Your tROAS campaign is hitting 400%. The dashboard looks healthy. Finance is asking why pipeline doesn't match. This is the moment most marketing leaders blame the platform, but the platform isn't lying. It's following the instructions you gave it.
Value inflation is what happens when the dollar figure Smart Bidding optimizes toward doesn't reflect what the business actually collects in revenue. It's not fraud. It's not a Google bug. It's almost always a self-inflicted tracking issue that compounds over time, and it matters more now than it did three years ago because tCPA and tROAS accept whatever value you set as truth. If the numbers are inflated, the algorithm will aggressively pursue more inflated numbers, chasing ghost conversions, raising CPCs to win auctions for "high value" clicks that were never actually high value, and burning budget in the exact direction it was instructed to go.
The account looks like it's performing well on the surface. That's the problem.
Where the Drift Starts
Value inflation rarely shows up as one glaring error. It's usually three or four small distortions stacked on top of each other, each one nudging the algorithm a little further from reality. A recent Search Engine Journal analysis identified the most common culprits: double-counted micro-conversions (a "form submit" and a "thank you page view" both firing as separate conversions for the same lead), mis-weighted primary versus secondary goals, inflated offline conversion imports, and conversion value rules created years ago and never revisited.
The double-counting problem is particularly insidious. If your form submission and your thank-you page view both carry their own value, you've just doubled your reported conversion volume for a single lead. Smart Bidding doesn't know the difference. It sees two high-value signals and bids accordingly.
The primary versus secondary goal confusion is equally damaging. Google's own documentation is clear: primary actions are used for bidding optimization, secondary actions are for observation only. But in practice, newsletter signups, PDF downloads, and add-to-carts often get marked as primary conversions alongside actual purchases. The bid signal gets diluted with noise that has nothing to do with revenue.
The Offline Import Trap
For B2B marketers, offline conversion imports are where the real damage accumulates. The workflow sounds reasonable: upload deal values from your CRM so Smart Bidding can optimize toward actual revenue. The execution is where things break.
Google's enhanced conversions for leads uses hashed user data to match offline conversions back to ad clicks. The system works well when the data is clean. The problem is that most CRM data isn't clean. Deal values get uploaded before deals close. Pipeline values get treated as revenue. A $50,000 opportunity that eventually closes at $12,000 (or doesn't close at all) still sits in the conversion data at its original inflated value.
The lag compounds the problem. If your sales cycle is 90 days and you're uploading deal values at the opportunity stage, Smart Bidding is optimizing toward a signal that won't be validated for three months. By the time you discover the mismatch, the algorithm has already learned the wrong lesson.
The Audit Protocol
Before trusting a single tROAS target in an unverified account, run this sequence:
Step one: Map every conversion action to a revenue event. Pull your conversion actions list and ask a simple question for each one: does this action directly generate revenue, or is it a proxy? If it's a proxy, it should be secondary, not primary. Best practice guidance is explicit: primary conversions should be reserved for actions that directly correlate with revenue, while secondary conversions handle the observation-only signals.
Step two: Check for duplicate firing. Look at your conversion paths for the same user. If you see multiple conversion events within seconds of each other (form submit, thank-you page, confirmation email trigger), you likely have duplicate counting. One lead should equal one conversion, not three.
Step three: Audit your offline import values. Pull the last 90 days of offline conversions and compare the uploaded values against actual closed revenue. If the gap is more than 15%, your Smart Bidding is optimizing toward phantom value. The fix is either to upload only closed-won values or to apply a discount factor that reflects your historical close rate.
Step four: Review conversion value rules. Conversion value rules let you adjust values based on audience, location, and device. They're powerful when maintained, dangerous when forgotten. If you created rules in 2023 to weight enterprise leads higher than SMB leads, check whether those weights still reflect reality. Business mix changes. Value rules should change with it.
The Recalibration
Once you've identified the inflation sources, the fix is straightforward but requires discipline. Demote inflated actions to secondary status. Adjust offline import values to reflect actual close rates. Update or delete stale value rules. Then reset your tROAS target based on the corrected data.
The reset will feel painful. Your reported ROAS will drop, sometimes significantly. This is the point where many teams panic and revert. Don't. The lower number is the real number. The algorithm will now optimize toward actual revenue instead of phantom value, which means your CPCs may drop, your auction selection will improve, and your pipeline-to-spend ratio will start to make sense.
Current Smart Bidding guidance recommends at least 15 conversions in the past 30 days for tROAS to function properly. If your corrected conversion volume falls below that threshold, you may need to consolidate campaigns or temporarily shift to a maximize conversions strategy while you rebuild the signal.
The Governance Layer
Value inflation is a recurring problem, not a one-time fix. Build a quarterly audit into your operating rhythm. The checklist is simple: conversion action inventory, duplicate firing check, offline import variance analysis, value rule review. Thirty minutes per quarter prevents the drift that turns a healthy account into a phantom-value machine.
The CFO will eventually ask why reported ROAS doesn't match actual revenue. You can either explain the discrepancy after the fact, or you can fix the tracking before the question gets asked. One of those conversations is significantly easier than the other.
