Google Ads is spending but not converting — how to find the cause
Is conversion tracking even working? The status labels, in plain terms
Google shows the status of each conversion action through an interactive troubleshooter: Active, Needs attention, Misconfigured, Awaiting conversions, or Removed. Only "Misconfigured" signals a genuine technical fault. "Awaiting conversions" on an action under 48 hours old is normal, not a bug. What makes this confusing: Google itself uses five more, non-matching status terms for the same underlying problem on other Help pages.
The official interactive conversion tracking status troubleshooter lists five overview statuses:
| Status | What it means, per Google |
|---|---|
| Active | "everything is working as intended" |
| Needs attention | action items to address, hasn't stopped recording entirely |
| Misconfigured | "Conversions have stopped recording entirely due to a setup error or a broken tag" |
| Awaiting conversions | "No conversions have been recorded in the last 7 days" — typically an action under 48 hours old or a paused campaign |
| Removed | "manually deleted or archived and is no longer tracking data" |
The problem: this troubleshooter isn't the only Google page on the topic — and the others use different words. A separate troubleshooting Help page instead refers to "Inactive," "Unverified," or "Needs Attention," and states reporting can take "3-24 hours to update." A third page explains "Unverified" in detail: with high hourly volume, this Help page on the Unverified status says "you can expect the status to update within 10 minutes" — the same page gives a third, different window for "Tag inactive" or "No recent conversions": no conversions attributed to ad clicks "in the last 7-14 days." The same page also still carries the sentence "Starting July 30, 2021, conversion actions that haven't recorded data in the past 13 months will be removed" — live on the page as written, but clearly unchanged since 2021, so don't read it as a current announcement.
Tag Assistant, in turn, uses the labels "no recent conversions" and "Recording," and per the page takes about 30 minutes after debugging for the status to update in the account. And Tag Diagnostics scores tag quality with a sixth, completely different vocabulary — Excellent, Good, Needs attention, Urgent, No recent data — where "Urgent" is triggered "if the Google tag was previously detected but hasn't been detected in the last 48 hours, causing data to drop to zero."
Six vocabularies, at least six different time windows — that's the main reason accounts get misdiagnosed: someone searching for "Inactive" but only aware of the troubleshooter's five statuses won't find the term, and wrongly assumes the account is either fine or broken.
| Time figure | What it means | Source |
|---|---|---|
| 48 hours | Grace period for new conversion actions | Troubleshooter |
| 7 days | Window for "Awaiting conversions" | Troubleshooter |
| 3-24 hours | Reporting delay | Status troubleshooting page |
| 10 minutes | Update time at high conversion volume | "Unverified" status page |
| 7-14 days | Window for "Tag inactive" / "No recent conversions" | "Unverified" status page |
| ~30 minutes | Update after Tag Assistant debugging | Tag Assistant |
| 48 hours | Trigger for "Urgent" in Tag Diagnostics (tag-wide detection, not one conversion action) | Tag Diagnostics |
Worth checking before any of this: setting up web conversions itself requires specific permissions — Google states you need "Admin access in Google Ads and Edit access in the linked Google Analytics property," and that "only events marked as key events in Google Analytics are eligible for import." Missing either permission, or importing an event never marked as a key event, produces gaps that look like a tracking bug but are really an access or configuration gap.
Are you measuring what you think you're measuring? Conversion windows and attribution delay
Without customization, the click-through conversion window defaults to 30 days, the engaged-view window to 3 days, and the view-through window to 1 day. For Search and Display campaigns the window can be set anywhere from 1 to 30, 60, or 90 days — Google itself recommends at least 7 days, because shorter windows systematically miss real, but delayed, conversions.
