ/RevBros. I will add dated replies, code changes, technical results, net performance, support experience, and payment evidence as they occur. Earlier rows will stay visible even if the result is disappointing.

Privacy and evidence boundary

The WhatsApp conversation is paraphrased. I am not publishing the phone number, email address, GAM network code, account screens, payment details, click data, CTR, CPC, or the actual commercial revenue-share percentage. The public seller record is not repeated because readers do not need it to understand the test.

How a local representative contacted me before ChatGPT surfaced LKZ

A local monetization professional first contacted me on Friday, August 28, 2026. That contact happened before I later asked ChatGPT for publisher-monetization options. ChatGPT surfaced lkz.ai as a research lead; it did not introduce us. The same representative sent the GPT/Zenith and LKZ Prebid scripts or wrappers installed for this test and created the support group that same day. I treated the AI answer as a lead to investigate—not as vetting, certification, an OpenAI partnership, or any revenue guarantee.

I applied because the company has a real website and the process moved into my genuine Google Ad Manager account. I had also just completed My completed AdSense-to-Journey test, so I wanted any next test to begin with a public baseline, a rollback plan, and a record of what was promised before code changed.

This diary begins at the most useful point for another publisher: after account authorization but before ad-serving implementation. That makes it possible to separate an invitation, an ads.txt change, and an actual revenue test instead of blending them into one success story.

What RevBros–LKZ told me on WhatsApp

On August 28, the representative reviewed my traffic-country screenshots and recommended Manage Account (MA). The table below paraphrases the conversation. These are company statements, not measured outcomes.

The message that created the most confusion was that AdSense could continue alongside AdX, followed later by a plan to wire GPT as the replacement for default AdSense tags in relevant places. Both ideas can exist in a carefully designed Ad Manager setup, but they do not tell me which of my current placements will remain, become backfill, or be replaced. I need that written map before approving a cutover.

Company statements versus current evidence

MA, MI, AdX, GPT and ads.txt—in plain language

Google’s documentation is the controlling source for product mechanics. In a Manage Account relationship, the parent can manage inventory in the child network, with access except to billing. The child uses its own Ad Manager network, tags and child Ad Exchange account, while an agreed Primary Ad Exchange allocation can be paid automatically. Reporting remains available in the child network.

Manage Inventory (MI) is different: the parent manages delegated inventory in its own network and generally controls that serving setup. I did not choose MI, so I am not presenting the representative’s payment-method description as a tested fact for this site.

Google Publisher Tag is a JavaScript ad-tag library for Google Ad Manager. It defines slots and sends ad requests; it is not a button that universally converts every AdSense placement. Google also documents ways for AdSense to act as demand in Ad Manager. That is different from assuming the existing Auto Ads and manual adsbygoogle code will keep running unchanged.

An ads.txt line tells buyers which seller account is authorized to sell inventory. It does not install an ad tag, start an auction, disable AdSense, or prove that an AdX test is serving. Those require separate code and account configuration.

What each term changes

What I independently verified on Day 0

The parent shown inside my signed-in Google Ad Manager is RevBros–LKZ, and the delegation is Manage Account. I accepted the invitation, the child Ad Exchange account was enabled, and GAM shows an approved auto-payment allocation. The exact commercial percentage is confidential and is not needed to evaluate the method.

The requested Google seller line is live in the public root ads.txt beside the existing AdSense seller line. A fresh seller entry can take time to be crawled. Its presence proves authorization only.

A same-day public code inspection found the original AdSense publisher configuration and adsbygoogle placements still active. It found no GPT loader, no new AdX publisher identifier in the runtime page source, and no LKZ runtime code. In other words, the account relationship is real, but the performance test has not started.

What I found—and did not find—about LKZ online

The operational footprint is stronger than a random cold message. The WhatsApp number matched LKZ’s contact page; Lkz Systems LLC appears in a Wyoming filing list; Lkz Media Limited appears in a Hong Kong Companies Registry list; and the genuine in-product GAM invitation links the parent relationship to my account.

What I did not find was a meaningful body of independent publisher reviews, payout histories, or long-form implementation reports on the major review and publisher-community surfaces I checked. A SmartCustomer profile displayed zero reviews at the time of checking. That absence is not evidence of fraud and is not evidence of strong performance—it simply leaves more work for this live test.

LKZ’s own website currently promotes a “no lock-in” message and reports performance/publisher totals as company marketing. Those figures were not independently audited for this article. Its terms say services do not guarantee demand, fill, revenue, or results; publisher payment/share details may be account-specific; the publisher generally gives 30 days’ termination notice while the company may terminate without notice; publisher privacy/consent duties remain; default governing law is Hong Kong; and stated liability is capped at USD 500. Google separately says either MCM party can end the Google relationship at any time. I have asked how those layers work together.

I also did not find LKZ, RevBros, or the parent network name in Google’s public Certified Publishing Partner directory. An MCM relationship inside GAM is not the same thing as Certified Publishing Partner status, and I am not implying either Google or OpenAI endorses the company’s commercial performance.

