We expected some of our Search campaigns to be short of conversion volume. We did not expect all of them.
Twenty-four Search campaigns across thirteen lead generation accounts. Not one produced 30 conversions a month. Twenty-three produced fewer than ten. Twenty produced fewer than five.
Every dollar of Search spend in the export, $109,631 of it, was running in campaigns that could not give Google’s bidding systems enough data to work with.
The campaigns were not badly built. Several are well organised, tightly themed, sensibly negatived. That is the uncomfortable part. Careful campaign building is exactly how you end up here.
This is the third piece from a campaign-level export across our lead generation accounts. The first found that one account’s broken conversion action was distorting our PMax versus Search comparison by 5x. The second found that seven of ten Target CPA campaigns were below the volume that strategy needs. This one is the underlying cause of the second.
The dataset
We pulled a campaign-level report from our manager account on 24 July 2026, covering 1 June 2025 onwards, which is roughly 13.8 months. That returned 36 campaigns across 13 US lead generation accounts. This article is about the 24 Search campaigns among them, which carried $109,631 of spend between them.
The businesses are plumbing, fencing, iron doors, windows and doors, home remodeling, tree removal, counselling, psychiatry, payroll services and dental.
Conversions per month means total conversions divided by 13.8. Campaigns launched partway through the window are understated by that, and we flag it where it changes the argument. Our manager account reporting shows more spend than these campaigns account for, so this is a subset rather than the full book. No client is named.
The distribution
| Threshold | Search campaigns below it | Share |
|---|---|---|
| 50 conversions/month | 24 of 24 | 100% |
| 30 conversions/month | 24 of 24 | 100% |
| 10 conversions/month | 23 of 24 | 96% |
| 5 conversions/month | 20 of 24 | 83% |
Google’s Target CPA documentation recommends at least 30 conversions in 30 days. Target ROAS is generally held to need 50 or more. Maximize Conversions is more forgiving but still performs better with volume.
Against the most lenient of those, every Search campaign we run fell short.
What fragmentation looks like inside one account
The plumbing account is the clearest example because it is the largest, and because it was built the way most people are taught to build.
Twelve campaigns. Eight of them Search, one per service line.
| Campaign | Bid strategy | Spend | Conversions | Per month |
|---|---|---|---|---|
| Brand | Manual CPC | $10 | 0 | 0.0 |
| Leak detection & repair | Maximize Clicks | $1,883 | 10 | 0.7 |
| Piping & repiping | Maximize Clicks | $6,514 | 26 | 1.9 |
| Water filtration | Maximize Clicks | $2,588 | 28 | 2.0 |
| Water heater install & repair | Target CPA | $13,817 | 44 | 3.2 |
| Drain cleaning & repair | Maximize Conversions | $4,934 | 47 | 3.4 |
| Search near me | Maximize Conversions | $7,327 | 120 | 8.7 |
| Plumbing services | Maximize Conversions | $7,150 | 126 | 9.1 |
Eight campaigns. Four different bid strategies. Not one campaign above ten conversions a month.
Add them together and the Search side of this account produces roughly 29 conversions a month. One campaign, at that volume, would sit right at the threshold where Target CPA becomes defensible.
Split eight ways, none of them can learn anything.
This is the entire argument for consolidation in one table. The account has the demand. The account has the budget. What it does not have is enough data in any single container for the bidding system to find a pattern.
Worse, the water heater campaign has Target CPA running on 3.2 conversions a month, which is the exact scenario the previous article covered. Fragmentation created the low volume, and the low volume made the bid strategy unworkable.
Why careful people build accounts this way
Nobody fragments an account on purpose. It happens for reasons that all sound correct in the moment.
It mirrors how the client talks. The client sells drain cleaning, water heaters, repiping and filtration. Building one campaign per service feels like organising the account around the business.
It looks better in reports. A client wants to know how water heaters are performing. A campaign named “Water Heater Install & Repair” answers that question in one glance.
It came from the old playbook. Under manual bidding, tight segmentation genuinely helped, because you were setting bids yourself and needed granular control to do it. That advice is fifteen years old and it has outlived the conditions that made it true.
Budget control. Separate campaigns let you cap spend per service. This is the strongest of the four reasons and the hardest to give up.
It accumulates. Nobody built eight campaigns on day one. Somebody added a campaign for a new service, then another for a seasonal push, then a test that was never turned off. Thirteen months later there are eight.
Every one of those decisions is defensible. The combined result is an account where automated bidding cannot function.
