Ruslan Galba AI x Google Ads

18 July 2026 16:01

AI got 14,430 of my dictated messages this year. My coworkers got 58.

Screenshot attached, my dictation app keeps receipts. 1,455,560 words spoken at 158 a minute. The ratio looks like bragging until you open it up, so here's what those 14,430 messages actually said, and why the voice part is what nobody copies. Bookmark this and feed it to your AI agent.

What a work order sounds like in the ad accounts we run (20+ brands, three of us):

1. Feeds: "pull last month's search terms for the top 40 SKUs and rewrite every title that doesn't contain what buyers actually type." The agent returns a rewrite table. I review diffs, not drafts.
2. Builds: "campaign skeleton for the spring launch, same structure as the one that held 2.2x, negatives preloaded from the master list."
3. Creative: "20 headline variants against the returns-eat-your-ROAS angle. Keep the claim, rotate the tension."
4. Teardowns: "every landing page this competitor runs, and what offer sits above the fold on each."
5. Anomalies: "why did Tuesday's spend spike. Check budgets first, then auction, then feed."

None of that gets typed. I call it the eye-voice split: eyes review, voice dispatches. Hands are for espresso.

The dispatch rules, since this is the part worth saving:

• Outcomes, not steps. I say "find where the money leaked last week," not ten micro-instructions. Agents are excellent at HOW and starving for WHAT.

• Ramble on purpose. I dictate the whole messy client-call context straight into the agent. Rambling is bad writing and elite context, the model compresses better than I summarize.

• Fire at 60%, steer mid-flight. "Narrower. US only. Ignore brand terms." Voice makes iteration nearly free, typing made me precious about prompts.

• One lane per agent, all lanes at once. Feed agent, creative agent, audit agent, running parallel like air traffic control, none of them waits for another to land.

• Review is the job now. Everything returns as a diff or a draft, so the human hour moved to the end of the pipe. I budget it like ad spend, because unreviewed volume is just expensive noise.

The compounding is the uncomfortable part. Speaking is 3x typing speed, and the work between prompts is free. One operator dispatching work orders ships more ad iterations before lunch than a typing team ships in a week, and volume finds winners in ads whether anyone likes it or not. The keyboard was the moat, and it drained.

Founder check: ask whoever runs your marketing what share of their day is instructions to software that does the work, versus meetings about the work. That ratio is the roadmap, and it's checkable today.

Ruslan Galba AI x Google Ads

16 July 2026 19:44

Marketers keep posting the same thing: long education pages are outperforming product pages in skeptical categories.

Beauty first among them.

The discovery is real. The explanation usually stops at "it works." Here's the mechanism, because you need it to build one that survives review.

A standard PDP assumes trust and asks for money. But the beauty buyer has been burned by a decade of miracle serums - her default read of your product photo and five stars is "another one."

The education page works because it spends its first half earning the belief the PDP takes for granted.

The structure, five sections in order:

1. The problem, explained at the mechanism level. Not "dull skin" - what's actually happening and why it resists what she's tried.

2. Why the usual fixes disappoint. This is the section that buys trust, because it explains her own failed purchases back to her better than she could.

3. Your ingredient logic as evidence, not adjectives. Concentrations, the reason for each active, what the formulation deliberately leaves out.

4. The offer, arriving after belief instead of before it.

5. Objections, answered plainly - sensitivity, timeline to results, what it won't do.

Section five deserves special respect in this category: every sentence about results is an efficacy claim, and "visibly firmer in 14 days" needs a substantiation file behind it, not a copywriter's confidence.

The education format tempts you to say more. Regulators read the whole page.

Run the arithmetic on your own funnel: count how many words a first-time visitor gets between your ad and your buy button. Under 200, and you're asking a skeptic for trust you haven't built.

The limit, stated up front so you can plan for it: these pages take real work - research, drafting, a legal read - and they lose to the PDP for buyers who already know your brand.

Route cold traffic through education and warm traffic straight to product. Building one page for both audiences is how you get neither.

Ruslan Galba AI x Google Ads

16 July 2026 03:23

I ran the same cold creative on Meta Reels and YouTube Shorts. The CPM gap was 2x.

Most Shopify brands still run Shorts as a brand play.

The numbers say it prospects.

Same product, same audience signal. Here is what came back.

YouTube Shorts:

• 2.1% CTR
• $4.80 CPM
• 30+ seconds watched before the click

Meta Reels:

• 1.4% CTR
• $11.20 CPM
• 3 seconds of passive scroll

12B+ daily views on Shorts, CPMs 60-70% below Meta Reels.

The audience skews buyer, not browser.

Meta is the default cold channel for most DTC brands, so the comparison rarely gets run.

Demand Gen on YouTube reaches buyers who haven't searched for the brand, on the platform where they're already watching.

That's targeted interruption you can measure down to the purchase.

The catch, so nobody reads this as free money: Shorts burns creative faster than Meta does. The CPM edge pays for the extra production volume, or it doesn't pay at all.

If an agency runs your ads and has never shown you a Meta-vs-Shorts CPM test on your own creative, that's a one-line email to send today.

For this brand, cold traffic moved to YouTube.

Ruslan Galba AI x Google Ads

15 July 2026 19:00

YouTube Ads Manager said 0.8x ROAS. The incrementality test said 2.4x.

The platform wasn't lying on purpose. It was answering the wrong question.

Last-click attribution gives credit to the final touch before purchase.

So here's the journey it never sees.

A customer watches your YouTube ad, doesn't click, googles your brand a day later, and buys through branded Search.

Search takes 100% of the credit. YouTube gets zero.

The demand was manufactured upstream. The accounting books it downstream.

I learned this the expensive way.

Pulled a Shopify brand builder's YouTube budget because the platform said it was dead weight at 0.8x.

Two weeks later branded search volume dropped 40%, retargeting pools dried up, and direct traffic tanked.

YouTube wasn't the closer.

It was starting 60-70% of the journeys that the other channels were finishing.

Killing it didn't save money. It quietly defunded every channel downstream.

Relaunched with proper measurement. True contribution came back at 2.4x.

There are three levels of attribution, and most brands only ever see the first:

• Level 1, platform last-click, what Ads Manager reports, shows 0.8x, where the panic and the budget cuts happen
• Level 2, blended, Northbeam or Triple Whale folding in branded search lift and direct, shows 2.4x
• Level 3, incrementality, geo-holdouts and lift studies, the only number that answers what happens if you turn it off

Across the tests I've seen, YouTube drives roughly 3.4x more conversions than the platform reports.

The gap between Level 1 and Level 3 is where every bad budget decision lives.

A geo-holdout costs a few hundred a month to run.

The misallocation it catches is measured in tens of thousands.

