- Sep 1, 2023
- 152
- 52
Should fresh Google Ads accounts start with tCPA or tROAS right away, or is it still better to collect data first and switch later? There doesn't seem to be a clear consensus.
Some buyers still follow the traditional progression: Manual CPC or Max Clicks first, then Maximize Conversions, and only after that tCPA or tROAS. Others argue that this approach forces Google to learn from clicks instead of conversions and delays the point where the algorithm can start optimizing for actual business goals.
One Reddit user who claims to have managed more than £3 million in Google and Bing ad spend said he had tested both approaches across dozens of accounts. According to him, launching with Maximize Conversions and a tCPA target often produced better results than starting with click-based campaigns first.
The logic is easy to understand. If the end goal is conversions, why spend money training Google's systems on traffic that may never turn into leads in the first place?
At the same time, not everyone in those discussions had the same experience.
Several buyers reported that fresh accounts launched directly on tCPA either struggled to spend or pushed CPCs much higher than expected. In many of those cases, the issue wasn't necessarily the bidding strategy itself. The account simply didn't have enough information for Google to make confident decisions.
Most practitioners agree that tCPA and tROAS work best when Google already understands what a successful user looks like. The challenge is that a fresh account doesn't have that context yet. There are no historical conversions, no proven search terms, and no established patterns connecting keywords, ads, landing pages, and revenue. As a result, Google's algorithms often start from a position of uncertainty.
With tCPA, the system is trying to find users who are likely to convert around a specific acquisition cost. On an account with existing conversion history, Google can compare new users to people who converted in the past. On a brand-new account, it has far fewer signals to work with.
Things get more complicated once the conversation shifts to tROAS. Several PPC practitioners in Reddit discussions pointed out that Google can only optimize for return on ad spend if it understands which conversions actually generate revenue. That's relatively straightforward in ecommerce, where every purchase already has a value attached to it. In affiliate verticals, the picture is often much less clear.
One specialist described a setup that combines CRM data, Enhanced Conversions, and Offline Conversion Imports. The conversion is recorded in Google Ads first, but the real value gets assigned later, after the lead has been processed. That additional feedback helps Google distinguish between a lead the buyer simply considers promising and assigns a value of around $20, a user who deposits $100, and a customer who goes on to generate $500 in rebills within a week. Without those signals, all three conversions can look exactly the same inside the ad account.
That's why many buyers treat tROAS as something that comes after a campaign has already proven it can generate consistent conversions. Before Google can optimize for revenue, it first needs enough information to understand where that revenue is actually coming from.
This is also where account quality starts to matter.
Several discussions around smart bidding focus entirely on campaign settings, but campaign settings are only part of the equation. Before Google can optimize efficiently, it needs enough conversion volume to identify patterns. If an account is heavily restricted, spends inconsistently, or runs into operational issues before enough data is collected, the learning process can take much longer.
Agency accounts often have an advantage here. Higher spending limits, stronger account trust, and smoother moderation make it easier to reach the conversion thresholds that automated bidding strategies depend on.
The exact numbers vary depending on who you ask, but the same benchmark appears surprisingly often. Around 30 conversions in a 30-day period is commonly mentioned as a reasonable starting point for tCPA, while tROAS is usually associated with higher data requirements. Some practitioners in Reddit discussions specifically mentioned 30+ conversions for tCPA and 50+ for tROAS before expecting either strategy to perform consistently.
Interestingly, that isn't very far from Google's own recommendations. The platform itself suggests having at least 30 conversions before relying heavily on a target CPA strategy. Still, most experienced buyers don't treat these numbers as hard rules.
The more important question is whether the campaign is generating stable and predictable results.
If leads have been coming in at roughly $70–80 for several weeks, search terms are under control, and the landing page isn't changing every other day, Google has enough context to work with. In that situation, setting a target close to the actual CPA usually produces better results than trying to force a dramatic improvement overnight.
One recurring theme in community discussions is that campaigns often stop spending after the switch because the target itself is unrealistic. A campaign that has never generated leads below $80 isn't likely to start producing them at $30 simply because a buyer enters a lower number into the settings. From Google's perspective, there is no historical evidence that such a target is achievable, so the system starts participating in fewer auctions while searching for opportunities that fit the goal.
One Reddit user suggested approaching the transition more gradually. If the desired CPA is $30, he preferred starting somewhere around $40–45 and lowering the target over time as the campaign continued generating conversions. His argument was that giving the algorithm some room to operate usually works better than immediately forcing it into a narrow range.
