Google Ads Audit: Where Your Budget Is Leaking and How to Find It

Google Ads rarely loses your budget all at once. More often the leak is quiet: a few euros a day on irrelevant searches, impressions inside mobile apps, duplicated brand clicks, and campaigns optimizing toward conversions that are not actually being measured correctly. Added up over a month, these small leaks are often the difference between an account that turns a profit and one that barely breaks even.

This guide shows where budget most commonly leaks in Google Ads and how to find it yourself — step by step, with the exact places to check. At the end you will find a short audit checklist and answers to the most common questions, so you can work through your account in under an hour.

David Ogilvy said that half of every advertising budget is wasted — the problem is knowing which half. In Google Ads, that half is measurable: it shows up in the search terms report, in the device breakdown, and in the settings nobody has opened since launch. Alex Hormozi builds on the same idea — the first profit comes not from more budget, but from stopping the spend that produces no result.

What “wasted spend” means in Google Ads

Wasted spend is every unit of budget that leaves the account without contributing to a profitable sale or inquiry. It is not just obviously wrong clicks — a large share of the leaks comes from default settings that Google suggests because they widen reach, and wider reach means higher spend.

The distinction matters: high spend is not a problem in itself if it produces profit. The problem is spend without contribution. That is why the audit does not ask “where are we spending a lot” but “where are we spending with nothing to show for it” — and those are two different questions.

Leaks usually cluster in a few areas: irrelevant search terms, broad match running without control, weak account structure, missing negative keywords, spend inside the Performance Max “black box”, wrong location and device settings, and — the most expensive of all — advertising optimized against broken measurement. We will go through them one by one.

1. The search terms report: the first place to check

The fastest win in almost every account hides in the search terms report — the real queries your ads were shown for. The keyword you set up and the phrase you actually paid for are often two different things.

This is where the raw truth shows: queries that have nothing to do with your product, informational searches with no purchase intent (“how to make”, “free”, “review”, “second hand”), and queries for competitor brands or incompatible products. Every one of those clicks is paid for.

In practice: sort the search terms by cost and work from the top down. For every phrase with meaningful spend and zero conversions, ask yourself one question — “if this person were standing in front of me, would I want to pay for them?”. If the answer is no, the phrase is a candidate for a negative keyword.

2. Broad match without control

Broad match gives Google the most freedom to decide which queries to show you for. With strong measurement and smart bidding it can work — but launched without control and without negative keywords, it is the most common source of wasted spend.

For most accounts, the sensible approach is to start with a more controlled match type (phrase), watch which real queries come in through the search terms report, and expand toward broader matching only where the data justifies it. Broad match is not off limits — it simply requires a buffer of negative keywords and working measurement before it gets budget.

The telltale sign of a leak: a large share of spend comes through broad match, and the search terms report is full of irrelevant queries. That is a signal the match type is outrunning your control.

3. The negative keywords that are missing

Negative keywords are the account’s brakes. Without them, your ads show for queries you would never knowingly pay for. Most leaking accounts have no maintained negative keyword list — or have an old one that has not been touched in months.

A few types of negatives that are almost always worth adding:

• Searches without purchase intent — “free”, “giveaway”, “second hand”, “DIY”, “jobs”, “review”, whenever that is not exactly what you sell.
• Incompatible products and categories — queries for things that look related but that you do not offer.
• Irrelevant brands — brands you do not carry but that your ads show for anyway.
• An account-level list — one shared list of obviously irrelevant terms applied across the whole account, so you do not repeat it for every campaign.

Negative keywords are not a one-time task. The search terms report produces new candidates every week — maintaining the list is what keeps the spend clean over time.

4. Account structure: when everything sits in one campaign

A weak structure does not spend money directly, but it makes every other leak harder to find and more expensive to fix. When brand and non-brand searches share one campaign, when products with different margins share one budget, or when a single campaign mixes objectives, the budget is allocated by the algorithm, not by you.

A few principles that put the account in order:

• A separate campaign for brand searches. People searching for your name already know you — mixed in with cold traffic, they distort the averages and hide the true cost of acquiring a new customer.
• Split by margin and by objective. High-margin and low-margin products have different break-even points; when they share a budget, the winners quietly subsidize the losers.
• One clear priority per campaign. A campaign chasing sales, traffic, and awareness at the same time usually achieves none of the three well.

A clean structure is what makes the audit possible: when every campaign has one job, you can see immediately which one is working and which one is leaking.

5. Performance Max: spend inside the “black box”

Performance Max allocates budget automatically across Search, Display, YouTube, and other channels, with minimal visibility into where exactly the money goes. It can work very well — but precisely because of that limited transparency, it is a place where spend leaks unnoticed.

A few things that put control back in your hands:

• Negative keywords and brand exclusions. Without them, Performance Max often spends part of the budget on brand searches you would have won anyway, and then takes credit for those sales.
• Checking the channel distribution. A large share of spend can flow to Display and cheaper, low-contribution impressions, while high-intent search gets little.
• The quality of the data you feed it. Performance Max learns from your conversion signals — with weak or broken measurement, it optimizes toward the wrong target, and that is expensive.

In practice: do not take Performance Max’s reported results at face value. Compare them against the real growth in total profit, and check whether it is claiming credit for sales that actually come from brand searches.

6. Location, devices, and partners: the default settings

Some of the quietest leaks sit in settings nobody opens after launch. The defaults are built for broad reach, not for your unit economics.

