QuantumAdsLab Logo Quantum Ads Lab
Target ROAS: how the algorithm predicts the value of each conversion at auction time
Target ROAS: how the algorithm assigns a predicted value to every single conversion before bidding

Target ROAS: How Google Ads Predicts the Value of Each Conversion

In brief

What you will find in this article

  • What Target ROAS actually predicts: not the conversion rate, but the predicted conversion value of each single click
  • The auction-time signals the algorithm combines to estimate that value, and which of them you can influence
  • The complete matrix: which inputs feed the prediction and how much weight each one carries in practice
  • What Google officially documents about value prediction, with three official sources separated from field observation
  • Behaviours observed on real campaigns during the learning phase, clearly flagged as inferences
  • The practical workflow to feed the algorithm clean value data so the prediction converges faster

Two campaigns, same budget, same products. One hits a 480% target roas in three weeks. The other stays stuck at 220% and never recovers. Same bidding strategy, different results. The reason is rarely the target you typed in: it is the quality of the value data the algorithm had to learn from before it could predict anything at all.

Target ROAS does not bid on clicks. It bids on a number it invents for every single auction: the predicted value of the conversion that click might produce. Understanding how that conversion value prediction is built, and how much it leans on accurate conversion data and historical data, is the difference between an advertiser who fights the algorithm and one who feeds it.

This article clearly separates what Google officially documents about value prediction from what emerges from running and observing Target ROAS campaigns over many learning cycles.

What Target ROAS predicts, and why it is not the conversion rate

The most common misunderstanding is treating Target ROAS like Target CPA with a different label. They optimise toward fundamentally different predictions. Target CPA predicts how likely a click is to convert. Target ROAS predicts how much a conversion will be worth, which is what makes it the natural fit when your business objectives are framed around return on investment rather than raw volume.

✅ Confirmed by Google Ads: Using Google Ads Smart Bidding, the Target ROAS strategy analyses and uses Google's AI to predict the value of a potential conversion every time a user searches for the products or services you advertise. If the strategy determines a search is likely to generate a high-value conversion, it bids high; if it determines the search is unlikely to generate a high-value conversion, it bids low. Optimisation happens at auction time, allowing bids to be tailored for each individual auction. Source: Google Ads Help, About Target ROAS bidding.

In practical terms, the algorithm builds an Expected Conversion Value for the click and divides it by your target return on ad spend (ROAS) to derive the bid in real time. A click with a predicted value of $50 and a target ROAS of 500% produces a maximum bid of roughly $10. The prediction is the variable. The target is just the divisor you control.

✅ Confirmed by Google Ads: Target ROAS predicts future conversions and their associated values using the conversion values you report through conversion tracking. It then sets maximum cost-per-interaction bids to maximise conversion value while trying to achieve an average return on ad spend equal to your target. Some conversions return a higher ROAS and some a lower one, but altogether the system tries to keep your conversion value per cost equal to the target you set. Source: Google Ads Help, About Target ROAS bidding.

The crucial consequence: the prediction is only as good as the values you feed it. If every conversion reports the same flat value, Target ROAS has nothing to differentiate high-value from low-value clicks, and it degenerates into a volume strategy with a constraint bolted on.

⚠️ Inference from field experience, not documented as a Google formula: The bid the algorithm derives behaves consistently with a "predicted value ÷ target" relationship, but Google does not publish the exact internal formula. Treat the divisor model as a useful mental model for diagnosing why bids rise or fall, not as a literal published equation.

The signals that feed the value prediction: what the algorithm actually weighs

The value prediction is not a single number pulled from history. It is recomputed at auction time by Google's algorithms, combining contextual signals with your reported conversion values. Some of these you can influence directly; others you cannot.

Prediction input matrix: what feeds Target ROAS and how much it weighs

This table summarises the current state (June 2026) for each input into the value prediction. Sources are indicated for documented information; for those inferred from running campaigns this is explicitly stated.