Another interesting point raised in the discussions is that there isn't a single "correct" launch sequence. Some buyers still prefer the traditional route of Manual CPC, then Max Clicks, then Maximize Conversions before touching tCPA. Others skip parts of that process entirely.
One practitioner who commented on a Reddit thread said he often launches fresh Search campaigns with Maximize Conversions from day one, provided that conversion tracking has already been tested and verified. He also recommended relying on native Google Ads conversion tracking rather than imported GA4 events whenever possible, arguing that cleaner signals tend to produce better optimization results.
The same user mentioned that keyword control becomes especially important during the early stages. His preferred setup combines Exact Match and Phrase Match keywords while aggressively building negative keyword lists as search term data comes in. Broad Match, in his experience, gives Google too much freedom when the account has little or no conversion history, often leading to expensive experiments that the advertiser ends up paying for.
That opinion wasn't unique. Several buyers in the discussion described a similar workflow: start with tightly controlled traffic, review search terms daily, expand winning queries into dedicated keywords, and continuously eliminate irrelevant intent. By the time tCPA enters the picture, the account already has a much clearer understanding of what good traffic looks like.
Several buyers in the discussions mentioned that they spend more time in the Search Terms report than anywhere else during the early stages of a campaign. The reasoning is simple: before Google can learn which users convert, the advertiser needs to make sure the campaign is attracting the right intent in the first place.
A search query that produces a conversion often gets promoted into its own Exact Match keyword. Queries that look promising but haven't generated enough data yet stay under observation. Anything clearly unrelated gets added to the negative keyword list. It's not particularly exciting work, but it's one of the fastest ways to improve signal quality before introducing stricter automation.
This is also where Maximize Conversions starts making sense for many advertisers. By that point, the account already has cleaner search terms, a growing list of negatives, and at least some conversion history for Google to work with. Instead of trying to control every bid manually, buyers can let the algorithm focus on finding additional conversion opportunities while continuing to monitor traffic quality.
One Reddit contributor described Maximize Conversions as a middle ground between fully manual bidding and tCPA. Unlike tCPA, it doesn't force the system to stay within a specific acquisition cost, which gives Google more flexibility while it's still learning. At the same time, it's already optimizing around conversions rather than clicks, making it a more natural stepping stone toward automated bidding.
Many advertisers assume that a campaign that suddenly stops spending under tCPA needs to be rebuilt from scratch. Several experienced buyers disagreed with that approach. In their view, the first step is figuring out whether the campaign actually has enough data to support the target that's been set.
One BlackHatWorld member pointed out that tCPA often needs more conversion history than advertisers expect. If spend dries up shortly after enabling the strategy, his recommendation was to temporarily return to Manual CPC or Maximize Conversions, collect additional data, and revisit tCPA later with a more realistic target.
That advice appeared repeatedly in different forms across both Reddit and BHW. Rather than treating smart bidding as a switch that instantly improves performance, most experienced buyers seem to view it as something that gradually becomes more effective as the account accumulates more information.
The operational side of the process rarely gets discussed, but it has a direct impact on how quickly that information can be collected.
While building the foundation for tCPA or tROAS, affiliates are usually testing multiple variables at the same time: offers, landing pages, GEOs, keyword groups, creatives, and budgets. Every interruption slows down the learning process. An account suspension, delayed funding, spending restrictions, or moderation issues can easily delay the point where enough conversion data exists to make informed optimization decisions.
That's one reason agency accounts continue to be popular among media buyers running larger testing volumes. The goal isn't just to launch campaigns faster. It's to reach meaningful conversion volume before operational issues start getting in the way. For teams testing multiple offers at once, reducing that operational friction often means reaching the 20–30+ conversion range much sooner, which is exactly the point where strategies like tCPA and tROAS start becoming viable options rather than theoretical ones.
Most experienced buyers seem to follow the same principle: get tracking right, collect clean conversion data, understand which search terms actually drive results, and only then start tightening CPA or ROAS targets.
The faster that foundation is built, the faster Google's algorithms can start making useful optimization decisions.
That's why many experienced media buyers pay attention not only to campaign settings but also to account quality. Trusted accounts with established history tend to provide more stable delivery and cleaner learning conditions for Google's algorithms. When enough conversion data can be collected consistently, strategies like tCPA and tROAS usually perform far more predictably than they do on completely fresh accounts.
Some buyers still follow the traditional progression: Manual CPC or Max Clicks first, then Maximize Conversions, and only after that tCPA or tROAS. Others argue that this approach forces Google to learn from clicks instead of conversions and delays the point where the algorithm can start optimizing for actual business goals.
