Why does platform-reported ROAS lie for long sales cycles?
Meta and Google attribution windows max out around 28 days for most accounts, sometimes less depending on campaign settings. A consultant selling a $15,000 engagement, a franchise selling a $50,000 territory, or an immigration firm signing a retainer for an EB-5 case doesn't close in 28 days. The deal is still sitting in someone's inbox on day 40, still waiting on a decision-maker on day 65, and by the time it signs, the platform has already closed the attribution window and quietly reported that ad spend as a loss.
The dashboard doesn't know the deal exists. It only knows what happened inside the window it was built to measure, and high-ticket sales cycles routinely run three times longer than that window. Nobody adjusts the dashboard for that. It just keeps reporting a loss on a campaign that's about to produce the biggest deal of the quarter.
This is worse than it sounds because it compounds. A business that trusts the dashboard month after month is training itself to distrust its own best-performing channels, simply because those channels' wins arrive too late for the platform to see.
What is signed-revenue attribution, and how is it different from platform ROAS?
Signed-revenue attribution ties actual signed revenue, not clicks or form fills, back to the specific campaign, ad set, and creative that originated the lead. It requires a CRM connection or a manual close-loop process where every signed deal gets tagged with its source before it counts toward performance reporting. The number that comes out the other end isn't "clicks that converted." It's revenue that closed, mapped to the spend that produced it, regardless of how many weeks sat between the two events.
This only works if the tracking survives the full sales cycle. A lead source tag that gets lost somewhere between the ad and the CRM, or a sales team that doesn't log source consistently on every call, breaks the chain, and the whole exercise collapses back into guessing dressed up as data.
Platform ROAS also blends every campaign into one number by default. Signed-revenue attribution forces separation by campaign, which is the only way to know that one ad set is quietly funding the whole account while three others are burning budget on leads that never had a real shot at closing.
Does a longer sales cycle mean attribution is impossible?
No, it means attribution has to run on a longer clock than the platform's default. Primary Pond, a client running long-cycle lead generation, produced 2,169 leads at $1.87 cost per lead, and those leads generated $589,762 in revenue against $404,251 in spend, a 1.46x return. None of that return was visible inside a 28-day platform attribution window. It only became visible once revenue got traced back through the CRM to the original campaign, months after the leads first came in.
That's the mechanism, not the pitch. The point isn't "we tracked things carefully." The point is that without tracing revenue past the platform's window, that 1.46x return would have looked like a straight loss for two consecutive months while it was actually building toward a win the whole time.
A return that only became visible once revenue was traced past the platform window.
Why do immigration and franchise clients need this more than most industries?
Both run sales cycles that routinely blow past standard attribution windows, and both sell something expensive enough that a wrong attribution call carries real budget consequences. An immigration firm's cost per signed case dropped from $2,372 to $1,064 for one client, verified against 1,391 signed cases, but that number only exists because signed-case data got traced back to case type and campaign. The platform never reported it, because the platform doesn't know what a signed case is.
Franchise recruitment runs on a similarly long clock, often 90 days or more from first inquiry to signed franchisee, which puts the whole cycle well outside what any ad platform natively measures. A prospect who registers interest in month one might not sign a franchise agreement until month four, after due diligence, financing conversations, and a site visit. The platform lost interest in that lead in week four.
Shorter sales cycles can survive on platform-reported numbers because the entire cycle happens inside the attribution window. High-ticket doesn't get that luxury. The deal is still moving through legal review, board approval, or a franchisee's own due diligence long after the platform has stopped counting anything.
What breaks when a business trusts platform ROAS for a 60 to 90 day cycle?
Budget gets pulled from the campaigns that are actually working, because the campaign that produced this month's signed deals ran two or three months ago and looks dead in the dashboard today. Meanwhile, a campaign that produced a burst of cheap, unqualified leads this week looks like a winner in the platform, because its clicks are still inside the window even though none of those leads will ever close.
This is how businesses end up killing their best channel and scaling their worst one, purely because one gets measured on a clock that fits its sales cycle and the other doesn't. It usually takes a full quarter of declining signed revenue before anyone connects the dip back to a budget decision made three months earlier, based on numbers that were never true.
This is how businesses kill their best channel and scale their worst one, purely because one gets measured on a clock that fits its sales cycle.
How do you build a profit-traceable attribution system without a huge tech stack?
Start with a single source field on every lead record, populated at the moment of first contact, not edited later by whoever closes the deal. Connect that field to whatever CRM tracks the deal through to signed. Report on cohorts by the month the lead came in, not the month revenue landed, so a deal that closes in month three still gets credited to the campaign that sourced it in month one.
None of this requires enterprise attribution software or a data team. It requires a source field that doesn't get lost, a sales team that logs consistently, and a reporting cadence that looks back far enough to match the actual sales cycle instead of the platform's default window. Most businesses already own the CRM they need. What's usually missing is the discipline to keep the source tag intact for 60 to 90 days without someone overwriting it halfway through.
What should replace ROAS as the number leadership actually watches?
Cost per signed deal, traced to campaign, reported on a delay that matches the sales cycle. Not cost per lead, not platform ROAS, not a blended return across every campaign averaged together into one comforting figure. A number that answers one question: for the money spent this month, how much signed revenue eventually came back, and from which campaign.
That's the only number that can't be gamed by a campaign generating cheap clicks with no real path to a signed deal. It's slower to report than a platform dashboard, because it has to wait for the sales cycle to actually finish before it means anything. It's also the only one telling the truth about where the profit came from.