What fragmentation actually costs you
Learning never completes. Smart Bidding needs conversion volume per campaign, not per account. Below threshold, campaigns sit in a permanent state of recalibration, and every edit restarts the clock.
Budgets get stranded. Eight campaigns with eight daily budgets means money sitting idle in the campaigns with no demand today while the campaign with demand hits its cap. A consolidated campaign moves budget to wherever the demand is, automatically, every day.
Your reporting is noise. At three conversions a month, one extra lead is a 33% swing. You cannot tell a real trend from a coincidence, which means every optimisation decision is being made on a sample too small to support it.
You optimise against randomness. The natural response to a bad month in a small campaign is to change something. At that volume, most of what you are reacting to is variance, and each reaction resets learning. This is how accounts get busy without getting better.
Bid strategies drift apart. In the account above, four bid strategies are running across eight campaigns serving one business. Nobody chose that. It accumulated, and now the account behaves inconsistently for reasons that are difficult to explain to anyone.
What happens when people actually consolidate
We are publishing our own problem here, not our own solution, so it is worth looking at what practitioners who have already done this report.
Boris Beceric, a Google Ads consultant and coach, told Search Engine Land that most advertisers try Smart Bidding too early without enough conversion volume, and that consolidating campaigns so more data flows through a single campaign is what usually helps. That is the diagnosis. The numbers below are what happens when you act on it.
The team at PPC Mastery have published two consolidation results worth knowing about. In an ecommerce account, merging 44 non-branded Search campaigns into 7 produced a 345% increase in conversion value and a 22% ROAS improvement. More relevant to us, a lead generation account consolidated from 19 campaigns down to 4 saw conversions rise 253% and cost per acquisition fall 60%. Notably, the person who ran the ecommerce consolidation says that doing it again today, they would use even fewer campaigns.
The most extreme published example is the Hagakure approach documented on Search Engine Land, where a retailer’s fashion brand campaigns went from 450 down to three. The top eight brands, accounting for around 75% of search sales, were grouped into a single campaign with a dedicated ad group per main keyword. That is a scale most accounts will never need, but the direction is the same one.
The honest counterweight comes from Adchieve, who ran Hagakure restructures across a number of accounts and reported strong improvements in most but not all. The accounts where results came out slightly worse were the small ones with few clicks and conversions, where the algorithms had less to work with and chance played a bigger role. That is a caution worth taking seriously, because it means consolidation is not a guaranteed win on a genuinely tiny account. It is a way of giving automation enough data to function, and if the total demand is not there, consolidation cannot manufacture it.
Our plumbing account has the demand. Twenty-nine conversions a month across eight containers is a structure problem, not a volume problem.
What consolidation actually means
Consolidation does not mean one campaign for everything. It means the smallest number of campaigns that still respects genuine differences.
Three things justify a separate campaign. Everything else is an ad group.
Different budget requirements that genuinely must be enforced. If the client has committed a fixed monthly spend to one service line and it cannot be exceeded, that is a real constraint.
Different geography with different economics. Serving a metro area and a rural area with different costs per lead and different competition is a real difference. Serving two adjacent suburbs is not.
Different conversion goals or bid strategies. Emergency service optimising for calls behaves differently from planned service optimising for booked consultations. If the goal genuinely differs, separate.
Everything else, including different services, different keyword themes, and “the client wants to see it separately,” belongs in ad groups inside one campaign. You keep the reporting granularity, because ad group reporting exists, and you give the bidding system a single pool of data.
For the plumbing account, the honest structure is roughly: one non-brand Search campaign with ad groups per service, one brand campaign kept separate because brand economics are completely different, and the local PMax campaigns as they are. That is three or four campaigns instead of twelve.
How to run this check on your own account
Step 1. Campaigns view, last 30 days, sorted by Conversions ascending.
Step 2. Count how many campaigns produced fewer than 30 conversions. Then count how many produced fewer than ten. The second number is the one that will tell you something.
Step 3. For each cluster of low-volume campaigns, ask the three questions above. Different enforced budget? Different geography with different economics? Different goal or bid strategy? If all three answers are no, they are ad groups.
Step 4. Add up the conversions across the campaigns you would merge. If the combined figure clears 30 a month, consolidation is not a preference, it is the difference between automated bidding working and not working.
Step 5. Check the conversion column is honest before you trust any of this. Segment by Conversions > Conversion action. A campaign that appears to clear the threshold on phone taps or page views has not cleared anything.
Step 6. Before merging, export your search terms, negatives and ad copy per campaign. Consolidation is reversible in principle and painful in practice if you did not keep records.