I'd rather spend $500 to find out the truth than cut $50K of upstream demand on a number that was structurally wrong before I ever read it.

The brands making budget calls on Level 1 think they're being cautious.

They're optimizing a P&L against fiction and calling it discipline.

I check Level 2 before I touch any awareness channel's budget now, because the cut I almost made on that brand would have been the most expensive thing I did all quarter.

Ruslan Galba AI x Google Ads

15 July 2026 12:57

Grüns - a gummy brand sold for $1.2B a few months ago.

And every food founder I know took exactly the wrong note.

The wrong note: "consumables are hot, we're a consumable, we're next."

The honest note: supplements are winning because of unit economics your snack brand does not have. 70-80% gross margins, tiny dim-weight shipping, subscription behavior the customer WANTS because running out feels like a health failure.

Your hot sauce runs 45-55% gross before shipping a glass bottle. Running out of hot sauce feels like a Tuesday.

Different physics. Copying the playbook without the margins is how food brands buy growth that eats them.

So here's the split - what actually transfers and what doesn't.

Copyable:

• The hero-SKU discipline. Every big supplement exit is one product with a routine attached, not a catalog. Food brands hoard SKUs like a pantry before a storm - cut to the one thing people reorder.

• The subscription OFFER, not the subscription assumption. Coffee earns it naturally. Snacks earn it with a bundle cadence ("the monthly box"), not a checkbox at checkout.

• The 90-day scoreboard. Supplement operators live on cohort value because first orders lose money. Food first orders lose money too - most founders just haven't done the math that proves it.

Not copyable:

• The margin that forgives mistakes. North of 70% gross you can misprice CAC for a quarter and live. At half that, the same mistake is the business.

• The health-anxiety reorder loop. Nobody panic-reorders chocolate.

Which means the food version of the playbook is stricter, not looser: tighter shipping breakpoints, bundles engineered to clear the free-shipping line, and a second order you design for instead of pray for.

Before you copy anyone's playbook, run the one-line check: gross margin after fulfillment, per unit, on your best seller.

If it starts with a 4, you're playing the harder game.

Ruslan Galba AI x Google Ads

14 July 2026 01:45

Everyone on X is asking "which AI video model is best."

Wrong question. It's why your ad account is bleeding.

I run YouTube ads for ecommerce brands. Last night we shipped dozens of video ads with 60+ shots, 5 hook variants each, through a full agentic AI pipeline.

For less than a stock footage subscription.

There is no best model.
There are best models per SHOT.

We route models per shot class, not per video:

1. Trust shots get the premium model.

The 5% of seconds that carry the proof: the water beading, the before/after match cut, the texture close-up.

One physics glitch there kills the whole ad, because viewers stare at those frames hunting the AI tell. Kling 3.0, $0.07/sec, no debate.

Maybe $6 of a whole production. Never save money here.

2. B-roll gets the billing arbitrage.

Some models bill per second, some bill PER VIDEO. Veo 3.1 Lite: about $0.175 for an 8-second 1080p clip.

That's $0.022/sec. Cheapest quality-per-second, because nobody does the division.

Workbench shots, driveway shots, scene-setting vignettes?

Nobody is pixel-peeping those.
Route them there.

We author b-roll beats at exactly 8 seconds to exploit the billing unit.

3. Talking heads only go to lipsync-capable models.

Most fake it. Two do it properly.

A silent mouth-flap is not a style, it's a broken file.

4. Sound costs extra. Pay it once.

Native audio is a +40% surcharge on some models. Pay it ONLY where the sound IS the appeal (the sizzle, the spray, the seal).

Everything else renders silent under one continuous voiceover.

Net effect: same perceived quality, half the spend, 5 hooks instead of 1, because hooks share the rendered body and only the opening clip changes.

A variant costs us $2-5. People pay editors $300 for what is a different first 8 seconds.


Now the part for people who don't generate video at all:

The highest-ROI step in our pipeline costs a dollar and needs zero rendering:

Before anything gets produced, the script gets torn apart by simulated buyers.

Not "does my team like it."

The skeptic burned by the last product. The spouse who controls the card. The 24-year-old who smells an ad in half a second.

Open-ended questions only:

- "retell this tomorrow, what survived?"
- "what did this assume about you that's wrong?"
- "at which second do you leave?"

Two rounds killed a line our whole team loved and rebuilt the entire opening.

Found after production, each of those is a full re-shoot. Found at script stage - a text edit.

You can do this today with any LLM and zero budget.

Most people won't.
It feels slower than creating.
It's the fastest thing we do.

Those "5 AI video prompts that go viral" threads? Recycled advice with the model name swapped. They age in days.

The questions underneath don't:

- Which shots carry the trust? Route premium.
- Which shots are wallpaper? Route cheap.
- What does the buyer retell tomorrow? That's your ad.

Stop asking which model is best. Roast the script before a single frame exists.

Ruslan Galba AI x Google Ads

02 July 2026 23:34

Google finally said the quiet part out loud: good SEO is good AI SEO.

For paid operators that moves where the work sits. Getting your product cited inside an AI answer is a feed and structured-data problem before it's a content problem. The model reads attributes, specs, and clean product data - the same Merchant Center discipline that wins Shopping.

So the feed you already maintain for Shopping is the asset that surfaces you when a shopper asks an AI what to buy. One file, two channels, and the second one is growing fast.

Most teams are about to find out their product feed was their SEO strategy the whole time.

Ruslan Galba AI x Google Ads

01 July 2026 23:22

We waited three years for the PMax channel report.

It shipped at Marketing Live, and it's the first time Google has shown where the budget actually goes.

The first thing it reveals on most accounts is uncomfortable: a big share of that glowing PMax ROAS was branded search wearing a costume. Conversions the brand was already winning for free, billed back as PMax performance.

I'm pulling the report on every account this week and re-running the math with brand stripped out. A few are going to look very different once the costume comes off.

That stripped-down number is what should set next quarter's budget, not the blended headline.

Ruslan Galba AI x Google Ads

30 June 2026 23:39

Google's whole pitch at Marketing Live was "just tell the AI your goal and step back."

Here's the part that doesn't make the keynote slide: the machine optimizes for Google's revenue, not your margin. Those two line up right until they don't, and the account can't feel the difference.

I'll automate the boring 90%. The agent can build, bid, and report all day. What I keep my hands on are the few levers Google quietly designs out of the interface: margin-aware targets, brand exclusions, and where the budget is actually allowed to go.

Hand over the work. Keep the calls that cost real money when they're wrong.

Ruslan Galba AI x Google Ads

29 June 2026 23:38

Google quietly handed Smart Bidding permission to bid on queries you never targeted.

They're calling it Smart Bidding Exploration.

