Ad Slurry: The problem of low-quality inventory and how to deal with it

The problem

You’ve heard it many times… PPC practitioners face a growing problem with the lack of visibility and control into exactly where our spend is going.

There were at least three pieces of news sharing this theme in the last few weeks alone:

• Display targeting behaviour was widened (with ‘and’ becoming ‘or’ for the different targeting methods…)

• The Content Network targeting optout became misleadingly greyed out… (more on this below)

• The removal of ‘similar audiences’ was announced

But the theme itself is not news (I first wrote about it four years ago, and it wasn’t news then either…)

Before we look at the best response to it, let’s take a deeper look at the dynamics of what’s really happening here, and why.

Are we bothered?

First – as clear as the pattern is – is it really a problem?

We can certainly find PPCers who embrace the automation of bidding and targeting decisions…

But yes – it is a problem that we can’t sufficiently see or direct where our ad spend is going… and it’s not just that some PPC professionals tend to be control freaks.

(I can neither confirm nor deny that suggestion… )

Because the truth of ‘where our ad spend is going’ is also changing, and not for the better.

Why do they do it?

Some degree of low-quality ad inventory has always slipped (or been slipped…) into our campaigns – and this has increased over the years, for perfectly logical reasons.

As one PPC commentator asked rhetorically while railing against one of the many ‘unwanted expansions’ of Google Ads targeting this week, “is $65bn profit per quarter not enough for them?”

Totally understandable sentiments – but the answer to that question is no; it’s not enough…

Not because of greed – but because, as a public company, growth is essential for Google’s financial health.

It’s not enough for Alphabet to make billions; it needs to make more billions each year in order to justify (and increase) its share price.

On that basis, Google’s decision makers actually have a legal responsibility (fiduciary duty) to make every reasonable* effort to maintain and protect that growth (*NB one man’s reasonable is another man’s PMax…)

…and as it happens, Alphabet’s latest financial figures gave them no cause to celebrate.

(credit to Neil Andrew as the source of the comment quoted above. Well worth following for PPC updates and tips.)

And so, in this pursuit of growth, as Google rushes to accommodate ever more advertiser dollars, it needs in turn to increase the inventory it has to sell…

This is what lies behind the gradual expansion we’ve seen, from the core, high- value inventory that used to be sufficient for all parties concerned – into the lower-quality inventory we now struggle to identify and keep at bay.

How do they do it?

This ‘extra’ inventory has shown up in various ways,

From content network as an add on (which is almost always best to remove, but which as mentioned above, Google has now made appear mandatory with a new greyed out opt-out box to tick…)

…to ‘Targeting Expansion’ in display campaigns

…to the removal of a simple method for excluding mobile app placements (“” – a happy memory for those long in the game…)

…to the problem of clicks from countries that we aren’t targeting. It became much harder to identify these after the change to the way location data was presented in 2021.

With Performance Max campaigns, the option to target users ‘in or regularly in’ our selected locations was first added and then quickly hidden from the campaign setup process….

Speaking of which, the fact that this option changed from ‘in’ to ‘in or regularly in’ was yet another small opening of the door to that lower-quality inventory.

I could go on (and have done elsewhere, often enough…)

But it’s too easy to rant against these constant incursions. Far more productive to understand the bigger picture.

And on that – the best exposition I’ve seen is from Rand Fishkin, founder of Moz and Sparktoro. Watch his excellent explanation here of the depressing ‘Ad Slurry’ principle… How platforms monetise lower-quality inventory with increasing aggregation and obfuscation (not naming any names 🤔).

To summarise (in case, understandably, you can’t bring yourself to click away from this page…), while ad inventory varies in its value – both to a given advertiser, and in general terms – ad platforms have an incentive to monetise as much of their inventory as possible. Advertisers have a converse incentive to pay only for the inventory that they deem profitable, and not a step beyond it.

Those platforms then use various methods (like those listed above…) to ‘blend’ that lower quality inventory with the higher-quality – such that it’s either hard or impossible to see – or at any rate, hard to avoid.

We don’t need to be conspiracy theorists to note this dynamic. It’s an entirely understandable – possibly inevitable – result of the way our incentives as advertisers and those of our ad platforms diverge.

What to do about it?

1 Be aware of the issue – and how it manifests itself in different platforms and campaign types…

Some are far worse than others. Performance Max for example is notorious for its lack of detailed insights. In those rare moments when the curtain is pulled back – the view can be alarming.

Take this case, of a high-end travel advertiser, using Performance Max-for-Travel campaigns, with a hotel feed and the usual set of assets.

On checking the ‘Performance Max campaigns placement’ report (one that it’s well worth being aware of…)

This is the set of highest-impression placements it revealed.

No comment needed.

But this leads on to the next item, which is

2 Get to know the controls at our disposal – reduced as they may be – to combat low-quality impressions and clicks when you see (or suspect) them.

In the case above, account-level placement exclusions are the answer…

This method allows the blocking of whole categories of app.

The gaming categories would often be good candidates for that…

3 Keep an eye on your metrics – and a keen eye on what they really mean.

For example, low CPCs aren’t necessarily a good thing

Even Conversion Rate can be misleading

and there are subtle ways in which ‘aggregated data’ can dilute the true picture.

Search partners traffic is a good example of this problem (but takes a bit of explaining…. Read tip #8 from my free guide, 10 advanced paid search tips)


4 Find and turn to a trusted partner to do this kind of advanced monitoring for you.

A one-off audit (the kind of thing I offer in one-to one sessions) may be what you need…

or if you’re looking to sharpen up your own PPC eyes, consider something like my paid course and membership Google Ads Level Up.

Share this post

Lastest Posts

Google Ads Updates - Q1 2024 Roundup

No slowdown in the pace of change with Google Ads updates in Q1 2024… Here’s…

Aggregation vs Articulation: Competing Virtues of PPC Management

Three ‘A’s Two concepts have been central to the recent evolution of Google Ads: aggregation…