One Reddit user who claims to have managed more than £3 million in Google and Bing ad spend said he had tested both approaches across dozens of accounts. According to him, launching with Maximize Conversions and a tCPA target often produced better results than starting with click-based campaigns first.
The logic is easy to understand. If the end goal is conversions, why spend money training Google's systems on traffic that may never turn into leads in the first place?
At the same time, not everyone in those discussions had the same experience.
Several buyers reported that fresh accounts launched directly on tCPA either struggled to spend or pushed CPCs much higher than expected. In many of those cases, the issue wasn't necessarily the bidding strategy itself. The account simply didn't have enough information for Google to make confident decisions.
Most practitioners agree that tCPA and tROAS work best when Google already understands what a successful user looks like. The challenge is that a fresh account doesn't have that context yet. There are no historical conversions, no proven search terms, and no established patterns connecting keywords, ads, landing pages, and revenue. As a result, Google's algorithms often start from a position of uncertainty.
With tCPA, the system is trying to find users who are likely to convert around a specific acquisition cost. On an account with existing conversion history, Google can compare new users to people who converted in the past. On a brand-new account, it has far fewer signals to work with.
Things get more complicated once the conversation shifts to tROAS. Several PPC practitioners in Reddit discussions pointed out that Google can only optimize for return on ad spend if it understands which conversions actually generate revenue. That's relatively straightforward in ecommerce, where every purchase already has a value attached to it. In affiliate verticals, the picture is often much less clear.
One specialist described a setup that combines CRM data, Enhanced Conversions, and Offline Conversion Imports. The conversion is recorded in Google Ads first, but the real value gets assigned later, after the lead has been processed. That additional feedback helps Google distinguish between a lead the buyer simply considers promising and assigns a value of around $20, a user who deposits $100, and a customer who goes on to generate $500 in rebills within a week. Without those signals, all three conversions can look exactly the same inside the ad account.
That's why many buyers treat tROAS as something that comes after a campaign has already proven it can generate consistent conversions. Before Google can optimize for revenue, it first needs enough information to understand where that revenue is actually coming from.
This is also where account quality starts to matter.
Several discussions around smart bidding focus entirely on campaign settings, but campaign settings are only part of the equation. Before Google can optimize efficiently, it needs enough conversion volume to identify patterns. If an account is heavily restricted, spends inconsistently, or runs into operational issues before enough data is collected, the learning process can take much longer.
Agency accounts often have an advantage here. Higher spending limits, stronger account trust, and smoother moderation make it easier to reach the conversion thresholds that automated bidding strategies depend on.
When Does It Actually Make Sense to Switch to tCPA or tROAS?
One recommendation keeps showing up across PPC communities: don't rush into aggressive targets before the account has enough data behind it.The exact numbers vary depending on who you ask, but the same benchmark appears surprisingly often. Around 30 conversions in a 30-day period is commonly mentioned as a reasonable starting point for tCPA, while tROAS is usually associated with higher data requirements. Some practitioners in Reddit discussions specifically mentioned 30+ conversions for tCPA and 50+ for tROAS before expecting either strategy to perform consistently.
Interestingly, that isn't very far from Google's own recommendations. The platform itself suggests having at least 30 conversions before relying heavily on a target CPA strategy. Still, most experienced buyers don't treat these numbers as hard rules.
The more important question is whether the campaign is generating stable and predictable results.
If leads have been coming in at roughly $70–80 for several weeks, search terms are under control, and the landing page isn't changing every other day, Google has enough context to work with. In that situation, setting a target close to the actual CPA usually produces better results than trying to force a dramatic improvement overnight.
One recurring theme in community discussions is that campaigns often stop spending after the switch because the target itself is unrealistic. A campaign that has never generated leads below $80 isn't likely to start producing them at $30 simply because a buyer enters a lower number into the settings. From Google's perspective, there is no historical evidence that such a target is achievable, so the system starts participating in fewer auctions while searching for opportunities that fit the goal.
One Reddit user suggested approaching the transition more gradually. If the desired CPA is $30, he preferred starting somewhere around $40–45 and lowering the target over time as the campaign continued generating conversions. His argument was that giving the algorithm some room to operate usually works better than immediately forcing it into a narrow range.
Another interesting point raised in the discussions is that there isn't a single "correct" launch sequence. Some buyers still prefer the traditional route of Manual CPC, then Max Clicks, then Maximize Conversions before touching tCPA. Others skip parts of that process entirely.