The pitch is free incremental volume. Until you've checked, the reality is it's fishing in water you deliberately left out, and you pay for every cast.

I'm not killing it on reflex. I'm reading the search-terms report first, tagging which exploration queries actually convert versus which just spend, then deciding from the data instead of the announcement.

Free volume is only free if it converts. The search-terms report tells you which one you've got inside a week.

Ruslan Galba AI x Google Ads

29 June 2026 16:30

Google just put a sunset date on Dynamic Search Ads. Here's what to lock before September flips it to AI Max.

AI Max for Search reached general availability back in April. The news at Marketing Live this month was the timeline for what comes next: Google confirmed the DSA wind-down. Auto-upgrade of broad match plus ACA starts September 2026. Full DSA migration got pushed to February 2027 after advertisers pushed back.

Calling it a DSA sunset undersells it. Google's folding DSA into AI Max with more reach and less of your control, and it happens whether you prep or not. So I'd rather flip it on my terms.

Here's what AI Max actually changes under the hood:

Final URL expansion. The system can send traffic to any page on your site it thinks matches intent, not just your designated landing pages. Powerful, and the fastest way to leak budget to your shipping-policy page if you leave it open.

Search-term matching. Broad keyword matching gets a Gemini layer on top, so you surface on queries you never added. More volume, more junk, depending on how clean your negatives are.

Text customization. It rewrites headlines per query. Great for coverage, a brand-voice risk if you sell anything regulated.

Google's own number: running the full suite (term matching + text customization + URL expansion) drives roughly 5-10% more conversion value at similar CPA versus matching alone. Real, but that lift assumes your account hygiene is already tight.

So before September, I'm locking three things on every account:

1. URL expansion controls. Exclude the pages that should never take paid traffic - cart, account, policy, thin collection pages. Whitelist the ones that should.

2. A hardened negative list. AI Max widens matching, so the negatives are now doing more work than the keywords. I rebuild them before turning it on, not after I see the waste.

3. Brand exclusions. Same reason as always. I don't want to pay AI Max to harvest branded search I already get for free, then call it incremental.

The new search-term by landing-page report that shipped late June is the one to watch. It finally shows where expansion is actually sending traffic. First thing I open on any AI Max account now.

The accounts that get hurt in September are the ones running on default match types and a thin negative list today. AI Max doesn't create that problem. It scales it. I'd rather walk in with the levers already set.

Ruslan Galba AI x Google Ads

28 June 2026 05:31

I let an AI agent run live ad ops for 30 days. Here's what broke - and the one save no human would've caught.

Everyone's posting the 5-minute demo right now. Agent checks creatives every 15 minutes, scales winners, pauses losers while you sleep. Looks like magic.

The demo is easy. The part nobody films is week 3, when the loop has enough rope to do real damage.

Here's what I actually ran. A Claude loop wired into the Meta and Google APIs, watching a live Shopify account in the mid-five-figures a month. Read performance, propose changes, execute inside guardrails. I built it because I wanted to find where it breaks before a client account did.

It failed in four places. Same root cause every time: the agent optimizes perfectly against the signal it can see, and it can't see the context that lives in my head.

1. It tried to pause a "losing" campaign that was actually our retargeting floor. Low ROAS on platform, high incrementality in reality. The agent doesn't know which conversions would've happened anyway.

2. It scaled a winner 80% in a day off three days of data. The math looked great. The variance was noise. A human waits for the sample. The agent wanted to act now.

3. It kept proposing budget shifts into branded search dressed up as "high ROAS." The same cannibalization trap I strip out of every account by hand.

4. It had no concept of margin. A 4x ROAS SKU at 10% margin loses money after COGS and fees. The platform doesn't pass margin, so the agent never saw it.

So I added four guardrails. Daily budget change capped at 20%. No pause without a 7-day window and an incrementality flag. Branded search locked out of the decision set. The margin table passed in as a hard constraint, not a suggestion.

After that, it earned its keep. Late one night it caught a tracking break I'd have missed. Conversions flatlined on one campaign at 2am, it flagged the anomaly and paused spend before the day's budget leaked into nothing. I was asleep. Someone reviewing in the morning eats the full loss.

That's the actual product. An agent that runs 24/7 inside constraints a senior operator wrote, and escalates the calls it can't judge.

The people winning with this in 2027 won't be the ones who automate the most. They'll be the ones who know exactly which four levers to never hand over. I run this across the brands at Tegra because I'm technical enough to build the loop and burned enough not to trust it raw.

Ruslan Galba AI x Google Ads

21 June 2026 13:18

Your best product has a 4.6x ROAS and Google is showing it for 10% of searches.

That's not a top performer. That's a top performer Google is hiding from 90% of the buyers who want it.

I audited a car accessories brand a few weeks back. Their top SKU was printing money. 4.6x ROAS, the team was thrilled. Then I pulled impression share: 10%.

Google was serving their best product for 1 in 10 eligible searches. The other 9 went to competitors with worse products and bigger budgets.

Here's the part most operators miss.

Impression share is the most underread number in the account. Everyone watches ROAS. Almost nobody checks how often their winners actually show up.

PMax makes it worse. It bundles your strongest SKUs with your weakest, spreads budget thin across the whole pile, and reports one blended number that looks fine. The aggregate hides the fact that your hero is starving while a dead SKU eats its impressions.

The scaling opportunity is already bought and proven inside the account. The blended number just hides it.

The math is simple once you look at the right level.

Headroom = (100% minus current impression share) times current revenue.

That car care SKU at 10% IS had 90% of its revenue still on the table - in a product already proven to convert at 9.6x. No new creative, no new audience, no new offer. Just room to breathe.

I ran this same pull across a portfolio of 20+ brands.

The pattern held in almost every account: the highest-ROAS campaigns consistently had the lowest impression share. 4.6x at 10% IS. 2.85x at 40% IS on a branded campaign that had been paused for weeks. The losers were getting all the visibility. The winners were getting starved.

Combined headroom across just three campaigns in one account: $30K+/mo in untapped revenue.

Here's the audit I run on every new account. Takes about 10 minutes.

1. Export the search terms report, last 30 days.
2. Sort by conversion value, descending.
3. For your top 10 products, pull impression share.
4. Flag anything above 2x ROAS and below 40% IS.

Those flagged products are your cheapest scale lever. They convert well, and Google simply isn't serving them enough.

Usual causes: budget shared across too many products, PMax bundling strong and weak SKUs together, or a bid strategy capping the campaign because low performers are dragging the average down.

The fix is to isolate the high-ROAS, low-IS products into their own campaigns with dedicated budget and target impression share floors. Give the winner room instead of making it fight its own losers for spend.

That car care SKU went from a quiet line item to the account's growth engine on the same ad budget.

