Bid suppression is the Smart Bidding behavior in which Google held bids below what your target allowed on budget-limited campaigns, so those campaigns won cheaper clicks and beat their Target CPA or Target ROAS. As of late August 2026, that behavior appears to be largely gone. Reporting the effect after the rollout completed, Search Engine Land wrote that Google's Smart Bidding update "appears to have ended bid suppression" — the mechanic that previously let budget-constrained campaigns secure cheaper clicks and out-perform their stated targets.
This is the downstream, data-in-hand sequel to a change B2B SaaS advertisers were warned about months earlier. We covered the announcement in Google Ads Bidding Target Optimization: Aug 17 2026, written before the rollout when the effect was still a forecast. Now the update has shipped — it began August 17 and finished its global rollout on August 27 — and the observed behavior confirms the concern: campaigns that were quietly over-delivering are being pulled up toward target. This post explains what bid suppression was, what the post-rollout data shows, why it matters most for lean SaaS accounts, and the exact audit to run now that it is live.
What bid suppression actually was
Bid suppression was never a feature or a setting. It was an emergent property of how Smart Bidding rationed a capped budget. When a target-based campaign was limited by budget, the system faced a trade-off: bid aggressively to hit the exact target on fewer auctions, or bid more conservatively to stretch the same money across more auctions. In practice it often did the latter, holding bids down so the campaign captured more conversions than a strict reading of the target would predict — which showed up in your reports as a realized cost well under your Target CPA, or a return well above your Target ROAS.
For advertisers, that looked like a free lunch: a campaign with a $600 Target CPA delivering conversions at $400, month after month. It was not really free — it reflected the system leaving your target's headroom unused because the budget cap was the binding constraint, not the target. But it was comfortable, and many SaaS accounts came to plan around those suppressed costs as if they were the true price of acquisition. That is precisely the assumption the August change invalidates. Understanding the mechanic matters because the same logic governs every target-based strategy you run, a topic we unpack in our guide to bidding strategies for B2B SaaS.
What Google changed on August 17
Google's stated change was narrow and specific: starting August 17, 2026, budget-limited campaigns using Target CPA or Target ROAS would, in Google's words, more consistently perform toward the bid target. The rollout was automatic — not opt-in — and applied across Search, Shopping, Performance Max, Demand Gen and Travel. The company framed it as a predictability improvement: delivery that lands closer to the number you set rather than scattering around it. It completed its global rollout on August 27, 2026, so by September the affected campaign population had been running under the new behavior for several weeks.
The phrase "more consistently perform toward your bid target" is doing quiet work. For a campaign that had been missing its target, the change might tighten delivery downward. But for the large group that had been beating its target through suppression, "toward target" means upward — the realized cost per conversion rises to meet the target you set. Google's own help documentation on changes to target-based bid strategies describes the intent as more predictable performance in line with your targets. The post-rollout reporting simply named the casualty: the cheaper clicks that suppression used to deliver.
Why this hits lean SaaS accounts hardest
The campaigns most exposed to the end of bid suppression are exactly the ones typical of B2B SaaS: budget-capped, running on automated target-based bidding, and quietly efficient. A seed-stage or growth-stage SaaS team rarely runs uncapped budgets — they set a monthly spend they can defend and let Smart Bidding work within it. That is the definition of a budget-limited campaign, which is the sole population this change touches. If your acquisition campaigns were comfortably under target, you were benefiting from suppression whether you knew the term or not.
The risk is not that anything broke — it is that your cost per lead or cost per SQL can climb without any visible cause in the account. You did not change a bid, a keyword, or a budget, yet September's numbers look worse than July's. That is a dangerous kind of drift because it is easy to misattribute to seasonality, competition, or creative fatigue. If you measure acquisition seriously — and for SaaS the metric that matters is cost per qualified pipeline, not cost per form fill, as we argue in cost per lead vs cost per SQL — you need to know whether a CAC increase came from this change or from the market, because the two demand different responses.
