Ads Not Showing in Google Ads: 12 Causes to Check

The campaign is live, every status is green, and your ads are not showing. Or there are dozens where you planned thousands. The first reflex is to raise the bid, the second is to raise the budget, and both are usually wasted — in roughly half of these cases the auction is not the problem at all.

Below are twelve reasons ads do not show, arranged in the order you should check them: from what takes a minute to fix to what takes a week. The order is deliberate — it runs from the most frequent and cheapest to diagnose toward the rarest and most expensive.

First, confirm there really are no impressions

Before hunting causes, rule out three situations where no problem exists:

You are reading the wrong date range. “Today” updates with a delay, and some campaign types take up to a day to report fully.
You are searching for your own ad in Google. The worst possible check: your own impressions accumulate without clicks, distort CTR and frequency, and the result is unrepresentative anyway. The Ad Preview and Diagnosis tool exists for this — it tells you whether you are eligible for a given query and, if not, why.
There are impressions, just fewer than expected. That is not “not showing”, it is an impression share question, and it is treated differently — see impression share and auction insights.

Rule: test every hypothesis about missing impressions with the diagnosis tool, never with your own browser search.

Reading the Ad Preview and Diagnosis tool properly

The tool is not there to show you what your ad looks like. It answers “am I eligible for this auction, and if not, what is blocking me?” Used carelessly it returns a useless answer.

Set every parameter explicitly. Query, location, language, device. By default it assumes your current context, while you are usually checking a different city on a different device.
Test a specific query, not a topic. “buy running shoes boston” and “running shoes” are different auctions with different outcomes.
Read the stated reason. The tool names the class of problem outright: bid below the first-page estimate, keyword inactive, ad under review, policy restriction, campaign limited by budget. That saves an hour of guessing.
Test several queries from different clusters. If one query fails and three succeed, the problem is local — one ad group or keyword — not account-wide.

One caveat: the tool answers for the moment you run it. If the budget runs out by lunchtime, a morning check looks clean and an evening check reports a budget limitation. Run it at the time of day when you observe the problem.

Group A. Money and statuses — five minutes

1. Billing problems

A declined payment, an expired card, a hit spending limit, an account suspended over billing — all while campaigns still look active. Always start here: payment settings, account status, outstanding balance. Number one by frequency and number one by speed of resolution.

2. Something is paused higher up

The ad is enabled, the group is enabled, the campaign is enabled — and the parent account in the hierarchy is paused, or the conversion action the bidding strategy depends on is. Walk the whole chain: account → campaign → ad group → ad → keyword. Paused keywords are more common than anyone expects.

3. Dates and ad schedule

An end date in the past, an ad schedule covering hours that have already passed, or a schedule running on an account time zone that differs from the target market. A classic variant: weekday 9-to-5 scheduling in a market whose demand peaks in the evening. The mechanics are in ad scheduling and dayparting.

Group B. Ad statuses and platform requirements

4. Ads disapproved or limited

The distinction matters. Disapproved means no impressions at all. Eligible (limited) means impressions happen, but not everywhere: not in certain countries, not for certain audiences, not on certain queries. The limited status renders green, which is exactly why it gets skimmed past.

Check at the ad level and again at the asset level: an ad group can be active while half its headlines and descriptions are disapproved. Disapproved assets do not enter the ad build, which costs both reach and quality — see ad assets and extensions.

5. Limited ad serving

A separate state applied to accounts without sufficient history or with outstanding platform requirements. Ads serve far less often than the budget allows, while statuses look entirely normal. The signature is recognisable: spend plateaus well below the daily budget, impression share is low, and there is not a single disapproval in sight. How it works and what lifts it is covered in limited ad serving.

6. Advertiser verification not completed

Identity and business verification is now mandatory for most categories. Until it is complete, some formats and geographies stay unavailable and serving is throttled. There is no workaround here — there is a procedure to complete correctly the first time, with documents that match the account details exactly. We handle it as a service through our agency account setup, but with careful document preparation it is entirely doable in-house.

Group C. The auction: bid, budget, quality

7. Bid below the entry threshold

Every auction has a minimum bid below which an ad is simply not admitted for a given query, position or format. The diagnosis tool says so directly when this is the cause. Do not raise bids blind: check what the simulator projects at different levels first — cheaper than experimenting with live budget. See bid, budget and target simulators.