Most of the scaling I do now starts here, before I touch a single bid. The growth was already paid for and proven. It was just buried in the aggregate where nobody thinks to look.

Ruslan Galba AI x Google Ads

20 June 2026 13:49

"Does SEO actually make my paid ads work better?" I get this every week.

The answer is yes, and almost nobody connects the three places it shows up.

Across the 20+ DTC accounts we run, the strongest paid performers almost always have real SEO underneath them. That overlap isn't accidental - both budgets feed the same relevance machine Google scores.

Three places the overlap shows up, in order of how much money it moves.

1. Quality Score is half SEO work.

Keyword relevance, landing page experience, and site speed are the three inputs Google scores. Those are the exact things an SEO team is fixing every week. When I inherit an account with a slow, thin landing page, the CPC on top-of-funnel informational terms is inflated and I'm paying a tax I can't bid my way out of. Tighten the page, speed it up, match the intent, and the same auction position costs less. The auction rewards relevance, and relevance is what SEO builds.

2. Organic authority de-risks the ad account.

I've watched stronger sites clear Merchant Center reinstatements faster than thin ones. When the organic footprint says "this is a real business with a real catalog and real reviews," the trust review moves quicker. A weak site with no organic signal sits in limbo. That's not a guarantee, but the pattern is consistent enough that I treat organic authority as account insurance now.

3. AI results reward the same signals.

AI Overviews, AI Mode, ChatGPT, Perplexity. They surface the brands with credible content, real mentions, structured data, and a clean review profile. Those are SEO outputs. So the work that lifts a Quality Score is the same work that decides whether a brand shows up when someone asks an AI instead of typing a query. Same content work, three separate distribution channels paying it back.

The strategic version: organic is the foundation that lets paid reach further. Advertorials and cold traffic plays only hold up when there's a credible site behind the click. Without it, the spend buys clicks to a page the algorithm and the buyer both distrust.

To be clear, good SEO doesn't guarantee good ads. I've seen well-ranked sites run terrible campaigns. But auditing accounts across the DTC brands we run, I almost never find a top performer sitting on a weak organic base. The best ones have both, quietly working together.

In every audit I run, paid and organic show up on the same invoice for trust. The page Google ranks is the page Google charges less to put in front of buyers.

Ruslan Galba AI x Google Ads

04 June 2026 02:54

"Ugly" presell pages are beating polished homepages on Google Ads, and it isn't close.

I've tested this across 40+ local service and lead-gen accounts. A plain presell page between the ad and the form cuts cost-per-lead 40-80% versus sending the click straight to a homepage or a bare contact form.

Not prettier pages. Not fancier design. Often the opposite.

(I know. It sounds backwards. Stick with me.)

Here's why it works.

A cold searcher doesn't trust you yet. Dropping them on your homepage, or worse, a "Request a Quote" form, is asking for the lead before earning the attention. It's handing someone a receipt before they've seen the menu.

A presell page builds the bridge. It educates first, asks second.

Here's the structure that actually converts:

1. Headline that names the problem, not your business. "AC short-cycling in this heat?" beats "[Company] Heating & Air."

2. Agitate. What it costs to wait. A failing compressor in July. A toothache that becomes a root canal. A refi window that closes when rates move.

3. Education. Teach one genuinely useful thing. This is where trust is earned. (Yes, give away real value. I know it feels wrong.)

4. Introduce your service as the answer to what you just taught.

5. Social proof. Real reviews, real photos, license numbers, years in the trade.

6. Comparison (optional). Position against the aggregator or the do-nothing option without trashing anyone by name.

7. One clear CTA. Call or book. One action, no confusion.

This isn't revolutionary. Most operators skip it because it feels like extra work, and because their web guy built one site for the whole business years ago.

The key insight: presell pages work because they match the temperature of the traffic. Cold searchers need warming. Your homepage assumes they're already sold. That mismatch is what's quietly draining the budget.

And here's the part that compounds. Google rewards relevance. A presell page built tight around the exact search intent earns a higher Quality Score. Higher Quality Score, lower CPC, more clicks from the same budget. On one HVAC account I took cost-per-booked-call from $9 to $5 with nothing but a presell page in front of the form. On a PI account where clicks run $80-$200, even a 30% Quality Score lift is real money back into the pipeline every single day.

One compliance line, because half my verticals are regulated: the social-proof and comparison sections are where PI law, insurance, and mortgage operators get themselves in trouble. Use real, attributable reviews, not invented testimonials. No "guaranteed settlement," no "lowest rate," no "guaranteed approval." Trust is built with the mechanism and real proof, not a promise a bar or DOI examiner can read.

A presell page is trust architecture, built before the form. The operators doing it right are quietly outbidding competitors on the exact same keywords for less.

Ruslan Galba AI x Google Ads

17 July 2026 03:54

Thirty-one entries from the claims blocklist our AI check runs on supplement ads.

Severities included, so you can steal the structure, not just the words. Bookmark it.

Severity 1 - blocked outright, no rewrite, no review. Disease claims, whatever verb they wear:

• cures
• treats
• mitigates (the statute's forgotten verb - it's in the same sentence as the other four)
• heals
• reverses
• prevents
• eliminates
• remedies
• diagnoses
• "fights [any disease]"
• "kills [pathogen/cell]"
• "beats depression" - a named disease puts the sentence here no matter how soft the verb
• any comparison to an Rx ("like the injections, without the needle") - an implied drug claim, and the exact pattern in the current warning-letter wave. No human review clears this one for a supplement.

Severity 2 - blocked pending rewrite. No disease named, but the sentence promises to fix a departure from normal:

• manages blood sugar
• lowers cholesterol
• reduces inflammation
• relieves anxiety (reads as diagnosis more often than not - when it does, it's severity 1)
• balances hormones
• restores testosterone
• fixes gut issues
• ends sleepless nights
• "clinically proven to [anything]" - the odd one out: you can't rewrite your way past this. Produce the studies or delete the words.

Severity 3 - routed to a human. Context decides, and pretending a regex can rule on these is how linters get people sued:

• supports / maintains / promotes (fine WITH normal-range framing, claim without it)
• customer testimonials with outcome numbers ("I lost 30 lbs" - publishing it makes it YOUR claim)
• before/after phrasing, even without images
• "without a prescription"
• "doctor-formulated" and credential adjacency
• money-back-if-it-works framings (implies a treatment outcome)
• "safe" as an absolute
• dosages positioned as protocols
• anything in second person about the reader's diagnosis

Three notes before you copy it.

This list handles claim classification - what your sentence says. Substantiation is a second lock on the same gate: a perfectly worded "supports" claim with no evidence file behind it is still an FTC problem.

The linter tells you which claims to defend, not whether you can.

