TLDR;
- Most advertisers assume that more impressions at a lower cost equals better performance, but this is a dangerous oversimplification.
- Analysis of a real Google Ads account shows that channels with massive impression volume (like Discovery) can be severely inefficient at generating qualified users compared to intent-driven channels like Search.
- The primary objective is to distinguish between low-quality shallow conversions (e.g., cheap app installs) and high-quality, considered actions (e.g., web registrations) to calculate a quality-weighted acquisition cost, which tells a completely different story.
- This article includes interactive calculators and benchmark visualisations to help you diagnose the quality of your own traffic mix and spot impression wastage in your campaigns.
We are all guilty of it. We open up an ad account, look at the dashboard, and our eyes are immediately drawn to the big numbers. Millions of impressions. A click-through rate that is respectable. A nice, low, blended cost per conversion that we can pop into a slide deck. But what if those headline figures are not just misleading, but actively hiding a serious problem? What if the vast majority of those impressions we are paying for are effectively worthless, and the conversions they generate are just noise?
This is not a theoretical question. It is a problem I see time and again when auditing accounts. There is often a massive, and costly, discrepancy between the sheer volume of ads being shown and the actual number of genuinely engaged, qualified users that come out the other end. So, I decided to properly investigate this question: What discrepancies actually exist between raw impression volume and qualified user engagement in multi-platform ad campaigns?
To get to the bottom of this, I am not going to use industry averages or vague benchmarks. I am going to pull apart a single, complete Google Ads account dataset for a B2B SaaS client we have worked with, ScoutIQ. They have an app that helps people scan and value books for selling online. The dataset covers a long period, from mid-2020 to early-2024, with a total ad spend of £3,387.34. Because it is a single account running multiple, distinct campaign types (Search, Performance Max, App Installs, and Demand Gen/Discovery), it is the perfect clean-room environment to see these different dynamics play out without the mess of cross-client variables. This is not a universal benchmark, but a detailed look at how different ad formats on the same platform behave for the same product.
So, how do you even measure qualified engagement?
Before we can analyse anything, we need to agree on what a qualified user actually is. This is where most reporting falls flat. A conversion is not a conversion. The data for this account gives us two primary conversion goals:
- Mobile App Installs: A user sees an ad, likely in the Google Play store, and clicks to install the app. The cost for this was incredibly low, averaging just £0.30 per install.
- Web Account Registrations: A user clicks an ad, is taken to the ScoutIQ website, and completes a sign-up form. This was a more expensive action, averaging £2.25 per registration.
Straight away, common sense and experience tell you these are not of equal value. The app install is a low-friction, almost impulse action. The web registration requires more effort and intent – the user has to visit a site, read something, and consciously decide to hand over their details. The 7.5x cost difference (£2.25 / £0.30) is not arbitrary; it reflects what the market and Google's algorithm know about the relative value and difficulty of acquiring these two types of users.
For this analysis, I am going to treat the Web Registration as our benchmark for a qualified user. It represents a much stronger signal of intent. This doesn't mean app installs are worthless, but it does mean we cannot treat them as equal. To quantify this, I am going to create a simple Quality-Weighted Conversion metric. If a low-friction App Install is worth 1 Quality Point, then a high-intent Web Registration is worth 7.5 Quality Points, based on that cost ratio. This lets us move beyond a simple headcount of conversions and start to measure the actual value being generated by our ad spend. With this, we can now properly dissect the performance of each campaign type.
Campaign Performance & Quality Matrix
This table breaks down the core performance and calculated quality metrics across all major campaign types run in the account. Pay close attention to the Impressions per Conversion and Cost per Quality Point columns to see the true efficiency of each channel.
| Campaign Type | Cost | Impressions | Conversions | Impressions / Conv. | Cost / Quality Point |
|---|---|---|---|---|---|
| Search | £1,538.32 | 27,853 | 592 | 47 | £0.35 |
| Performance Max | £665.89 | 25,562 | 554 | 46 | £0.16 |
| App Installs | £754.90 | 101,600 | 2,490 | 41 | £0.30 |
| Demand Gen (TOFU) | £287.54 | 21,434 | 16 | 1,340 | £2.40 |
| Demand Gen (MOFU) | £140.69 | 10,159 | 6 | 1,693 | £3.13 |
The table above lays the data bare, and the story it tells is stark. On the surface, you see five campaign types, all spending money, all getting impressions. But when you start looking at the efficiency metrics, a massive divide appears.
