Meta Killed 45-Day Float on April 1 and Agencies Forgot to Reprice
Meta required higher-spending advertisers to switch from credit cards to monthly invoicing or direct debit on April 1, 2026, eliminating the 45 to 60 day card float that agencies had quietly built into their working capital. Most retainer contracts haven't repriced. The agencies absorbing the timing gap on $50,000-plus monthly client spend are now extending Meta a working-capital loan their clients still aren't paying for.
The official line from Meta, via Payments Dive, called this an update affecting "a very small percentage of advertisers." The actual scope, per industry coverage, is accounts spending over roughly $50,000 a month, which is most independent agency books and a meaningful chunk of brand DTC. The deadline to switch was March 31, 2026, and Meta locked billing settings between March 30 and April 4 during the rollout.
The math agencies are quietly eating
Cards gave agencies two financial things they rarely separated in their P&L. The first is float, which is the gap between when ad spend hits the campaign and when the agency actually parts with the cash. The second is rewards, which are typically 2 to 3 percent cash back, and stacked across business cards can hit closer to 4 to 4.5 percent according to EcomCrew's coverage of the operator response.
Run the numbers on a $200,000 monthly Meta book, which is a normal-size mid-market agency client. The card gave you somewhere between 21 and 30 days of float on the statement closing date, then another 21 to 25 days before payment was due. So $200K in spend was sitting interest-free in your operating account for about six weeks. Pull that down to a 30-day invoice from Meta, and that working capital cushion compresses to roughly half. On a four-client agency at that spend level, you're looking at $400K to $500K of cash flow that used to be free, that now has to come from somewhere.
The rewards loss is more visible because it shows up on a statement. At $50,000 of monthly spend, a stacked rewards setup pulled in $12,000 to $18,000 a year per the same EcomCrew piece. Multiply that across a multi-client book and it stops being noise.
Why most retainers missed this entirely
A retainer signed in 2024 or 2025 was priced against an implicit assumption: the agency could either run its own card and earn the float and rewards, or have the client run a card and skip the cash flow exposure entirely. The economics of the first option were good enough that a lot of agencies, particularly the ones building toward acquisition, ran spend on agency-owned cards as a way to juice working capital and effective margin.
That's the model Meta just deprecated for anyone over the threshold. And per AdAmigo's agency playbook, the recommended new structure is client-owned billing, where the client takes on the Net 30 invoice directly and the agency only gets Partner access to manage. That's a sensible legal posture. It's also a quiet retainer event the agency probably hasn't announced internally. The cash benefit that was implicitly priced into the management fee no longer exists, and the contract didn't change.
Cedric Yarish, founder of AdManage.ai, told AdAmigo that "if a client ever feels like 'this agency owns our business,' they panic," which is the legitimate reason most agencies are sliding to client-owned billing. The reason they're not also raising the management fee at the same time is harder to defend. From what I've seen, it's usually that nobody on the account team actually sees the credit card statement, so the float gain never made it into anyone's mental model of the unit economics.
The hidden tax on Q4 inventory plans
The other thing the float quietly funded was Q4. A DTC brand on agency-owned billing could push a 30 to 50 percent budget bump into Black Friday week and not feel the cash hit until late January, because the agency card statement gave them another four to six weeks of breathing room on top of whatever the agency offered as Net terms. That stack is gone for everyone above the threshold.
Per Meta's billing rules under invoicing, payments are due 30 days from invoice date. That means a Black Friday push gets invoiced in early December and is due in early January, which is exactly when most retailers are running tight on cash from holiday inventory write-downs and refund volume. From what I've seen modeled out, this alone shifts the realistic Q4 spend ceiling for agency-managed accounts by 10 to 15 percent unless the brand has a credit facility ready.
It also changes who eats a bad week. On a card, a $400K spend week that didn't convert was an interest-free 45-day window to fix the funnel. On Net 30 invoicing, the same week is real cash going out the door three weeks later, regardless of whether the campaign worked.
The retainer repricing window is right now
The honest move is to model two numbers per client and bring them to the next quarterly review.
First, the lost cash benefit, which is float plus rewards. On agency-owned billing pre-April, a $200K-spend client was generating roughly $4,000 to $6,000 a year in rewards plus the soft benefit of the float. That benefit just became zero. Whether the retainer covers that or not is a real conversation, and it lands cleaner if you frame it as a billing-model change Meta forced rather than an ask for more money.
Second, the new working capital exposure if you're keeping any spend on agency-owned billing. A 30-day Net invoice from Meta against client retainers that pay Net 30 or Net 45 is a structural mismatch. Origin Web Studios told the same playbook that they cut payment issues 80 percent by moving to milestone-based triggers and front-loaded deposits, and that's a reasonable template. A 20 to 50 percent ad spend deposit from new clients is normal practice in 2026 and was unusual in 2024.
What I'd actually flag this week
If you're an agency operator with any client over $50,000 a month in Meta spend, the cleanest version of this conversation is: Meta changed billing terms April 1, the previous structure passed a hidden cash flow benefit to us that we never billed for, and we're proposing one of two adjustments. Option one, you take the Meta invoice directly and we keep the management fee flat. Option two, we keep agency-owned billing and add a 1 to 1.5 percent working capital fee on managed Meta spend.
Most clients pick option one once they see the numbers, which is fine. The point is to actually have the conversation. If you don't, you've absorbed the change without negotiating it, and the retainer math you signed against doesn't exist anymore.
For brands on the other side of this, the same change creates a clean opportunity to push for transparency clauses your agency probably doesn't want. We covered the markup transparency angle in our earlier piece on Meta's 21-month markup disclosure deadline, and the timing of the two events isn't a coincidence. Meta is gradually pulling agencies out of the cash flow chain, both by removing the float and by mandating spend disclosure on the invoice itself.
The category I'd watch over the next two quarters is mid-market agencies with thin balance sheets. The ones that quietly funded their growth by stacking cards on a few large clients are the ones with the least margin to absorb a structural change to how Meta gets paid. Some of them won't reprice, and a meaningful percentage of those probably get acquired or fold by year-end. Not because the float change is fatal on its own, but because it's the kind of change that exposes a model that was already running close to the line.
Notice Me Senpai Editorial