The platform's gate is separate again - restricted concepts are their own list and their own fight.

And a verb list is only half a linter. "Formulated for people with type 2 diabetes" contains zero blocked verbs and names a disease - the other half is a disease-name list, and it's longer than this one.

To use it tonight: paste your five live ads and your landing page into a doc and search it against severities 1 and 2. Anything that hits severity 3, a human reads before it runs again.

Ruslan Galba AI x Google Ads

16 July 2026 12:31

Everyone's selling "an AI media buyer that works while you sleep" this month. Nobody shows you what it actually finds.

Here's what ours flags in apparel accounts, ranked by how often we find it - and none of it is bid magic.

1. Dark variants. Size-level availability that never synced, so a third of the catalog reads out-of-stock to Google while it sits in the warehouse. The algorithm doesn't bid on inventory it thinks is dead. This is the single most common finding, and no human checks it weekly because it's boring.

2. Price mismatches on markdown. The site says 40% off, the feed still says full price. Google reads the click-to-page gap and disapproves the item - or its automatic updates rewrite your feed price without asking. Humans catch either in week three of the sale. An agent catches it the first morning.

3. Returns missing from conversion values. The account optimizes on revenue, the brand keeps 60-75% of it after returns, and the bidder happily scales the size-guessers. Wiring refunds back in is a one-time job that most accounts never do because nobody owns it.

The pattern across all three: the machine isn't smarter than your media buyer. It's more willing to do the unglamorous checks every single day.

One rule the "while you sleep" pitches skip: everything it drafts waits for a human signature before it spends.

The dark-variant check: open Merchant Center, filter to out-of-stock, and count how many of those sizes your warehouse says it has.

Takes ten minutes. In apparel accounts we open, the answer is rarely zero - and it's the cheapest revenue you'll recover this quarter.

Ruslan Galba AI x Google Ads

15 July 2026 22:35

The same 30-second AI video ad costs $2.10 or $9.40 to render.

Same script. Same visual quality. The only difference is knowing which model renders which shot. Worth adding to your bookmarks.

We just spent a week reverse-engineering every current video model - Veo 3.1, Kling 3.0, Seedance 2.0, Sora 2 Pro, Grok 1.5, Wan 2.7, Hailuo 2.3, Gemini Omni - against their actual API pricing. Not the blog-post pricing. The real request schemas and rate cards.

Here's what nobody tells you:

1. "Which model is best?" is the wrong question.

A performance ad is not one video. It's 8-12 shots with completely different jobs:

- Proof shots (the product actually working): ~5% of your seconds, ~80% of your persuasion
- Talking head: the only place lip sync matters
- B-roll: nobody scrutinizes the workbench shot
- Filler: 1-3 second cuts between beats

Route each shot class to a different model and your cost drops 40-60% with zero quality loss where it counts.

2. Minimum billable duration is the hidden tax.

Every model has a billing floor, and they never advertise it:

- Wan 2.7 bills a true 2 seconds
- Kling 3.0 floors at 3s
- Seedance 2.0 floors at 4s
- Grok floors at 6s
- Sora on API floors at 10s

That "cheap" $0.045/sec model? A 4-second talking head bills 10 seconds. You paid $0.11/sec and never noticed.

For fast-cut ads (1-3s per shot), this one number matters more than the per-second price.

3. Per-video billing is an arbitrage.

Some tiers bill per clip, not per second. Veo 3.1 Lite: ~$0.17 for an 8-second 1080p video.

Author your b-roll AT 8 seconds, harvest 3 usable cuts from each clip, and you're paying ~$0.05 per cut. Cheaper than every per-second model on the market.

4. Lip sync is binary. Enforce it in code.

Only 2 of the 9 models can render a person speaking with usable lips. Everything else produces the uncanny half-sync that instantly reads as AI.

Our pipeline literally refuses to render a talking head on a non-lipsync model. One hard rule, zero embarrassing ads.

5. The #1 leaderboard model shouldn't render your ads.

Gemini Omni is #1 on blind arena Elo right now. We still don't generate with it:

- 3x the cost of the right model per shot
- hard-blocks prompts containing real brand names
- quality degrades after 4 sequential edits (we tested)

But as an EDITOR it's unbeatable: $0.84 to change one element in an existing clip vs $1+ to re-render the scene. Next-gen model, wrong job description.

6. Never trust the docs. Render one probe.

We ran a $0.09 test render before trusting any of this. Found two things no documentation mentions: one model returned a 10-second clip for a 6-second request, and another attaches a silent audio track that breaks naive pipelines.

One coffee's worth of API credits beats a week of confident assumptions.

---

The full routing table is in the screenshot.

Steal it, wire it into your pipeline, and stop paying premium rates for shots nobody looks at.

Ruslan Galba AI x Google Ads

15 July 2026 16:07

60 hours of agent work in a day is now real. The scarce resource is a clean 12-minute review.

The number is OpenAI's, from its heaviest users. Three of us manage $10M+ a month across 20+ brands, so I can tell you where that stat stops being impressive and starts being dangerous.

One person can't physically work 60 hours before dinner.

One operator can review 60 hours of parallel work - if the system surfaces the right decisions.

That second sentence is doing all the work, and it's the part everyone skips.

Teams are still teaching people to write prettier prompts. Workshops, internal prompt libraries, certification decks.

It genuinely annoys me, because the operators actually compounding are designing the review receipt first.

One row from ours looks roughly like this:

Job: investigate a PMax spend anomaly

Evidence: change log, query delta, inventory state

Proposed action: staged, never live

Authority: Google owner approval required

Expiry: discard if the account state changes

Delegation design is the skill now. The model keeps working long after the chat window stops being interesting, and fluency in 2026 means deciding what deserves to run - not phrasing it nicely.

The economics follow. When one operator runs research, analysis, QA, and production in parallel, headcount stops being a clean proxy for capacity.

Ours has been stuck at three on purpose.

Founder check, one email: ask the agency you pay by headcount to show you one completed agent job with its evidence, proposed action, approver, and expiry. "We check everything manually" means 60 hours of unread homework, billed to you.

Your value moves toward judgment - choosing the job, setting the boundary, killing bad output before it touches money.

I still write prompts. I just spend most of the week deciding what deserves to run.

Ruslan Galba AI x Google Ads

14 July 2026 22:04

Every AI video model comparison on X compares vibes. Not one of them compares invoices.

So I pulled the actual per-second math on every current model, and the "cheap" models aren't cheap, the "premium" ones aren't premium, and the best deal on the market is hiding inside a billing unit nobody bothers to divide.

The full breakdown is in the screenshot and at the end of the post. Bookmark it if you'd like.

Now the three findings that make this table worth money:

1. Per-video billing is arbitrage.

Per-second models punish long clips. Per-video models reward maxing the clip length.