8. Budget exhausted or spread too thin

Two different scenarios. First: the daily budget runs out by noon and the afternoon is silent. Second: budget is split across so many campaigns that none can meaningfully enter the auction. The tell for the second is a dozen campaigns each collecting a handful of impressions a day and none accumulating any usable data.

9. Weak quality and landing page experience

Quality affects auction eligibility, not just cost per click. A loose query → ad → page chain leaves you technically in the auction and practically never winning it. Page performance is a factor of its own: a slow landing page hurts both quality signals and conversion rate.

10. The campaign is in learning

After a bidding strategy change, a large budget move or a structural edit, a campaign re-enters learning, and impression volume can dip temporarily. That is not a fault. The mistake is stacking new edits before it stabilises — each significant change restarts the clock. What resets learning is listed in the Smart Bidding learning period.

Group D. Targeting and demand

11. Stacked restrictions collapsed your reach

Each restriction looks reasonable on its own, and together they leave an addressable audience of two hundred people: narrow geography, plus an age bracket, plus one device, plus a schedule, plus exclusion lists, plus exact match. Evaluate the product of your constraints, not each one separately.

Check the presence-versus-interest location setting specifically: “people in your targeted locations” and “people regularly in or who showed interest” produce very different volumes. And audit exclusion lists — occasionally an exclusion covers the entire target audience.

12. The demand simply is not there

Keywords marked “Low search volume” do not serve until volume rises. The same goes for ultra-narrow clusters: at 30 searches a month, zero impressions is not a configuration problem. Volume should be checked before launch — see Keyword Planner and sizing a keyword set.

Symptom to cause: a diagnostic table

What you see	Likely causes	Where to look
Zero impressions account-wide	1, 2, 4	Billing, statuses, disapprovals
Zero impressions in one campaign	2, 3, 11, 12	Schedule, targeting, search volume
Impressions far below plan	5, 7, 8, 9	Limited serving, bid, budget, quality
Spend never reaches daily budget	5, 7, 11, 12	Serving limits, bid threshold, reach
Impressions stopped abruptly	1, 3, 4	Payment, end date, new disapproval
Impressions fell after edits	10	Learning period — give it time
Impressions only in the morning	8	Daily budget exhausting early

Ads not showing: why the order matters more than the list

Everyone has a list of twelve causes. The difference between a thirty-minute diagnosis and half a day of wandering is discipline about sequence. The logic: check what makes serving physically impossible first (money, statuses, dates), then what restricts eligibility (policy, platform requirements), then the auction (bid, budget, quality), and only then targeting and demand.

Breaking that order has a measurable price. The typical story: someone starts by widening targeting and raising bids, burns two days and part of the budget, then discovers a disapproved ad — meaning no bid would have produced impressions in that window anyway. Worse, those edits restarted learning, so after the real fix the campaign spends another week climbing back to normal volume.

Every edit made before the diagnosis adds a week to the recovery. Diagnosis costs thirty minutes; premature edits cost weeks.

PMax and Demand Gen behave differently

In automated formats there are no keywords to inspect, but there are asset and signal requirements.

Insufficient assets. If an asset group falls short of minimum headline, description, image and logo requirements, entire placements are physically unavailable. That shows at asset-group level, not campaign level.
Disapproved images or video. A single rejected asset can close off a whole channel.
Over-narrow audience signals and aggressive exclusions. The system uses signals as a starting point rather than a boundary, but heavy exclusions genuinely restrain it.
Minimum budget requirements. Several formats have a floor below which a campaign effectively never enters delivery.
Missing or failing product feed. For shopping-driven campaigns, an unapproved feed means zero impressions with a fully green campaign.

Three real-world patterns

“New account spending $3 a day on a $100 budget”

Statuses clean, no disapprovals, bids at market. The usual cause is limited ad serving on an account with no history and outstanding platform requirements. Raising bids does nothing here, because the constraint is not the auction. The fix is completing verification, letting the account build history on a small steady budget, and avoiding sudden spend jumps.

“It ran for a week, then hard zero”

A cliff points to an event, not a process. Check three things in order: payment (a declined card is the most common), campaign end date, and fresh disapprovals from an automated re-review. If all three are clean, read change history — something upstream may have been paused.