The post-rollout audit to run now
Because the change is already live, this is a diagnosis exercise, not a pre-deadline one. Start by isolating the affected population: every campaign that was limited by budget in recent months and uses Target CPA or Target ROAS. Those are the only campaigns bid suppression applied to, so they are the only ones where its end can move your numbers. Everything uncapped or on a non-target strategy can be set aside for this specific review.
For each affected campaign, pull realized CPA or ROAS for a window before August 17 and a matched window after August 27, and look for unit cost that climbed toward the stated target while conversion volume held or rose. That signature — steady or higher volume at a higher price per conversion — is the fingerprint of suppression ending, distinct from a demand drop (which shows falling volume) or a tracking break (which shows falling conversions). Segment by device and match type if you can, since suppression tended to be most generous on the cheaper corners of the auction. Fold this comparison into your recurring review using our Google Ads optimization checklist so it becomes a repeatable check rather than a one-off.
Resetting targets to real economics
The durable fix is the same discipline that would have insulated you from this change in the first place: a Target CPA or Target ROAS that encodes your actual unit economics rather than a round number that felt safe. Your target should fall out of the math — customer lifetime value, the lead-to-customer conversion rate, and the gross margin you need to protect — so that when the system steers delivery toward it, it is steering toward a ceiling you genuinely accept. When the target is honest, suppression ending is a non-event: the campaign was already priced at your true ceiling. When the target is loose, the system now spends into the slack, which is why a stale target is more expensive in September 2026 than it was in July.
For SaaS specifically, the target you feed the system is only as good as the conversion values behind it. A closed customer is worth far more than a trial signup, and if you bid to a flat cost-per-conversion target while treating every conversion as equal, you will systematically misprice your best traffic — a problem the end of suppression amplifies rather than creates. This is the case for value-based bidding over flat tCPA, which we lay out in Maximize Conversion Value vs Target CPA. If your offline and CRM data are wired correctly, the system can bid to the revenue a click eventually produces rather than the count of clicks, which is a far more defensible target to steer toward.
Budget increases versus target changes
Because suppression only occurred on budget-limited campaigns, one honest response is to remove the budget constraint rather than move the target. If a campaign was over-delivering purely because the cap forced conservative bidding, raising the budget lets it capture the incremental volume that suppression was previously funding through cheap clicks — and for a genuinely volume-constrained, profitable campaign, that can be the right call. But it only works when the target is sound. Adding budget to a campaign with a loose target just spends more money toward a number that does not reflect your economics, which compounds the drift instead of fixing it.
The correct sequence is target first, budget second. Reset the target so it encodes your true CAC ceiling, confirm the campaign is still profitable at incremental volume, and only then decide whether it earns more budget. This is also the moment to revisit how much you are spending overall against what the channel can absorb, which we frame in how much you should spend on Google Ads. Treating the end of bid suppression as a prompt to re-derive both your targets and your budgets — rather than a fire to put out — is how you turn a silent CAC risk into a deliberate, defensible acquisition plan.
Your action plan
Put concretely: first, list every budget-limited campaign on Target CPA or Target ROAS, because those are the only ones bid suppression touched. Second, compare realized CPA or ROAS before August 17 against the weeks after August 27, and flag campaigns whose unit cost rose while volume held — the fingerprint of suppression ending rather than a demand or tracking problem. Third, for each flagged campaign, decide whether the stated target still equals the CAC you are willing to pay, and tighten any that were set loosely so the system steers toward your real ceiling.
Fourth, verify the inputs the whole system depends on — that conversion tracking is firing cleanly and conversion values reflect real economics — because tightened delivery built on bad data is worse than loose delivery. Fifth, document every target you change with the date and rationale, so when you compare Q4 performance you can attribute shifts to the right cause. Most of this is a few focused hours for a typical SaaS account, far cheaper than discovering in November that your blended CAC drifted and not knowing why. If you would rather have a second set of eyes on which targets to reset, our Google Ads audit covers exactly this kind of target and tracking review.