How we would sequence the migration
Consolidation done carelessly costs you performance for a month and gets blamed on the idea rather than the execution.
Build the new campaign alongside the old ones. Do not edit eight campaigns into one. Create the consolidated campaign, ad groups mapped to the old campaigns, keywords carried across with their match types intact.
Carry the negatives. This is the step people skip and regret. Every negative in every old campaign exists because somebody found wasted spend. Consolidate the negative lists before you launch, or you will re-buy every bad search term the account already learned to avoid.
Start on Maximize Conversions. Not on a target. The new campaign has no history of its own and needs to establish a real cost per lead before anyone constrains it.
Set the budget to the combined total of the campaigns it replaces. Not lower. Consolidation is not a cost-cutting exercise and starving the new campaign will make it look like consolidation failed.
Run both for a short overlap, then pause the old ones. A few days is enough. Leaving them running longer means competing with yourself in the auction.
Do not touch it for 30 days. The learning period is real, the first two weeks will look worse than the old setup, and the entire benefit arrives after the system has enough pooled data to work with. Every account we have consolidated has had a nervous fortnight followed by a better quarter.
Then, and only then, consider a target. Once the consolidated campaign is consistently above 30 conversions a month, the sequence from the Target CPA article applies.
The alternative if you genuinely cannot consolidate
Sometimes the client constraint is real and the campaigns have to stay separate.
In that case, use a portfolio bid strategy. It pools conversion data across the campaigns you include and bids against the combined pool, which gets you most of the learning benefit while keeping the campaigns and their separate budgets.
It is a genuine solution and not a hack. But it is second best. Portfolio strategies still leave you with fragmented budgets that cannot flow to demand, and reporting that is split across containers with too little data each.
Use it when structure cannot change. Do not use it to avoid a conversation about structure that should happen.
What we are doing about our own accounts
We are consolidating, starting with the plumbing account, and we will publish what happens.
The honest position is that we found this in our own book, not in somebody else’s. These accounts were built by us, over time, for reasons that made sense at each step. The pattern was invisible until we exported everything into one table and looked at conversions per month per campaign, which is not a view the Google Ads interface encourages you to take.
If you have never run that specific check on your own accounts, we would guess your numbers look more like ours than you expect.
Methodology and limitations
This report covers 36 campaigns across 13 US lead generation accounts running between 1 June 2025 and 24 July 2026, with $163,732.04 in total spend. The 24 Search campaigns analysed here represent $109,631 of that.
Conversions per month means total conversions divided by 13.8 months, so campaigns that launched or paused partway through are understated. We do not have reliable launch dates for every campaign and preferred a stated approximation to adjusting some and not others.
The thresholds referenced come from Google’s Target CPA documentation, which recommends at least 30 conversions in the last 30 days for evaluation.
The sample is small: thirteen accounts, one carrying over half the spend, all US, mostly home services in the southeast. This is a description of one agency’s accounts rather than an industry benchmark. We publish it because 100% was not a number we expected to find, and because the practitioners quoted above suggest we are not unusual.
One further caveat matters here. As documented in the first article in this series, one account in this set had a phone number click set as its primary conversion. If anything that means the real conversion volumes in that account are lower than shown, which makes the fragmentation problem worse than these figures suggest rather than better.
FAQ
How many conversions does a Search campaign need per month? At least 30 if you want to use a target-based bid strategy, per Google’s own documentation. Maximize Conversions works below that but performs better with more. Below roughly ten a month, your reporting is too noisy to optimise against regardless of bid strategy.
Does consolidating campaigns mean losing reporting detail? No. Move the campaigns to ad groups and you keep the segmentation in reporting. What you lose is the ability to set a separate daily budget per service, which is usually the actual reason people resist.
Will consolidation hurt performance at first? Expect one to two weeks of a learning period, and do not judge anything inside 30 days. In our experience the first fortnight looks worse and the following quarter looks better.
What if the client insists on separate budgets per service? Use a portfolio bid strategy across those campaigns. It pools the conversion data while keeping the campaigns separate. It is second best, but it is a real solution.
Should brand and non-brand stay separate? Yes, always. Brand traffic converts at a completely different rate and cost, and blending it into a non-brand campaign corrupts both your bidding and your reporting.
Does this apply to Performance Max? The same logic applies, but PMax aggregates across placements so it tends to accumulate volume faster than a narrow Search campaign. Segmenting PMax excessively, particularly for lead gen, creates the same starvation problem.
How do I know if my campaigns should be merged? Different enforced budget, different geography with different economics, or different conversion goal. If none of those apply, they should be ad groups.
About The Author
