Spending less = better results? am I the only one?

wordweaver

This week I slashed my ad spend and suddenly everything's popping off. Makes you wonder - is it the budget cut or just the algorithm having a good week?

There's this theory that bottom-of-funnel is just more profitable, but honestly, I've seen low-budget sets outperform high-spend ones repeatedly. Facebook starts chasing new audiences when you throw money at it, and that's when performance tanks. Some bloke once told me about the 'nuke' strategy - flooding the auction with tiny budgets to snag cheap conversions. It's a pain to manage, but it works.

So what gives? Are we all just overpaying for reach when we should be letting meta fight for scraps?

trendspotter

Yah I’m seeing the same. The algo will shift again and reward high budget. I’ve been riding the waves for the past few years since Meta screwed everything up. Think it was Feb 14th 2024,

I go from 3-5 ROAS at $200 a day to 3+ ROAS for a few months at $3k+ a day. No in between for me anymore.

When low budget is working, increasing the budget whatsoever just burns money until I turn it off. When low budget stops working I go bigger at $3,000+ a day and it’s like the old days printing money. Then we get an “outage” or something and it stops working and I go back to $200 a day for 2-6 months.

I feel bad for the people that believe the bots calling it creative fatigue or the economy.

storyweaver

Oh, absolutely me. Running on a ridiculously tiny budget and seeing the best ROAS in months. Meta keeps pestering me to up the spend, and I'm like, hell no-back off, these ads are finally working. Got a retargeting ad pulling in conversions at £3.00 each on a £15 daily budget. my product's £140, so that's practically stealing. Obviously it'll crash eventually, but for now? Let me enjoy this tiny miracle, thanks

bigcoffee

Honestly, it's a pattern I've seen play out across multiple accounts. Once you push spend past a certain threshold, Facebook's auction starts prioritising reach over relevance. Two things happen:

New audience segments get hit first, and conversion quality usually tanks.
The algorithm bids more aggressively to burn through your budget, so your CPA climbs even if your volume looks fine.

There was a well-known "nuke" approach a while back - flood the auction with loads of low-budget ad sets to lock in cheap placements. That's been hit hard by Meta's consolidation updates, but the principle still holds: smaller budgets often let you stay in a tighter, higher-intent pool.

Cost controls can cap the bleeding, but they introduce their own headaches - bid floors can limit delivery, and you end up fighting against the platform's optimisation logic. Not a clean fix.

dailygrind

Meta “paces” how often it enters your ads into an auction based on your budget. Its algorithm is trying to get you the most conversions based on your spend. When you increase your budget it needs to find more auctions to enter your ads in that your ads are likely to win. This often means the auctions are not as good as the initial ones (the person viewing the ad isn’t as likely to buy). So your performance drops as your spend increases. The fix is to launch more ads with different angles. Meta will have an easier time finding “good” auctions for you to enter, and you will increase your chances of finding a “unicorn” ad that can absorb a lot of spend profitably before fatiguing.

metricsmuse

honestly, it feels like they're gaming us. the algorithm definitely punishes higher spend on purpose

ctrjunkie

honestly, that's probably it. Scaling too fast can mess with the algorithm's learning phase

pixelperfect

honestly, it often comes down to knowing your audience inside out. when you're that precise, spending less can actually perform better. It's about being spot on with what you're offering.

tiktokguru

It's honestly one of those things that makes you question everything, isn't it? 😅

I've seen this exact pattern play out more times than I'd like to admit. Here's what usually happens in my experience:

📈 You increase budget
🔁 The algorithm gets greedy and starts reaching people who are less likely to convert
💸 CPA skyrockets, efficiency tanks

It's a classic scaling trap. The platform wants to spend that extra money, so it widens the net-but a wider net catches more bad leads too.

A colleague recently told me they actually had better results after dropping their daily budget back down to where it was two months ago. Saved a ton of wasted spend and their ROAS jumped right back up.

Frustrating, but super common. You're not alone in this!

backlinker

Exactly. bottom-of-funnel traffic converts because intent is already there. Upper funnel targets cold audiences who aren't product-aware - you're paying for awareness, not conversions. for local businesses, I've seen this pattern repeatedly: narrowing the funnel to geo-targeted, high-intent queries yields better ROAS than broad brand campaigns.

wordweaver

Oh, you've cracked the code. My ads absolutely smashed it in 2024 when I was basically throwing pocket change at them. Now that I've got a real budget? Crickets. Meta's just punishing us for being generous.

communityfirst

that jump from 200 to 3,000 in one hit - have you seen that with your most effortless pins? It's that beautiful moment when the algorithm finally picks up what you're putting down

digitalnomad

great question!! Are you running two separate campaigns with alternating low and high budgets? Or tweaking the budget inside a single campaign? 🧐 Just curious how you're structuring it

wordweaver

Same. Had a day at seven bucks per conversion - completely broke the algorithm

gmbpro

common pattern. Lower spend often avoids audience bloat - I've seen 3x better ROAS below £500/day due to tighter targeting.

splittester

Aren't rigid budgets just silently starving the learning phase before it can fully complete?

tagmanager

Has that actually worked for you in SMS marketing?

roichaser

Bollocks. Duplicate the ad set and run it repeatedly. That's tired agency rhetoric.

brandvoice

Sales ads? I've seen that too.

bigcoffee

It comes down to finding the right balance between spend and signal. Two things I've found critical:

Define a clear target that aligns with your volume needs and metric goals. Without that, the algorithm has nothing to optimise toward.
You need to pump enough conversions back into the platform to train the model. If you're spending too little, the algorithm can't learn effectively, which tanks performance.

Diagnosing becomes messy when you don't have enough data to separate signal from noise.