Yana Lyashenko
Google Ads AI Architect · Founder of ADWService
I build Google Ads systems for e-commerce businesses, where each campaign isn’t just a set of settings, but a component of an architecture designed for profitable scaling.
Since 2014, I have been working exclusively with Google Ads for online stores. My focus is on Performance Max and Google Shopping: not just launching campaigns, but building an architecture where Google AI learns to sell your specific products.
I developed my own Bucket Architecture™—a system for structuring PMax campaigns through 4 levels of AI training: conversion labeling, global pre-training, local fine-tuning, and task decomposition by product segments. This isn’t from Google’s documentation—it’s a practical system developed based on hundreds of real accounts.
12+
years at Google Ads
316
active MCC accounts
34.7K
YouTube subscribers
$84M+
customer conversions per year
Performance Max
Google Shopping
Merchant Center
Feed optimization
Bucket Architecture™
Google Ads AI
e-Commerce scaling
Search advertising
YouTube ads
GA4 analytics
ROAS management
Specialization
What I do better than most
A narrow focus on Google Ads for e-commerce—not spreading yourself too thin across all channels.
Google Ads Architecture for E-Commerce
I’m building an account structure where each campaign plays a specific role—donors, cash cows, scaling, and new markets. It’s not just a collection of campaigns, but a system.
Training Google AI using Bucket Architecture™
PMax’s proprietary 4-step structuring methodology: conversion labeling → pretraining → account fine-tuning → task decomposition by product.
Google Shopping and Performance Max
Specializing since 2014. Feed setup, campaign structuring by category, and targeting based on margin and conversion volume.
Merchant Center and feed optimization
Unblocking, feed audits, and attribute optimization for Shopping. Over 300 successful Merchant Center unblocking cases.
Scaling ROAS without increasing CPA
Performance based on business metrics: profitability, seasonality, and demand structure by SKU. ROAS as a tool, not just a goal.
International e-Commerce
Launching and managing Google Ads campaigns in the US, UK, EU, AU, and CA markets. Specifics for each market: product feeds, competition, target audiences, and budgeting.
Original methodology
Bucket Architecture™ - how it works
This is not official Google documentation. It is a practical guide on how to train Google AI to sell your specific products.
L1
Conversion tracking
Google AI evaluates only two criteria: whether a conversion occurs and what its value is. Assigning the correct labels is the foundation of everything.
L2
Global pretraining
Google’s massive algorithm processes billions of auctions. We can’t change it—but we can optimize it effectively through signals and structure.
L3
Local fine-tuning
Learning period + continuous online learning. Every change to your account is a new prompt for your store’s AI.
L4
Bucket campaign structure
Task decomposition: donors, cows, nursery, scaling by geography and margin—each bucket as a separate fine-tuning model.
Why is this important for the customer to know?
Most agencies launch Performance Max and wait for results. But PMax isn’t a black box. It’s an AI model that needs to be trained properly.
When we build the architecture, we’re essentially teaching Google AI to understand your business: which products are profitable, which audience converts, and where to allocate your budget.
The result: controlled scaling instead of chaotic budget drain.
Case Study • Home Goods Niche
How Bucket Architecture™ boosted sales from 300 to 740 a day
Watch the video: how we turned Google Shopping’s seasonal slump into steady growth—without increasing the budget.
How can you boost sales during the off-season? A home goods store was stuck at 300 sales per day and couldn’t seem to make any progress.
We implemented Bucket Architecture™ in Google Shopping / Performance Max. We divided the product range into buckets: donors, cash cows, and nursery—each with its own fine-tuning.
300 → 740 sales per day. A 2.4-fold increase without increasing the advertising budget.
01
Owners of online stores with a product range of 100+ SKUs
02
For marketers and PPC specialists working in e-commerce
03
For those launching Performance Max or Google Shopping for the first time
My YouTube channel • 36,200 subscribers
Explaining Google Ads for e-Commerce in plain English
36.2K
community of subscribers
1 500+
a video from 2016
4.0M
views
- How to Train Google AI in Performance Max
- The Bucket System for Google Shopping
- Merchant Center and Feed Setup
- ROAS: How to Scale Without Increasing CPA
- Demand Gen as a PMax Booster
- Google Ads for Shopify, Prom.ua, Horoshop, WooCommerce
Confirmation
Regalia and recognition
Ringostat TOP-30 · 2025
We ranked 20th, considering that we are a PPC-focused agency, among Ukraine’s digital agencies that work across all channels. There were 494 participants and 42 experts on the jury.
Clutch · 5.0/5.0
Verified reviews. Quality, Schedule, Cost — all 5.0.
Merchant Center Unlock Course
• 21 modules • PDF + video
The first course in Eastern Europe. Over 300 online stores have resumed operations
YouTube channel
Community of 36,200 subscribers. ADWService | Google Ads & E-Commerce
TOP eCommerce Marketing Agency
Ukraine 2026
TOP Display Advertising Company
Ukraine 2026
MCC Account 2025
316 accounts
active e-commerce sites under management
Conversion Value
$84M+
customer revenue/year
Markets
10+
(UA · US · UK · EU · AU · CA)
* Google Ads MCC, 2025. Screenshot upon request.
For existing Shopify 2.0 merchants
Ukrposhta and Payoneer, together with their partners, have created a training workshop. The workshop is organized by Ukrposhta in collaboration with Payoneer as part of the E Export School initiative.
The topic of Yana Lyashenko’s presentation “Tools for Promoting Products on Shopify Using Google Ads”
Shall we talk
about your Google Ads?
Consultation or sprint —
depends on
where you are now and where you want to go.
Or Telegram @Adwservice_bot · We’ll get back to you within 2 hours
RESULTS • ADWService
Scale of the practice (Google Ads MCC data for 2025)
Advertising expenses: UAH 449,250,059.71 for 2025 (≈ UAH 37.44 million/month)
Conversions recorded: 1,751,316.87 (≈ 145,943/month)
Average metrics: CTR 1.83%, Avg CPC UAH 8.22, Conversion rate 1.57%
Google Ads MCC data · 2025 · includes accounts from Ukraine, the US, the EU, Australia, and Canada
Note: Conversion value and Conversions depend on the tracking and attribution settings in your accounts (attribution model, conversion import, event duplication, etc.).