The Search, Performance Max, and App Install campaigns are all remarkably efficient at turning an impression into some kind of action. They each require only 40-50 impressions on average to generate a conversion. Now, look at the Demand Gen campaigns (formerly Discovery). The top-of-funnel campaign needed a staggering 1,340 impressions for every conversion. The middle-of-funnel retargeting campaign was even worse at nearly 1,700 impressions. This is the discrepancy in its purest form. You are paying to show your ad to over a thousand people just to get one person to take action.
But the story gets more interesting. While the Impression to Conversion ratio for Search and App Installs looks similar, we know the quality of those conversions is wildly different. The Search campaign primarily drove high-value web registrations (£2.25 CPA), while the App campaign drove low-cost installs (£0.30 CPI). This is where our Cost per Quality Point metric comes in. Assuming Web Registrations from Search and PMax are worth 7.5 points and App Installs are worth 1 point, we can calculate a truer cost of value. The App Install campaign's cost per point is simply its CPI: £0.30. The Search campaign's CPA of £2.60 divided by 7.5 quality points gives us a Cost per Quality Point of approximately £0.35. Suddenly, they look very comparable from a value perspective. And Performance Max, with its blended approach, looks even better at just £0.16 per quality point. The Demand Gen campaigns, however, are an anomaly on this metric, costing £2.40 to £3.13 to acquire a single point of quality - nearly 10 times less efficient than the other channels.
Are you paying for users or just filling a leaky bucket?
This brings us to a vital point about blended metrics. The headline figures for this account were 3,543 users at £0.96 each. It sounds brilliant. Sub-£1 user acquisition is fantastic for almost any SaaS business. But the detail reveals a different picture. Of those 3,543 users, a massive 2,482 around 70% came from the low-cost, low-intent App Install campaign. Only 30% were from the higher-intent, more costly web registration path.
So while the blended average looks good, the user base is heavily skewed towards the lowest quality cohort. For a SaaS business focused on activation, retention, and lifetime value, this is a dangerous situation. You might be hitting your top-line user acquisition targets, but you are filling your funnel with users who are far less likely to stick around or ever pay you. This is the illusion of low-cost volume. It feels like you are winning because the numbers are big and the costs are small, but you are just pouring water into a leaky bucket.
To really get a feel for this, I have built a small calculator. You can plug in your own numbers for two different conversion types—a low quality one like an app install and a high quality one like a web signup. It will show you your blended CPA, but more importantly, it will calculate a Quality-Weighted CPA based on the cost difference. It helps to reveal the true cost of acquiring a valuable user, not just any user.
User Quality Mix & True CPA Diagnostic
Enter the volume and cost for your low-intent (e.g. app installs) and high-intent (e.g. web signups) conversions to see how your user mix impacts your true acquisition cost.
Where does real engagement come from? The power of intent
If cheap impressions from passive channels like Discovery are so inefficient, where does the best engagement come from? The data points to one clear answer: user intent. The Search campaign is the perfect example. Here, we are not interrupting someone's browsing; we are answering a direct question. An impression is only served when someone actively types in a query like book scanner app or how to sell books online.
These impressions are, by their very nature, pre-qualified. The user has a problem and is actively looking for a solution. It is no surprise, then, that this channel is so much more efficient. While it required 47 impressions per conversion, those conversions were almost all high-quality Web Registrations. To dig deeper, we can look at the ad groups within the Search campaign, which were structured by user intent:
- TOFU (Top of Funnel): Broad terms like selling online.
- MOFU (Middle of Funnel): More specific, high-intent terms like book scanner app.
- BOFU (Bottom of Funnel): Brand keywords and competitor names.
You would expect a neat progression, with costs getting lower as intent gets higher. The data mostly bears this out, but with a twist. The brand keyword ad group had the lowest CPA at £1.75, which makes perfect sense. But surprisingly, the broad TOFU ad group, with a £2.55 CPA, actually outperformed the more specific MOFU (£2.77) and competitor (£3.55) ad groups. This is a great reminder that you cannot make assumptions. In this case, it is likely that the sheer volume of searches for broader selling online terms, combined with Google's improving machine learning, allowed it to find converting users efficiently at scale, even with lower average intent. Nonetheless, all these Search CPAs are clustered in a tight, efficient range, a world away from the £17+ CPAs of the Discovery campaigns.