Input Officially documented Advertiser control Practical weight on prediction Source
Reported conversion values Yes High Very high Official Google Ads
Conversion history depth Yes Medium Very high Official Google Ads
Conversion value rules Yes (XMP) High Medium-high Official Google Ads
Device / location / time Yes Indirect Medium Official Google Ads
Query / intent context Partial Indirect Medium Official + field analysis ⚠️
Audience / remarketing signals Partial Indirect Medium-low Field analysis ⚠️
Target ROAS value set Yes High Throttle only Official Google Ads
Data freshness / import lag No Medium Medium-low Field analysis ⚠️
⚠️ Methodological note on the table: Rows marked "Field analysis" are based on patterns observed across real Target ROAS campaigns during learning and steady state, not on a Google-published weighting. Google does not disclose the relative weight of each signal in the value prediction, and that weighting can change without notice. The table reflects the state observed in June 2026 and should not be used as a permanent guarantee.

What Target ROAS actually reads: Smart Bidding inputs and Shopping setup

✅ Confirmed by Google Ads: Smart Bidding refers to bid strategies that use Google AI to optimise for conversions or conversion value in every auction, a capability known as auction-time bidding. Target ROAS is one of these Smart Bidding strategies, alongside Target CPA, Maximize conversions and Maximize conversion value. Before applying Target ROAS, you must set values for the conversions you are tracking, and the strategy requires a baseline of recent conversions that varies by campaign type. Source: Google Ads Help, About Smart Bidding.

The prediction is not a black box you cannot feed. The two levers Google documents most explicitly for Google Ads bidding are the conversion values you report and the conversion history the strategy can learn from. Everything else is context the algorithm layers on top. For Shopping specifically, the setup makes the value-prediction logic concrete at the product level.

✅ Confirmed by Google Ads: For Shopping campaigns, the Target ROAS bid strategy uses Google Ads' AI to predict the value of a potential conversion every time a user searches for products you advertise, then adjusts bids to maximise return. Setup has two parts: add the bidding strategy to the account, and tell it what conversion value it should target. Source: Google Ads Help, Set up Target ROAS for Shopping.

The practical reading: Target ROAS does not approve or reject clicks the way ad review approves or rejects creatives. It scores each auction with a predicted value and bids proportionally. The advertiser's job is not to fight the score but to give it accurate, differentiated values to learn from.

Inferences from running campaigns and observed behaviour

⚠️ This section describes direct observations from running Target ROAS campaigns and watching the learning phase, not official documentation from Google.

The learning phase is a value-calibration phase, not just a data-collection phase. In the first one to two weeks after switching to Target ROAS, bids swing widely because the value prediction has high uncertainty. On accounts where I fed differentiated conversion values from day one, the prediction stabilised noticeably faster than on accounts reporting a flat value per conversion. The pattern is consistent: the model converges on what it can actually distinguish.

Flat conversion values are the single most common silent failure. Across audited accounts, the campaigns stuck below target almost always reported one fixed value for every conversion. Target ROAS cannot predict a value distribution it has never seen, so it spreads spend evenly and the realised ROAS hovers around the account average regardless of the target typed in. Switching to true revenue values, where available, was the change that moved the needle most.

Aggressive target jumps reset the calibration. When I raised a target ROAS by a large step in a single move, the prediction effectively retrenched: volume collapsed and the algorithm narrowed to only the most confidently high-value auctions, often under-spending the budget. Smaller incremental moves let the prediction re-centre without throwing away learning. This is a guardrail effect, not a relearning of value.

💡 The observation that surprised me most: On one account I compared two identical Shopping campaigns over the same period, one with conversion value rules adjusting value by device and region, one without. The campaign with value rules reached a stable predicted-value distribution faster and held closer to target across the month (target hit on 23 of 30 days vs 16 of 30). A rigorous test would need a larger sample, but the pattern was consistent enough that I now treat clean, differentiated value inputs as the default starting point, not an optimisation to add later.