“Only one campaign out of five is starved”

Since the problem is local, look for what makes that campaign different: geography, schedule, match types, its own exclusion lists, a narrower audience, a separate budget. Nine times out of ten a five-minute side-by-side settings comparison finds it.

The first 24 hours after launch

Hour 0. Verify statuses at every level and payment status. Confirm dates and schedule.
Hours 1–2. Run three to five core queries through Ad Preview and Diagnosis. It either confirms eligibility or names the blocker.
Hours 3–6. Review ad and asset statuses for disapprovals and limitations.
End of day. Compare actual spend with the daily budget. Near-zero spend with active statuses means the answer is in Groups B and C — not in a higher bid.
Days 2–3. If impressions arrive but conversions do not register, verify measurement before changing campaigns — the order is in diagnosing conversion tracking.

And the governing rule of the first week: change one thing at a time. Otherwise you will not know what helped, and the bidding strategy will have restarted learning several times over.

One-page checklist

Billing: active method, no declined payments or outstanding balance.
Statuses: account, campaign, ad group, ad and keyword all enabled.
Dates and schedule: end date not in the past, time zone matching the target market.
Ads: no disapprovals, no “limited” status where it matters.
Assets: minimum requirements met, nothing rejected.
Platform requirements: verification complete, no serving limitations.
Budget: spend does not cap out mid-day; budget is not spread across dozens of campaigns.
Bid: the diagnosis tool does not report a below-threshold bid.
Targeting: geography, devices, audiences and exclusions reviewed together, not individually.
Demand: cluster volume is sufficient for the expected impression count.
Edits: nothing in the last 7–14 days that restarted learning.
Measurement: conversions record correctly, otherwise you are diagnosing volume against bad data.

When the account itself is the constraint

Some limitations come from account state and history rather than campaign settings: a brand-new account with no spend record, unmet platform requirements, limited serving. None of that is solved with bids. It is solved by building history deliberately, completing verification properly, and operating inside the platform’s rules. If the goal is to start on a structure with confirmed billing and a clean hierarchy, our Google Ads agency accounts cover that, and the broader operating approach is what our agency team works from.

FAQ

The campaign is enabled but has zero impressions. Where do I start?

Billing and statuses across the whole hierarchy, then the Ad Preview and Diagnosis tool on a specific query. Those two steps close most cases in ten minutes.

Why shouldn’t I check by searching Google myself?

Your own impressions accumulate without clicks, damage CTR data, and the result is unreliable anyway because of personalisation and location. The diagnosis tool answers without the side effects.

What does “Eligible (limited)” mean?

The ad may serve, but with restrictions: not in all countries, not to all audiences, or not on all queries. You will get impressions — noticeably fewer than expected.

How long should first impressions take after launch?

Ad review typically completes within one business day. If a day passes with active statuses and healthy billing and still nothing, run the full diagnosis.

What do I do with a “Low search volume” keyword?

Leave it in the account and do not build a campaign on it. The status clears automatically if volume rises. Add reach with broader phrasing, not with bids.

Will a higher bid help when the budget is exhausted?

No — the opposite. With a capped budget, a higher bid buys fewer clicks for the same money. Fix budget and allocation first, bids second.

Why doesn’t spend reach the daily budget?

Usually narrow reach, a bid threshold, or limited ad serving. The daily budget is a ceiling, not a commitment by the system to spend it.

Impressions dropped right after I changed bidding strategy. Normal?

For the first days, yes — that is learning. It is not normal if the dip persists beyond two weeks, or if you keep editing and restarting learning each time.

Can one disapproved asset stop an entire ad group?

The whole group, rarely — but reach and quality always suffer, because the system has fewer valid combinations to build from. In formats with mandatory asset requirements, missing approved assets can close entire placements.

How do I tell quality from bidding as the cause?

By the combination: impressions exist, impression share is low despite a high bid, and relevance and landing page experience metrics sit below average. If raising the bid barely moves reach, the bid is not the constraint.

Does the presence-versus-interest setting really change volume?

Substantially. “People in your targeted locations” delivers far less reach than the option that includes interest in the location. In narrow geographies it is a frequent cause of near-zero volume.

What if none of the twelve causes fit?

Assemble the evidence: status screenshots, the diagnosis tool output for a specific query, impression share and spend data, and a list of edits in the period. That set usually surfaces the cause in one pass — and it is exactly what support will ask for anyway.