Where I am stuck before any tag change

I accepted the invitation before I fully understood MA, MI and how GPT would alter my live AdSense stack. I am now waiting for the promised team follow-up and an explanation in a local language. No additional implementation approval should be inferred from the account invitation or ads.txt update.

The unanswered questions below are my deployment gate. I want written answers, a limited first test, complete reporting access, and a tested restoration path before a partner changes monetization code across a multilingual site.

Correction—August 30, 2026: A local monetization representative first contacted me on Friday, August 28, 2026—not August 26. The same representative sent the GPT/Zenith and LKZ Prebid scripts or wrappers installed for this test and created the support group that same day. The group exists; response time, technical ownership, local-language support quality and escalation reliability are still being evaluated. This correction supersedes the earlier August 26 contact row and pending-group wording below while preserving them as part of the diary history.

The break-even math after a partner share

A higher gross CPM does not automatically mean higher publisher income. The relevant comparison is net revenue after the confidential partner allocation and any additional fees. The generic break-even formula is required gross multiplier = 1 / (1 - partner share).

For a deliberately unrelated example, at a hypothetical 10% partner share, gross revenue must rise by about 11.1% merely to preserve the same net amount: 1 / 0.90 = 1.111…. This is an educational example, not KBT’s agreement, and it cannot be used to infer KBT’s private percentage.

Any header-bidding revenue outside the child Primary Ad Exchange auto-payment needs its own label and payment route. I will not combine AdSense, AdX gross, partner deductions, and third-party demand into one flattering number.

My fixed measurement and rollback plan

I am publishing the rules before results so the goalposts cannot move. Immediately before the first GPT-serving date, I will preserve the last 14 and 28 complete AdSense days in one timezone. Partial launch or rollback dates will not be treated as full comparison days.

The primary business metric will be net page RPM after the confidential share and any other fees. I will use independent GA4 sessions and pageviews as traffic controls, segment country/device/top-page mix, and separately track impressions, coverage/fill, viewability, ad density, Core Web Vitals/latency, policy warnings, and reader complaints.

Every tag, placement, pricing rule, refresh rule and demand-source change will receive a timestamp. Day 7 is a technical-health checkpoint, Day 14 is directional only, and Day 30 is the first serious performance decision. Day 60 and 90 updates happen only if the test remains live.

Before cutover, I will archive the exact working AdSense stack and prepare a narrow rollback. If net results, reliability or user experience materially worsen, I will restore that known-good version rather than extend losses just to reach an arbitrary date. This follows the reversible-change discipline documented in A reversible production-change method.

Dated, append-only update log

This table will grow when the company replies, when code changes, when ads begin serving, and when enough complete days exist for a fair comparison. Earlier entries will not be silently rewritten. Each future result will state the evidence source, traffic context, and next gate.

Attribution note: The wrapper installation, ads.txt expansion, CSP repair and AdSense In-page-format change all happened on August 28. Because several monetization variables changed on the same day—and Google warned that its setting could take up to one hour to propagate—I will not credit or blame LKZ/AdX for an immediate revenue movement. A fair conclusion requires complete comparison periods and separate AdSense, AdX gross, partner deductions and traffic controls.

Day 1 checkpoint: the first real revenue comparison

On August 29, at about 9:46 p.m. Pakistan time, I read both signed-in dashboards after the wrapper had been live throughout nearly the entire August 29 calendar day. These are same-day estimated figures, not midnight-final or settled earnings. The two systems served separate inventory, so impression counts are context—not a like-for-like A/B test.

Disclosure update—August 29, 2026: for this measured checkpoint I am disclosing the fixed 20% Primary Ad Exchange auto-payment allocation so the net publisher comparison can be audited. The earlier Day-0 privacy boundary remains part of this diary’s history; account identifiers, private dashboard screenshots, click data, CTR, CPC and payment details remain private.

Google Ad Manager reported $1.75 before the MCM auto-payment allocation: 20% ($0.35) to the parent and 80% ($1.40) estimated for this child account. Direct AdSense added $0.78, making the estimated amount attributable to the site owner $2.18. Simply adding $1.75 and $0.78 would overstate owner revenue.

The directional efficiency signal favored GAM: its reported eCPM was $0.82, or about $0.656 after the 20% allocation, versus $0.42 AdSense impression RPM. GAM viewability was 64.7% versus 54.75% in AdSense. However, GAM also showed about 1,190 unfilled opportunities; a simple served-versus-unfilled proxy is 64.4% filled and 35.6% unfilled.

The verified recent AdSense high was $2.36 on August 26 ($2.31 was August 27), so the $2.18 combined owner estimate was $0.18 below that benchmark at the checkpoint. Turning off AdSense without replacement would have reduced the observed owner total by $0.78, or 35.8%. GAM alone would have needed $2.725 reported before MCM to match the hybrid $2.18, and $2.95 to match $2.36.

