A client came to me spending ₹9 lakh a month across Meta and Google. Good product, decent margins, and a marketing team of three people who genuinely worked hard.
I pulled the account. Forty-one active ad sets. Fourteen of them had spent money in the last 30 days without a single conversion. Nobody had touched the campaign structure in five months because nobody wanted to be the one to "break something that's working."
It wasn't working. It was just running.
We killed 26 ad sets in one afternoon, rebuilt the account into four campaigns instead of eleven, and redirected the budget those dead ad sets were quietly eating. Same total spend. CAC dropped 34% in five weeks.
Nothing about the creative changed. Nothing about the offer changed. We just stopped feeding money to things that were dead.
Where the Budget Actually Goes to Waste
Everyone assumes wasted ad spend comes from bad creative or wrong targeting. Sometimes it does. But across the accounts I've audited, the bigger leak is almost always structural — the way the account itself is built, not what's inside it.
Meta and Google's algorithms need volume to learn. Fifteen conversions a week spread across six campaigns means each campaign is starving — none of them ever leave the learning phase, and you keep paying the "still figuring this out" tax indefinitely.
I've seen founders build a separate campaign for every product SKU because it "feels more organized." It's not organization. It's fragmentation, and it's expensive.
I still open Search Term reports on client accounts and find spend going to "free," "jobs," "salary," and competitor brand names with zero intent match. On one account this was 22% of monthly search spend — money going to clicks that were never going to convert, on keywords nobody had ever reviewed.
This isn't a one-time setup task. Search terms drift every month as new junk queries match your broad terms.
Meta ads have a shelf life. CTR drops, frequency climbs past 3, and CPMs quietly creep up — but the campaign keeps running because it was "the winner" three months ago.
One D2C account I worked with had run the same three creatives for 11 weeks straight. Frequency had hit 6.2. CPA had crept from ₹420 to ₹910 over that period, and nobody had noticed because the dashboard still said "active" in green.
"All website visitors, last 180 days" is not a retargeting audience — it's a leftover pile of people, most of whom forgot you exist. Someone who visited once 5 months ago and never came back is not the same prospect as someone who added to cart yesterday.
Treating them identically wastes budget on the cold end and under-serves the hot end.
The blue "Boost Post" button inside Instagram is not the same tool as Meta Ads Manager. It skips proper objective selection, audience layering, and placement control in exchange for one-click convenience.
I've taken over accounts where 60%+ of the "ad spend" for a quarter went through Boost, with no conversion tracking attached at all. The team genuinely could not tell me what that money bought.
The Campaign Structures That Keep Outperforming
Across the accounts I've run, a handful of structures show up again and again on the ones that actually scale. Not because they're clever — because they give the algorithm clean data and give you fewer places to hide problems.
The 3-campaign Meta setup
One campaign for cold prospecting using broad targeting with Advantage+ audience expansion turned on. One campaign for warm retargeting, segmented by the recency tiers above. One campaign for a lookalike built from your best customers — not all purchasers, your top 20% by LTV.
That's it. Three campaigns, clean budget allocation, no internal competition for the same auction.
Branded search protection on Google
Run a dedicated exact-match campaign on your own brand name, separate from your generic search campaigns. It's usually your cheapest conversion — often ₹5–15 a click — but if a competitor is bidding on your name and you're not defending it, you're bleeding customers who were already looking for you.
The 70-20-10 creative testing split
70% of budget to proven creative that's currently converting. 20% to iterations on that winner — new hooks, new first frames, same core message. 10% to genuinely new concepts you're not sure will work.
This keeps the account performing today while still feeding the pipeline for next month's winners. Accounts that skip the 10% eventually run out of fresh creative and performance decays with no replacement ready.
Always-on retargeting, separate from campaign launches
Retargeting shouldn't turn on and off with your promotional calendar. It should run continuously, on its own budget, regardless of what prospecting campaigns are doing that week. The people in your retargeting pool don't stop being warm leads just because you're between launches.
What I'd Tell Someone Auditing Their Own Account Today
Open Ads Manager. Sort every ad set by spend, last 30 days. Look at the ones with spend but no conversions. That list is your first fix, before you touch creative, before you touch targeting, before you touch anything else.
Most accounts I audit could cut 20–30% of "wasted" spend without losing a single conversion, just by killing what's already dead and consolidating what's fragmented. That's not a growth hack. It's just paying attention to an account most teams stopped looking at closely months ago.
Do that first. Then, and only then, start testing new creative and new audiences with the budget you've freed up.
✅ No ad set running 14+ days with zero conversions
✅ Negative keyword list reviewed in the last 30 days
✅ Creative frequency checked weekly, refreshed above 3–4
✅ Retargeting segmented by recency, not one bucket
✅ Zero spend going through Boost Post
✅ Branded search campaign running separately from generic search
✅ At least 10% of creative budget testing genuinely new concepts
Run this against your own account this week. Most of it takes an afternoon, not a redesign.
Next up, I want to write about something nobody asks me about but should — how AI tools are actually changing the day-to-day of running these accounts in 2026, and where they help versus where they quietly make things worse.
See you there.
— Suraj
Frequently Asked Questions
Across the accounts I've audited, 20–30% of monthly spend is usually recoverable just by killing dead ad sets, cleaning negative keywords, and consolidating fragmented campaigns — before any change to creative or targeting.
Weekly, at minimum, on any cold-audience campaign. Once frequency crosses 3–4 for a given creative, expect CTR to drop and CPMs to climb. Have replacement creative ready before that happens, not after.
No. Split budget based on where your data says the marginal rupee converts best right now, not a fixed ratio. Review this monthly using blended CAC by channel, and shift budget toward whichever channel is currently more efficient.
Only for pure engagement or reach goals with no conversion tracking need — and even then, Ads Manager gives you the same objective with proper audience and placement control. There's rarely a good reason to use Boost over Ads Manager once you're spending real money.
Comments
Post a Comment