One practitioner who commented on a Reddit thread said he often launches fresh Search campaigns with Maximize Conversions from day one, provided that conversion tracking has already been tested and verified. He also recommended relying on native Google Ads conversion tracking rather than imported GA4 events whenever possible, arguing that cleaner signals tend to produce better optimization results.
The same user mentioned that keyword control becomes especially important during the early stages. His preferred setup combines Exact Match and Phrase Match keywords while aggressively building negative keyword lists as search term data comes in. Broad Match, in his experience, gives Google too much freedom when the account has little or no conversion history, often leading to expensive experiments that the advertiser ends up paying for.
That opinion wasn't unique. Several buyers in the discussion described a similar workflow: start with tightly controlled traffic, review search terms daily, expand winning queries into dedicated keywords, and continuously eliminate irrelevant intent. By the time tCPA enters the picture, the account already has a much clearer understanding of what good traffic looks like.
Building the Foundation Before Smart Bidding
Once the first conversions start coming in, most of the routine work shifts away from bidding strategies and toward traffic quality.Several buyers in the discussions mentioned that they spend more time in the Search Terms report than anywhere else during the early stages of a campaign. The reasoning is simple: before Google can learn which users convert, the advertiser needs to make sure the campaign is attracting the right intent in the first place.
A search query that produces a conversion often gets promoted into its own Exact Match keyword. Queries that look promising but haven't generated enough data yet stay under observation. Anything clearly unrelated gets added to the negative keyword list. It's not particularly exciting work, but it's one of the fastest ways to improve signal quality before introducing stricter automation.
This is also where Maximize Conversions starts making sense for many advertisers. By that point, the account already has cleaner search terms, a growing list of negatives, and at least some conversion history for Google to work with. Instead of trying to control every bid manually, buyers can let the algorithm focus on finding additional conversion opportunities while continuing to monitor traffic quality.
One Reddit contributor described Maximize Conversions as a middle ground between fully manual bidding and tCPA. Unlike tCPA, it doesn't force the system to stay within a specific acquisition cost, which gives Google more flexibility while it's still learning. At the same time, it's already optimizing around conversions rather than clicks, making it a more natural stepping stone toward automated bidding.
Many advertisers assume that a campaign that suddenly stops spending under tCPA needs to be rebuilt from scratch. Several experienced buyers disagreed with that approach. In their view, the first step is figuring out whether the campaign actually has enough data to support the target that's been set.
One BlackHatWorld member pointed out that tCPA often needs more conversion history than advertisers expect. If spend dries up shortly after enabling the strategy, his recommendation was to temporarily return to Manual CPC or Maximize Conversions, collect additional data, and revisit tCPA later with a more realistic target.
That advice appeared repeatedly in different forms across both Reddit and BHW. Rather than treating smart bidding as a switch that instantly improves performance, most experienced buyers seem to view it as something that gradually becomes more effective as the account accumulates more information.
The operational side of the process rarely gets discussed, but it has a direct impact on how quickly that information can be collected.
While building the foundation for tCPA or tROAS, affiliates are usually testing multiple variables at the same time: offers, landing pages, GEOs, keyword groups, creatives, and budgets. Every interruption slows down the learning process. An account suspension, delayed funding, spending restrictions, or moderation issues can easily delay the point where enough conversion data exists to make informed optimization decisions.
That's one reason agency accounts continue to be popular among media buyers running larger testing volumes. The goal isn't just to launch campaigns faster. It's to reach meaningful conversion volume before operational issues start getting in the way. For teams testing multiple offers at once, reducing that operational friction often means reaching the 20–30+ conversion range much sooner, which is exactly the point where strategies like tCPA and tROAS start becoming viable options rather than theoretical ones.
Final Thoughts
Looking through the discussions on Reddit and BlackHatWorld, the biggest takeaway is that tCPA and tROAS don't fail because the strategies are flawed. More often, they fail because the account doesn't have enough reliable data yet.Most experienced buyers seem to follow the same principle: get tracking right, collect clean conversion data, understand which search terms actually drive results, and only then start tightening CPA or ROAS targets.
The faster that foundation is built, the faster Google's algorithms can start making useful optimization decisions.
That's why many experienced media buyers pay attention not only to campaign settings but also to account quality. Trusted accounts with established history tend to provide more stable delivery and cleaner learning conditions for Google's algorithms. When enough conversion data can be collected consistently, strategies like tCPA and tROAS usually perform far more predictably than they do on completely fresh accounts.