• Location setting. By default, Google often shows your ads to people outside your target area who merely “show interest” in it. If you only sell in a specific region, it is reasonable to restrict targeting to actual presence there, not interest.
• Device breakdown. If mobile traffic brings many clicks and few conversions, it is quietly eating your budget. Do not adjust blindly — check the data first, then reallocate the budget.
• Impressions in mobile apps. Clicks on ads inside mobile apps are frequently accidental and almost never produce a real sale for an online store — yet they cost the same as every other click.
• Search Partners and default expansions. The partner network sometimes delivers cheaper but also weaker traffic; it is worth checking separately to see whether it contributes or merely dilutes the averages.

These settings get checked once and then periodically. Half an hour of review here often saves more than a week of optimization elsewhere.

7. Measurement: the most expensive leak of all

Everything so far assumes your data is correct. If measurement is leaking, the entire account optimizes toward the wrong target — and that is the most expensive possible problem, because you multiply it with every unit of budget you invest.

Signs that measurement is off:

• Duplicate conversion counting. The same event counted more than once — a purchase and an “add to cart” recorded as two conversions inflate the results.
• Missing conversion value. If you report a “purchase” without the actual amount, smart bidding cannot tell a 20-euro order from a 200-euro order.
• Counting unwanted actions as conversions. Phone clicks, page visits, or form submissions with no real business outcome, recorded as conversions, steer the algorithm in the wrong direction.
• Missing server-side measurement. A share of conversions is lost when scripts are blocked; without a server-side setup, the data is incomplete in both directions.

The sensible order is to verify measurement before optimizing campaigns. There is no point in trimming negatives and bidding precisely on data that cannot be trusted.

An illustration: how small leaks add up

The figures below are illustrative, meant to show the logic of accumulation — they are not metrics from a specific account and not a promise of results:

Each number looks small on its own. Combined, they are often the difference between a profitable account and a losing one — and almost all of them are found through a review, not through more budget.

Step by step: a quick audit in under an hour

1. Open the search terms report for the last 30–90 days and sort it by cost. Flag the phrases with spend and no conversions.
2. Add negative keywords for the irrelevant queries — and consider a shared account-level list.
3. Check your keyword match types. See what share of spend flows through broad match and whether it is backed by negatives.
4. Review the structure. Are brand searches separated? Are products with different margins or different objectives mixed into one campaign?
5. Open the location and device settings. Is location restricted to actual presence? Does mobile traffic bring conversions, or only clicks?
6. Exclude impressions in mobile apps and check Search Partners separately.
7. Check Performance Max for negatives / brand exclusions and for its channel distribution.
8. Check measurement last — and most important. No duplicate conversions, conversion value is being passed, only real business actions are counted, and server-side measurement is in place.

Go through these eight points once, then revisit the search terms and negative keywords periodically — they are the part that starts leaking again if left unmaintained.

A short audit checklist

• I have reviewed the search terms report and removed the irrelevant queries.
• I have a maintained negative keyword list (plus a shared account-level one).
• I know what share of spend flows through broad match and whether it is under control.
• Brand searches sit in a separate campaign from cold traffic.
• Products with different margins and different objectives do not share one budget.
• Location is restricted to actual presence, not interest.
• I have checked the device breakdown and excluded impressions in mobile apps.
• Performance Max has negative keywords, and I know how its spend is distributed across channels.
• Measurement is correct: no duplicates, with conversion value, only real actions, with a server-side setup.

Frequently asked questions

How often should I audit my Google Ads account?

A full review of the structure and settings — once a quarter is usually enough. The search terms report and the negative keywords, however, need more frequent attention, because new queries come in every week. Part of the account is audited periodically, the other part continuously.

Does broad match always waste budget?

No. Broad match wastes budget when it runs without negative keywords and without working measurement. With a buffer of negatives and reliable conversion signals, it can deliver results. The problem is the control, not the match type itself.

Is Performance Max worth it, or is it just a “black box”?

It can work very well, but it demands a level of transparency that does not come by default: negative keywords, brand exclusions, and a check on the channel distribution. Taken at face value and left uncontrolled, it often claims credit for brand sales and dilutes the results.

How do I know whether the wasted spend comes from the campaigns or from measurement?

If the reported results in the dashboard look good but real profit is not growing, suspicion falls on measurement first. That is why it gets checked before anything else — optimizing on wrong data only scales the mistake.

Should I stop everything that has no conversions?

Act on data, not on reflex. A query with spend and zero conversions over a long enough period is a candidate for exclusion; but decide on sufficient data and on the trend across several time windows, not on one weak day. Low volume calls for more patience before you cut.

Next step: see exactly where your budget is leaking

If you want to see these checks applied to your own account — which search terms are leaking, where negative keywords are missing, how Performance Max distributes its spend, and whether your measurement is feeding correct data — you can start a free audit. It runs over 500 checks across 12 categories (tracking and measurement, campaign structure, keywords and match types, negative keywords, wasted spend, bidding and budget, location and devices, and more), ranks the findings by their impact on profit, and points out what to fix first.

The audit is read-only — it does not touch your account and does not execute changes. Behind it sits a system with 24/7 monitoring and predictive analysis that watches the account continuously and raises a flag early — when a few euros have been wasted, not when the damage is already on the invoice.

Request your free audit: https://audit.hpanov-digital.com/

Adfure · Ecom Vision LTD · VAT BG208156804 · adfure.com · office@adfure.com