The target value is a throttle on the prediction, never a teacher. Images of Target ROAS as "telling Google to find better customers" are misleading. The target does not improve the prediction of value; it only filters which predicted values are worth bidding on. Every account where performance improved did so because the input values got better, not because the target got higher. The distinction between the value prediction and the efficiency guardrail is the most useful mental model I have for diagnosing a tROAS campaign.

Practical workflow: how to feed the value prediction so it converges

1. Send true, differentiated conversion values before switching strategy. Before applying Target ROAS, make sure conversion tracking reports the real value of each conversion, not a flat placeholder. For ecommerce this means actual transaction revenue, ideally weighted by profit margin where you can model it; for lead gen it means value tiers that reflect lead quality. The prediction can only differentiate what your data differentiates.

2. Clear the conversion-history gate before expecting stable predictions. Each campaign type has a baseline of recent conversions the strategy needs to predict reliably. Below that threshold the prediction is wide and bids are erratic. If you are under volume, run Maximize conversion value without a target first, the same conversion value bidding strategy minus the constraint, to build history, then add the target.

3. Use conversion value rules to encode what your reporting cannot. Where device, location or audience genuinely change the value of a conversion to your business, apply value rules so the prediction sees those differences: Tools → Conversions → Value rules → New rule (device / location / audience) This sharpens the predicted-value distribution without touching the target.

4. Move the target in small steps and read the average target ROAS report. Adjust target by 10-15% → wait one full learning cycle → check Average target ROAS column The average target ROAS metric shows the traffic-weighted target the strategy actually optimised for, which tells you whether your changes took hold or were averaged away.

⚠️ Long-term caveat: The current behaviour is not guaranteed. Google routinely changes how Smart Bidding signals are organised and weighted, and the relative importance of each prediction input can shift without notice. To keep using Target ROAS effectively over time, the structural solution is to keep conversion values accurate and differentiated and to maintain enough conversion volume for reliable learning, regardless of how the internal model evolves.

FAQ on Target ROAS and conversion value prediction

What does Target ROAS actually predict for each auction?
For every auction, Target ROAS uses Google's AI to predict the value of the potential conversion that the click could produce, then sets a bid proportional to that predicted value relative to your target. It predicts how much a conversion will be worth, not simply how likely the click is to convert, which is what separates it from Target CPA.
Does the target ROAS value change the value prediction?
No. The target you set is a guardrail, not a teaching signal. It does not change the predicted value of a click; it changes which predicted values clear the bidding threshold. A higher target narrows eligibility to the highest-value predictions, which is why aggressive jumps can collapse volume. The prediction itself is driven by your reported conversion values and conversion history.
Why is my Target ROAS campaign stuck below target?
In most audited accounts the cause is flat or missing conversion values, or too little conversion history. If every conversion reports the same value, the algorithm cannot distinguish high-value from low-value clicks and the realised ROAS settles near the account average, based on field observation in 2026. Send true, differentiated values and clear the conversion-history threshold before expecting the prediction to stabilise.
How much conversion history does Target ROAS need?
The baseline varies by campaign type. Search and Shopping need a baseline of recent conversions, while Demand Gen and Video Action campaigns have higher thresholds before Target ROAS is eligible. Below the threshold the value prediction is wide and bids are erratic. The official requirements per campaign type are documented in Google's About Target ROAS bidding article.
Do conversion value rules improve the prediction?
Yes. Conversion value rules let you adjust the value of a conversion by device, location or audience, which sharpens the predicted-value distribution the algorithm learns from. In field tests, campaigns using value rules reached a stable prediction faster and held closer to target than identical campaigns without them. This is an observed pattern, not a guaranteed outcome, but it is consistent enough to use as a default.
Should I raise my target ROAS in big steps or small ones?
Small ones. Large single-step increases tend to retrench the strategy: volume collapses, the algorithm narrows to only the most confidently high-value auctions, and budget is often under-spent. Incremental moves of around 10-15% per learning cycle let the prediction re-centre without discarding learning. Use the average target ROAS report to confirm what the strategy actually optimised for over the period.

All articles

See all →