Search Campaign: Cost Per Conversion by User Intent
This chart breaks down the cost-effectiveness of different ad groups within the main Search campaign, segmented by the user's position in the buying funnel.
Visualising the true cost of an impression
Ultimately, this whole investigation comes down to one thing: not all impressions are created equal. An impression on a high-intent search keyword is fundamentally different from an impression on a passively-viewed display or video ad. The former is a potential conversation; the latter is often just digital noise. To make this crystal clear, let's look at Impression Wastage – the number of impressions you have to buy to get a single conversion.
The differences are astronomical and put the whole discrepancy into sharp focus. The following chart visualizes the Impressions per Conversion for each campaign type. It is a measure of raw efficiency. How many times did we have to put our ad in front of someone's face before one of them finally did something?
Impression Efficiency Benchmark by Campaign Type
This chart shows the average number of impressions required to generate one conversion for each campaign type. Lower is better.
Data source: Analysis of ScoutIQ Google Ads account (2020-2024).
The chart says it all. The Demand Gen campaigns are a black hole for impressions. This does not mean they have no value – they can play a role in brand awareness and keeping your name top-of-mind. But if your goal is efficient user acquisition, as it is for most SaaS businesses, relying on these channels is like trying to fill a swimming pool with a teaspoon. You are spending the vast majority of your impression budget on people who will simply never convert. The intent-driven and specialized campaigns, meanwhile, are surgical. They find the right people with minimal waste.
What does this all mean for your ad strategy?
The findings from this single account analysis are clear and have major implications for anyone running multi-channel ad campaigns, especially in the B2B SaaS world.
First, stop obsessing over raw impression volume and blended CPAs. They are surface metrics that hide the truth. You must segment your reporting by campaign type and, more importantly, by the quality of the conversion action. If you have multiple conversion goals, you need a way to weight them to understand the true value you are acquiring.
Second, understand the job of each channel. Do not expect a broad awareness campaign like Demand Gen to be an efficient driver of high-intent signups. Its job is reach. Your Search campaigns' job is to capture existing intent. Your App Install campaign's job is to drive low-cost volume to the top of the funnel. Judge them on the right metrics. The problem arises when you blend all the results and think the average represents reality.
Third, always prioritize budget towards intent where possible. Capturing someone actively looking for your solution is always going to be more efficient than trying to convince a cold prospect they have a problem they did not know about. As the data shows, the largest portion of spend in this account (£1,538, or 45% of the total) rightly went to the Search campaign, which drove the highest quality users. In contrast, the App Install campaign drove 70% of the users for only 22% of the spend (£755). This is a classic volume versus value trade-off. This particular business made a strategic decision to acquire a large volume of low-cost users via the app, likely to build an initial audience and gather data, while simultaneously using Search to capture high-value users. It is a valid strategy, as long as it is a conscious one and you are not fooling yourself about the quality of that 70% cohort.
Budget vs. User Acquisition Breakdown
This visualization contrasts the percentage of total budget spent against the percentage of total users acquired, highlighting the strategic trade-off between acquiring high-quality users and high-volume users.
The discrepancy between serving an ad and getting a real user is not a small rounding error; it is a fundamental flaw in how many businesses measure success. By digging into the data, moving beyond surface metrics, and applying a quality-weighted lens to your conversions, you can start to see the true picture of your advertising performance. You can allocate budget more effectively, optimize for value instead of volume, and build a system that does not just generate clicks, but engineers real growth.
If you are looking at your own dashboards and suspect that your big impression numbers might be hiding a similar story of inefficiency, it is probably time for a deeper look. Getting this right is not about tweaking a few bids; it is about fundamentally rethinking how you measure success. We offer a free initial consultation where we can apply this same level of detailed analysis to your accounts and help you figure out where the real value—and the waste—is hiding.
Cite this Research & Data
If referencing this research, data, or strategy in your publication, please use the attribution below.
Lukas Holschuh
Founder, Growth & Advertising Consultant
Great campaigns fail without expertise. Lukas and his team provide the missing strategy, optimizing your entire advertising funnel—from ad creatives and copy to landing page design.
Backed by a proven track record across SaaS, eLearning, and eCommerce, they don't just run ads; they engineer systems that convert. A data-driven partnership focused on tangible revenue growth.