If your b-roll beats are authored at exactly 8 seconds, Veo Lite undercuts everything.

Structure your shot list around the billing unit, not the other way around.

2. The audio surcharge is a tax most people pay for nothing.

Native audio adds up to 40% per second on some models.

If you're laying one continuous voiceover over the footage anyway (you should be), you're paying for audio you delete.

Render silent.

Pay the surcharge only on the beat where the sound IS the ad.

3. There is no winner in this table. That's the point.

- The proof shot goes to Kling.
- The talking head goes to Sora or Seedance.
- The wallpaper goes to Mini.
- The 8 second b-roll goes to Veo Lite.
- The revision goes to Omni.

A single production should touch four of these models.

Anyone running everything through one model is either overpaying on 60% of their seconds or under-delivering on the 5% that carry the trust.

The model wars are content for spectators.
The billing table is content for operators.

Bookmark the table.

The prices will drift, the logic won't: normalize to per-second, route by shot, never pay for audio you'll replace, and let the per-video models subsidize your b-roll.

KLING 3.0
$0.07/sec silent, $0.10/sec with audio
15 sec max. The physics king. Weak lipsync (~22%).

SEEDANCE 2.0
$0.10/sec with video input, $0.165 cold
12 sec max. Real lipsync. Multi-shot consistency is its whole pitch.

SEEDANCE 2.0 MINI
$0.03 to $0.06/sec
The cost floor. Your background shots do not need more than this.

VEO 3.1 LITE
$0.175 PER VIDEO for an 8 sec 1080p clip.
Do the division: $0.022/sec. That's the cheapest quality-per-second in the entire market, and it's invisible because it's billed per clip instead of per second. This is the single most valuable row in this table.

HAILUO 2.3
$0.15 per 6 sec clip. About $0.025/sec. Same trick, second cheapest.

WAN 2.7
$0.08 to $0.12/sec. Solid middle. No audio.

SORA 2 PRO
$0.045/sec at 720p.
The only model that holds a 25 second continuous take, and the best lipsync available. Somehow also one of the cheapest. Nobody talks about this.

GEMINI OMNI
$0.84 per generation WITH video input. It's not a generator, it's an editor. Feed it a finished clip and change one element with a sentence, instead of re-rendering the whole scene. This is a different product category and almost nobody has noticed it shipped.

Ruslan Galba AI x Google Ads

12 July 2026 23:47

"Non-branded campaigns don't scale profitably."

They do. Most operators are just measuring them wrong and structuring them worse.

I've taken non-branded from 15% to 60%+ of total revenue across 20+ e-commerce brands.

The breakthrough was never better keywords. It was intent.

A searcher typing "buy organic skincare" and a searcher typing "how to reduce acne" are two completely different people.

One has a wallet open. The other is reading.

Transactional intent converts 4-6x higher than informational.

So when you bid them the same, send them to the same page, and run the same copy, Google averages the two audiences together and the math goes flat.

That's the real reason most non-brand "doesn't convert."

Here's the case that made it obvious.

A brand was spending $15k/mo on non-branded Search. Breaking even at best.

I pulled their ad groups. Transactional keywords like "buy organic skincare" sat in the same campaign as "how to reduce acne."

Same bids, same pages, same copy.

I split them. Transactional got its own campaign, its own conversion-optimized pages, its own budget.

Informational got pushed down or moved to a separate funnel.

CPA dropped 40% from the separation alone. No new budget.

Google just stopped averaging performance across two audiences that had nothing in common.

The structure I run now sits on three tiers:

• Transactional ("buy X," "X discount"): aggressive bids, pages built to convert on the spot.
• Commercial investigation ("best X," "X vs Y"): moderate bids, comparison and education pages. These people are choosing, not browsing.
• Informational ("how to X"): low bids or a separate Display funnel. Cheap reach, not a closing channel.

One messaging rule carries all three: non-brand means they don't know you.

The ad has to answer "why should I buy this product," not "why should I buy from you." Lead with their problem, not your logo.

Now the part that kills good campaigns: ROAS.

Non-brand brings in new customers.

Their first-purchase ROAS will always look worse than brand, because brand is just harvesting people who already decided.

But non-brand customers often carry 2-3x the LTV, because they're not one-and-done deal seekers who only show up for a discount code.

Judge non-brand on first-touch ROAS and you'll cut every campaign that actually grows the business.

I measure on 90-day customer value, then decide.

The scaling triggers I use:

• Transactional CPA stable for 2 weeks - increase budget 20%.
• Commercial investigation showing positive 90-day LTV - move it up to transactional budget levels.
• New winners surfacing from broad match - add them to exact match immediately.

Brand harvests demand that already exists. Non-brand is the only budget line building demand you don't have yet, and it scales the day each intent gets its own room.

Founder check, two minutes: open your non-brand campaign and read ten search terms. If "buy X" and "how to fix X" live in the same ad group, you just found the reason it "doesn't scale."

Ruslan Galba AI x Google Ads

02 July 2026 16:25

An agent built me a full advertorial funnel in a weekend. It also got two things dangerously wrong.

I've been testing the thing half my feed is excited about. Point an agent at a product, and it spins up the whole funnel - advertorial, listicle, comparison page, a quiz. Work that used to take my team two to three weeks. It did a usable first pass in about two days.

I want to be honest about both halves of that, because the demo crowd only shows you the first half.

What it nailed: structure and speed. It produced a clean advertorial arc, a comparison table that didn't embarrass itself, a quiz with branching logic wired up. As a production tool, it collapsed weeks into days. That part is real and I'm not going to pretend otherwise.

What it got dangerously wrong, twice:

One. The offer. The agent wrote compelling copy around an offer that didn't actually make sense. It assumed a discount structure the brand couldn't afford and never questioned whether the math worked. It optimizes the words, not the economics behind them. A funnel that converts on an unprofitable offer is just a faster way to lose money.

Two. The claims. It generated benefit statements that were, to put it kindly, ambitious. For a supplement brand that's not a copywriting flourish, it's an FTC problem. The model has no instinct for which claims get a brand a letter. Every line needed a human who knew the regulatory floor.

So here's where I landed after the test. The bottleneck in funnel building used to be production - writing, designing, wiring it all up. The agent genuinely moves that. But the work just relocated. It now sits on the offer, the post-click math, and the compliance line, which was always the hard part and never the typing.

I made a version of this mistake the expensive way a year ago, before any of these tools existed. Scaled a funnel with a beautiful front end and broken back-end economics. Cost me a real number to learn that production was never the constraint.

These agents are a genuine upgrade to how fast I can build. They're also very good at confidently building the wrong thing. I run every agent output through the same two questions now before a dollar goes near it: does the offer math survive contact, and will a claim get us a letter. Those two stay human.

