Binghott claims Meta CBO campaign spent $1.5M+ in 30 days
Binghott says one Meta CBO campaign on Incremental Attribution spent more than $1.5M in 30 days at 5% lower cost per conversion. A follow-up screenshot showed $823,110 in June spend across 1,532 ads, with 200+ above $1,000.

TL;DR
- One CBO campaign hit $1,508,114.36 in 30 days on a $50,000 daily budget, while binghott said the account moved from nine campaigns to one Incremental Attribution CBO and cut cost per conversion by 5% the $1.5M CBO thread.
- The follow-up proof point was a deep ad bench, not two hero ads: binghott said one June campaign spent $823,110 across 1,532 ads, with 200+ ads above $1,000 the 1,532-ad screenshot.
- His control layer was cost caps, budget minimums, and maximums, with high caps walked down over time and caps used to “buff or nerf” distribution the cost-cap walkthrough the buff-or-nerf reply.
- The caveats were explicit: CBO optimizes to cost, struggles on low volume, can fight required product or geo spend, and gets messier when margins vary the CBO caveat thread.
- The bigger workflow shift was away from campaign fiddling and toward creative, product, offer, copy, CRO, landing pages, and site-change monitoring the creative-and-site reply.
Meta’s Advantage+ campaign budget doc describes CBO as one central budget that redistributes spend in real time to ad sets with the best opportunities. Its incremental attribution help page says the model optimizes and reports on incremental conversions, and also says bid controls cannot be used. That last note is awkward next to binghott’s reply saying an older critique predated IA having cost caps available, while Ben Radack’s LinkedIn response agreed with the CBO plus cost-cap guardrail stack but kept an ABO lane for ads that fail after being moved into a scaling CBO.
The $1.5M CBO
Binghott, a buyer who says he has worked on Facebook ads for 18+ years and studied billions of dollars in spend, framed the move as account consolidation. The brand had been spending six figures across nine campaigns in March, then spent $1.5M+ in the last 30 days through one CBO campaign on Incremental Attribution, according to his thread.
The screenshot attached to the thread showed the visible campaign row at $1,508,114.36 spent with a $50,000 daily budget. Meta’s developer documentation gives the dry version: Advantage campaign budget automatically finds available opportunities across ad sets and distributes budget in real time.
Binghott’s sharper claim was that less media buying work created more room for creative, offers, copy, CRO, products, and landing pages. That is the useful part for creative teams: the campaign structure is being treated as plumbing, not the main performance lever.
The 1,532-ad bench
The follow-up answered the obvious objection that CBO dumps budget into one or two ads. Binghott said the June campaign spent $823,110.02 across 1,532 ads, with the top ad at $80,000, ad #30 still above $5,800, and 200+ ads above $1,000 in the screenshot post.
His interpretation was that “ugly” ad-level CPAs can hide assist value. Three highlighted ads spent about $30,000 in June, then only $94 in July after he left them on and let Meta stop spending, according to the same post.
One reply showed why ad-level ROAS can mislead. A good-looking image ad used a discount code and looked bottom-funnel, but binghott said it was not scalable and should not spend more than it already had in the discount-code reply.
He also said the campaign ran “across a bunch of ad sets” in one reply, and that he batches ads for organization and control while preferring consolidation where possible in another reply.
Cost caps as knobs
Binghott’s control system was not “trust Meta blindly.” It was CBO plus cost caps, min budgets, and max budgets.
His mechanics, as described across the replies:
- High cost caps behave like highest volume, then caps get walked down toward the desired CPA, according to his cost-cap walkthrough.
- Caps set too close to target can block spend from finding new audiences, he argued in a reply.
- Cost caps work better as distribution controls than hard pass-fail limits for every ad set, according to his “buff or nerf” framing.
- He preferred cost caps over bid caps because bid caps had become too rigid for his taste in the bid-cap reply.
- New creative batches could be organized as new ad sets, with min budgets used only when needed in his batching reply.
Meta’s bid strategy docs describe cost cap as a strategy that tries to get the most results while meeting a target cost per action, with no guarantee that the limit holds exactly.
Incremental Attribution gotcha
Incremental Attribution was central to the case study. Binghott said it tells Meta to optimize toward conversions the ads actually drove, rather than every last-touch conversion Meta can claim in the original CBO thread.
The migration detail was more specific than the headline. In replies, he questioned whether old and new campaigns overlapped, asked whether an account had been fully optimized for IA, and said he had seen IA work well when the old setup was paused and the new one turned on in one move in the full-switch reply in the IA optimization reply.
Meta’s own incremental attribution page says the model enables optimization and reporting based on incremental conversions, and says the attribution model cannot be changed after publishing. The same page says bid controls cannot be used with incremental attribution, while binghott said an earlier critique came from before IA had cost caps available.
That doc gap is worth flagging because the whole workflow depends on cost caps being available with IA in the account being discussed.
Where CBO breaks
Binghott’s caveat list was unusually useful because it named the failure modes instead of selling CBO as a trick:
- CBO optimizes to cost, not profit.
- Mixed-margin products can push spend toward cheap, low-value conversions.
- Required spend on a new product or geo can fight the system’s preferred allocation.
- Low-volume accounts give Meta less signal.
- Consolidation can reset learning and create a short-term transition cost.
He said ABO still makes sense for lower-spend or lower-conversion accounts, accounts that cannot feed Meta good data, and objectives like reach, brand awareness, or link clicks in the same thread.
The scale claim was not limited to the $50,000-per-day screenshot. Binghott later said he also has brands spending about $1,000 per day with the same structure in a follow-up, then told a lower-data account that less data meant less trust in Meta at the start in a reply about smaller accounts.
Site changes over campaign structure
Binghott’s bluntest line was that “CBO vs ABO isn’t your problem. Everything else is,” in a reply that pointed back to ads and the site his reply to Challenge_Mnstr.
He listed the external variables that can move numbers more than campaign structure: landing page changes, product launches, price updates, broken checkout, competitor promos, holidays, and news cycles in the original thread. That is also why he plugged URL Love It, a tool he said he is building to catch page changes that affect ad and business performance.
When one commenter brought up a performance shift, binghott asked what else changed around the same time, including site or ad changes in the mid-March reply. The media-buying take is almost boring: he said the structure matters less than creative, product, offer, and site work in the creative-and-site reply.
Catalog and vertical caveats
The example account was clean because it had one main product, according to binghott’s original thread. In a later reply, he said multi-SKU can still work when products have similar prices, but different margins may push the setup toward ROAS bidding or separate conversion events by SKU or category in his multi-SKU reply.
The vertical was financial services, according to his short reply. In the same multi-SKU reply, he added a human-error counterweight to the “trust Meta” argument: in one week, he said he saw experienced buyers label a campaign as targeting existing customers while it was excluding them, and label ad sets as Incremental Attribution while they were actually running 7dc1dv in the same reply.