My Day 1 decision: keep GAM and AdSense running together and make no mid-day switch. A hypothetical transfer of all current AdSense opportunities into GAM at the same fill and price only roughly breaks even, and that assumption may fail when inventory, demand, placements and viewability change. One near-full day is encouraging evidence for GAM efficiency, not proof that GAM-only earns more.

Directional dashboard context

Configuration phase B: AdSense controls changed again

On August 30, 2026 at 10:46 p.m. Pakistan time, I confirmed the final saved AdSense site settings after changes requested by the representative. Auto Ads remained enabled and Intent-driven formats stayed active. Auto optimize was turned off, every Overlay format was turned off, and Banner ads was re-enabled while Multiplex remained off.

Measurement consequence: this ends the unchanged configuration window described in the August 29 checkpoint. I will preserve that checkpoint, segment later results as a new configuration phase, and will not credit or blame GAM/GPT alone for changes in revenue, fill, viewability, page speed or layout. Directional comparison requires complete days with this exact setting unchanged.

Journey status

Already using LKZ, RevBros or another MCM partner?

Please use the moderated English comments below to share first-hand experience. Helpful context includes MA or MI, site category, broad traffic-country mix, test length, broad revenue-share range if you are free to disclose it, net RPM before and after, fill/viewability, payment experience, and support quality.

Do not post account IDs, emails, phone numbers, payment details, private screenshots, contract-confidential terms, clicks, CTR or CPC. Reader comments remain personal reports—not verified facts—unless I can independently check the evidence. I will compare credible reports with my own dated results and update this article as the journey continues.

Related first-hand KBT reports

Sources, review method and disclosure

Google documentation supports product mechanics; company pages and terms support statements about LKZ; filings support entity footprint; community threads are used only as reader-question signals, not as proof of KBT results. Searches were performed on August 28, 2026.

• Google: MCM account setup options
• Google: About Multiple Customer Management
• Google: Manage Account delegation
• Google: Manage Account auto-payment FAQ
• Google: End an MCM relationship
• Google: Create and understand ads.txt
• Google: Use AdSense in Ad Manager
• Google: Get started with Google Publisher Tag
• Google: Compare Ad Manager, AdSense and AdMob
• Google: Certified Publishing Partner directory
• LKZ contact page
• LKZ terms
• Wyoming domestic entity filing list
• Hong Kong Companies Registry list
• SmartCustomer: LKZ review profile
• Independent publisher discussion: can AdX/MCM improve revenue?
• First-hand account evidence: signed-in Google Ad Manager and AdSense report readback on August 29, 2026 at about 9:46 p.m. Pakistan time.
• First-hand settings evidence: owner-confirmed signed-in AdSense site-settings readback on August 30, 2026 at 10:46 p.m. Pakistan time.
• Kaboompics / Pexels · Photo 6334150
• Kaboompics / Pexels · Photo 6334146
• Kaboompics / Pexels · Photo 6334193
• Kaboompics / Pexels · Photo 6334203

Editorial disclosure

KeyboardTester.click received no payment, sponsorship, affiliate commission, free product, or other compensation from LKZ or RevBros for this article. The commercial relationship described is a revenue-share MCM arrangement under test; its exact terms remain private. The people in the photographs are illustrative and are not presented as customers or endorsers. If compensation or the relationship changes, I will add a dated disclosure update.

Frequently asked questions

• Will my Google AdSense still work after accepting an MCM invitation?

  At Day 0 on this site, yes: the existing AdSense runtime is still active. The invitation and ads.txt line did not disable it. What remains after GPT implementation depends on the written placement and demand configuration, so do not assume every current tag will stay unchanged.

• Does adding an AdX ads.txt line replace AdSense ads?

  No. ads.txt authorizes a seller account; it does not install GPT, create ad units, change page code, or disable the existing AdSense seller line and tags.

• Will Google AdX increase my AdSense income?

  It may change monetization, but there is no guaranteed increase. Compare net page RPM after partner share and other fees over complete, controlled windows—not a gross CPM claim or one unusually strong day.

• Is LKZ or RevBros legitimate?

  I verified an operational footprint and a genuine RevBros–LKZ Manage Account relationship inside Google Ad Manager. I found little substantive independent publisher review or payout evidence, and the terms disclaim revenue results. My current conclusion is operational relationship verified, performance and support not yet proven.

• What is the difference between MCM Manage Account and Manage Inventory?

  Manage Account gives the parent access to manage the child’s Ad Manager account except billing, while the child uses its own network, tags and child AdX account. Manage Inventory delegates inventory through the parent’s network. This site selected Manage Account.

• Should I accept a Google AdX MCM partner invitation?

  Verify the parent identity in Google Ad Manager, read the account-specific agreement, understand access and payment routing, get fees and tag changes in writing, preserve a baseline, start with a limited test, and prepare rollback. An invitation alone is not proof that the partner will raise net revenue.

Follow the evidence, not the promise. Bookmark this page for the company reply, implementation status, Day 7/14/30 checks and any payment update. If you already use an MCM partner, share a privacy-safe first-hand report in the moderated comments so other publishers can compare real experiences.