Ruslan Galba AI x Google Ads

01 July 2026 16:51

AI shopping traffic to US retail jumped nearly 400% this year. Your product feed just became your most important API.

Most operators still treat the feed as a Google Shopping chore. Titles, GTINs, the stuff you fix when disapprovals pile up. That mental model is about to cost people real revenue.

Here's what shifted in the last two quarters. AI referral traffic to US retail is up nearly 400% year over year. Traffic to Shopify stores from AI search is up around 8x, and orders from it close to 13x. eMarketer pegs AI-platform retail spend at almost $21B in 2026. The buyer is increasingly a model, and the model reads your feed.

Two rival rails are now live for this. OpenAI and Stripe shipped ACP - checkout right inside ChatGPT. Google's coalition is pushing UCP into AI Mode and Gemini. Shopify's Winter release syndicates one catalog out to ChatGPT, Perplexity, and Copilot. Same product data, many more surfaces consuming it.

So I've started restructuring feeds for a reader that isn't human.

A person scans an image and a price. A model parses structured attributes and makes a judgment call about fit. That means the fields most brands treat as optional are now the whole game: material, use case, compatibility, dimensions, what problem the product solves, who it's wrong for. The feed has to answer the question the shopper asked the AI, not just match a keyword.

The discipline transfers almost perfectly. The same Merchant Center hygiene that wins Shopping - clean attributes, accurate GTINs, rich product types, structured data on the PDP - is what gets a product surfaced and cited inside an AI answer. One asset, two channels, and the second channel is growing triple digits.

What I'm doing on accounts right now: auditing feeds for attribute completeness against the questions buyers actually ask, not against Google's minimum required fields. Adding the "who is this not for" data models use to qualify. Treating the PDP and the feed as one structured object a machine has to understand in full.

The brands that win the next two years are the ones whose product data is legible to a machine deciding what to recommend. I'd rather rebuild the feed now, while the traffic curve is still early, than explain to a brand in 2027 why they went invisible to the buyer that grew 400% in a year.

Ruslan Galba AI x Google Ads

30 June 2026 16:06

Everyone's posting AI ad wins this month. The median DTC brand makes $2.04 per $1 of ad spend. Let's talk.

My feed right now is a wall of screenshots. Agent built 5 ads in 5 minutes. $155K to $242K in 30 days. A €1.4M Shopify dashboard. All real, probably. All survivorship.

Here's the aggregate the highlight reel skips. Average e-commerce ROAS in 2026 sits around 2.87:1. The median is 2.04:1. Mid-market brands watched ROAS drop close to 10% year over year while fixed marketing costs climbed about 30%. Net margin for most DTC brands lands between 3 and 10%.

Sit with the median for a second. $2.04 back on $1 spent. After COGS, fees, shipping, and the platform's cut, that's a business running on fumes. No agent fixes a 2x ROAS account by making more ads faster.

The reason "just add AI" doesn't move it: the problem usually isn't the volume of creative. It's unit economics the ad account can't see.

Take a typical 2x account where the owner is convinced he has a creative problem. AOV $40, blended CAC $38, repeat rate under 20%. The first purchase is break-even at best, and the whole business is leaning on a second order that isn't coming. No amount of new creative fixes that.

The lever is the post-purchase flow, the bundle AOV, and the freight line. Move those and blended ROAS barely budges while contribution margin per order can nearly double. That's the number that pays salaries.

More AI ads would've poured faster water into a leaking bucket.

Creative volume is the cheapest input in the stack now. When everyone's agent makes 50 ads a week, ad output stops being the constraint. The constraint walks back upstream to the offer, the margin, and the second-order economics - the stuff no model can see from inside the ad platform.

The AI wins on your timeline are real. They're also the 5% who already had the unit economics to survive scale. For the median 2x brand, the gains come from the boring math under the ad account, not from running the ads faster.

Ruslan Galba AI x Google Ads

29 June 2026 19:36

Google Ads API: Major Changes in Product Reporting from June 15, 2026

Google has announced a long-awaited update - product reporting will become transparent for all campaign types, not just Shopping and partially Performance Max.

What will change?

1. Extended metrics for shopping_product - cost_micros and conversions will now be available for Video, Demand Gen, and App campaigns. Previously, only impressions and clicks could be extracted there, making it practically impossible to assess the effectiveness of a product outside of Shopping/PMax.

2. A single point for reporting - shopping_performance_view - all types of campaigns using Google Merchant Center will return data through this view. No more "hacks" for collecting metrics from different places.

3. Transparency of networks in Performance Max - shopping_performance_view will start returning data for all PMax networks, not just selected ones. Be prepared for a one-time jump in figures in the reports - this is not a bug, but finally a complete picture.

Important nuance: historical data before June 15, 2026 will not contain the new metrics. That is, year-on-year comparisons for PMax after this date will have an asterisk - your dashboards and automatic scripts will need to be either calibrated or marked with a change in methodology.

What to do now?
- Review your own scripts and GAQL queries that pull product statistics
- Be informed about a possible "jump" in Performance Max
- Consider how to correctly merge data "before" and "after" June 15

Ruslan Galba AI x Google Ads

28 June 2026 23:30

I built an agent that spins out 5 Meta ads in 5 minutes.

Then it hit me: so can the agency across town, and so can your competitor's intern.

Creative volume just dropped to zero marginal cost. When anyone can generate 50 ads a week, the ad itself stops being the scarce thing.

What's scarce now is knowing which 5 of those 50 deserve spend. That read comes from watching a thousand of them live and die in real accounts, not from a better prompt.

The model makes the ad. It still can't tell you which one the market will pay for. That's the job that didn't get automated this year.

Ruslan Galba AI x Google Ads

21 June 2026 22:41

Two hours per competitor to manually outline what they're running.

I needed that refreshed for 20+ brands, every week.

The arithmetic on that is a full-time analyst doing nothing else.

That's why I built the system.

SERP APIs scrape every live Google ad variant in the market - headlines, descriptions, sitelinks, callouts, landing pages, the full creative inventory across every keyword cluster. Meta Ad Library pulls every active creative, the landing page behind it, the refresh cadence, and how long each variant has been spending against impressions.

An agentic loop normalizes the corpus, scores positioning against the competitive set on a 100-point differentiation index, flags new entrants since the last refresh, and surfaces the angles nobody in the set is running yet.

One sweep: ~5,000 Google ads + ~9,000 Meta ads, analyzed across 20+ brand-by-market combinations. The version of this I used to do by hand was two weeks of an analyst's time. The loop finishes it before my coffee's done.

The screenshot above is the actual output from this morning. Differentiation Score per brand. Blue Ocean band per market. Notes on what shifted since the last refresh. "Tide launched 132 new ads since prior" is the line that changes how I run Shopping for a brand competing in the same SKU. "Sephora Meta 3 to 49" tells us a category leader just multiplied their creative volume across one market by 15x.

Every Google and Meta campaign me and @andreilunev write for the brands we run at @hellotegra starts from this output. We read the differentiation snapshot, then the per-brand notes, then we open the underlying creative library to look at what the new entrants are actually testing.

The bidding strategy, the creative angles, the negative keyword work, the audience structure - all of it sits downstream of what the competitive set is doing this week.

It's not just a snapshot either.

The same loop monitors the set continuously. Last month it flagged a competitor testing a new landing page angle. We had our counter version live within a week, before they'd finished proving theirs out.

When a competitor refreshes 130 ads, the auction shifts. When a category leader multiplies their creative output 15x, the cost curve moves.

A strategy that ignores the live competitive set is solving a problem from 90 days ago.

This is the layer most performance teams skip, because the manual version doesn't scale past two or three brands. Three of us run that depth across all the brands every week because the system above does the analyst work.

Ruslan Galba AI x Google Ads

20 June 2026 22:44

The forward-deployed engineer role Lenny is describing already exists in ads ops. Nobody is calling it that.

Dan Shipper's thesis from the Lenny podcast: AI agents will create massive new demand for SaaS seats. Not kill SaaS - generate demand. Agents need tools. Tools have seats.

Correct. But there's a filter most product people aren't naming.

The SaaS tools that capture agent-generated demand are the ones instrumented for non-human session patterns. An AI agent querying the Google Ads API doesn't behave like a human analyst refreshing a dashboard. It fires API endpoints at a different scale, in sequences no human click-flow produces, touching attribution data at a frequency that exposes every assumption baked into your conversion event taxonomy.

If your event schema was built for human traffic, agent traffic will produce wrong attribution signals at scale.

Silently.

I read Google Ads API docs like source code - looking for what the endpoint actually fires, not what the documentation says it does. Dev (8 years) before marketing (8 years). That combination is why I build API-level tooling for ad ops instead of managing dashboards.

The forward-deployed engineer in the ads stack sits between the MMP or Ads platform API and the bidding algorithm.

Job: postback architecture, conversion event taxonomy, API key management, signal quality at non-human I/O velocity. In DTC, SaaS, local services - this person determines whether your AI agent's campaign management produces ROAS gains or data corruption.

Three of us manage $10M+ in ad spend. The tooling is built at the API layer, not the UI layer.

The seats flowing to agents are already flowing - to operators who own the instrumentation layer. The ones who only own the dashboard layer are about to find out why that distinction matters.

Ruslan Galba AI x Google Ads

04 June 2026 13:13

You can reverse-engineer any competitor's Google Ads strategy in 10 minutes.

I do this for every new client. It's also something 90% of advertisers have never done once.

Here's the 5-step process:

Step 1: Find your real competitors.
Go to Insights & Reports > Auction Insights. Click into your best campaigns and ad groups. Look at impression share, overlap rate, and outranking share.

This shows you who's actually competing for your most valuable keywords. Not who you think your competitors are - who Google says they are. These are often very different lists.

Step 2: See their entire ad playbook.
Google Ads Transparency Center. Free. Plug in their brand name. Filter by region, format, and date.

You can see every live ad they're running. Their messaging. Their offers. Their creative formats. Whether they're using video, images, or just text.

This is their entire strategy laid out for you. Most brands don't even know this tool exists.

Step 3: Find abandoned keywords.
This requires SEMrush. Go to their domain > Paid Keywords > Position Lost.

These are keywords they used to bid on and stopped. Why did they stop? Maybe the CPA was too high for them. Maybe they shifted strategy. Either way, these are open lanes you can test for a fraction of the competitive cost.

I've helped clients capture six-figure revenue from competitor abandoned keywords alone.

Step 4: Analyze regional strategies.
Back in Ads Transparency, filter by location. Smart competitors geo-target aggressively with different offers per region.

If your biggest competitor isn't bidding in certain cities, that's an open lane. I've found entire markets where the top 3 competitors had zero presence.

Step 5: Exploit the gaps.
Go back through everything and ask: what aren't they doing?

No video ads? PMax and YouTube are wide open.
Generic Shopping photos? Lifestyle shots clear the field.
Lazy product titles? Intent-based titles outrank them.
Traffic dumped on the homepage? Dedicated landing pages take the conversions.

I do the above on auto with AI agentic system with APIs connections, but manual gets the same result. Just slower.

I never run this to copy a competitor. I run it to find the lane they left empty and build there before they notice.

Ruslan Galba AI x Google Ads

04 June 2026 02:49

Everyone says start with Shopping.

That's not always right.

The standard advice: Shopping campaigns are the best starting point for e-commerce. High intent. Product visibility. Let the algorithm work.

For some brands, that's true. For others, it's a fast way to burn money.

Here's when Shopping makes sense:

- You have healthy margins (40%+ gross). Shopping is competitive. If you can't absorb CPC pressure, you'll lose the auction or lose money winning it.

- You have strong product imagery. Shopping is visual. If your photos don't compete, neither do your ads.

- You have competitive pricing. Shopping is a comparison engine. If you're priced 20% higher with no differentiation, you'll get impressions but no clicks.

- You have enough SKUs. Broad catalogs perform better than single-product stores in Shopping.

Here's when Search makes more sense:

- Thin margins. Search lets you control exactly which keywords you bid on. You can target high-intent, low-competition queries that Shopping's algorithm might ignore.

- Custom or unique products. If people don't know to search for your exact product, Shopping won't find them. Search lets you target problem-aware queries.

- Strong landing pages. Search rewards good conversion experiences. If your product pages are weak but your custom landers convert, Search gives you more control.

A DTC furniture brand I ran started with Search on intent-based keywords instead of Shopping, because their margins were thin and their photography couldn't win a comparison grid.

We targeted competitor-aware queries ("X alternative"), solution-aware queries ("best standing desk for small spaces"), and problem-aware queries ("back pain from sitting"). Each keyword landed on a custom page built to match that exact intent.

Conversion rate went from 1.3% to 2.8% in five weeks. 115% revenue lift on the same budget.

Shopping came later, once we had the conversion data to build the feed against real buyer behavior instead of guesses.

The framework I run before either one: check margins, imagery, pricing, and landing pages. All four strong, I start with Shopping. Any one of them weak, I start with Search, bank the conversion data, fix the gap, then layer Shopping on top of a funnel that already works.

Most accounts that burned money on Shopping out of the gate had a thin-margin or weak-imagery problem the channel was never going to fix. The channel was fine. The starting point was wrong for their numbers.