Google states: "If you don't customize the click-through conversion window when you create a new conversion, the default window is 30 days" (Google Ads Help, About conversion windows). If you're advertising something with a long decision cycle — B2B, high-ticket purchases — and the window has accidentally been shortened to a few days, reports will show "no conversions" even though customers genuinely buy days later; the conversion simply falls outside the measured window. Google explicitly recommends: "windows of at least 7 days are recommended, because they provide a richer set of conversion data" (ibid.). Add the 3-24 hour reporting delay from the section above — checking the account right after a conversion happens often won't show it yet.
Consent mode and modeled conversions — what the numbers actually mean
Without correctly implemented consent mode, part of your conversions are simply missing, because users decline cookie consent. Google models that gap, but only above a minimum threshold: 700 ad clicks per 7 days, country, and domain grouping. Once that threshold is met, modeling recovers on average more than half of the lost click-to-conversion paths — with advanced settings, on average twice as many.
The threshold is hard and documented: "You have a daily ad click threshold of 700 ad clicks over a 7 day period, per country and domain grouping" (Google Ads Help, About consent mode modeling). Accounts with low traffic per country/domain often never reach this threshold — the gap then stays unmodeled, and reports permanently show fewer conversions than are actually happening. Once an account does reach it, a second Google page states: "conversion modeling through consent mode recovers more than half of ad-click-to-conversion journeys lost," and with advanced settings "you can recover 2 times more on average" (Google Ads Help, How to improve your modelling). This recovery isn't a 100% fix: "Google is unable to observe whether an unconsented conversion previously had an ad interaction" — models close the gap only "largely," not completely (consent mode modeling page). In practice, a falling reported CPA or rising conversion count can simply reflect more users consenting, or modeling kicking in — not the campaign performing better. Google's own example: at a 50% consent rate, observed conversions drop 19% (12 of 62), while modeling shows an 18% conversion rate uplift (ibid.). When cookies are declined, Google recommends sending "cookieless pings" for more accurate modeling — pings it states are never used to track users across devices or build remarketing lists.
Does the account have enough conversion volume for automated bidding?
Google says Target CPA can be activated with no conversion history at all — the often-cited "30 conversions" figure is a measurement window, not an activation requirement. Target ROAS, by contrast, has hard minimums that vary sharply by campaign type: 15 conversions in 30 days for Search/Shopping, up to 50 in 35 days for Demand Gen. Without that volume, a campaign stays stuck in the "Learning" phase.
Google separates two things that get blurred in practice: the requirement to turn a bid strategy on, and the recommendation for evaluating it. "Advertisers can start using Target CPA with no conversion history, and Target CPA is effective for campaigns of all sizes" — for evaluation, Google recommends only "measur[ing] performance for the last 30 days," in which "at least 30 conversions" should have been recorded (Google Ads Help, About Target CPA bidding). Target ROAS is stricter, and breaks down by campaign type:
| Campaign type | Minimum conversions for Target ROAS |
|---|---|
| Search & Shopping | at least 15 in 30 days (conversion tracking level) |
| Display | at least 15 total, combined |
| App | at least 10 per day or 300 in 30 days |
| Demand Gen | at least 50 in 35 days, 10 of which in the last 7 days |
| Video action | at least 30 in 30 days |
| Hotel | at least 50 per week (campaign level) |
| Travel | at least 50 in 7 days (campaign level) |
Source for all figures: Google Ads Help, About Target ROAS bidding. Before switching to Target ROAS, Google recommends waiting "4 weeks or 3 conversion cycles" (whichever is longer, until the campaign receives conversion values at a similar rate, ibid.) — a distinct figure from the general learning period. That learning period is triggered when a bid strategy is newly created, reactivated, or has a setting changed, or when campaigns, ad groups, or keywords are added to or removed from it; per Google it takes "up to 3 weeks or 1-2 conversion cycles" to calibrate to the new objective — "although it can be faster depending on the amount of conversion data present" (Google Ads Help, About the learning period). An account that constantly changes bid strategies, budgets, or structure never really exits the learning phase.
Search terms and negative keywords — the budget-leak check
The search terms report shows which actual searches triggered your ads — not just the keywords you bid on. Irrelevant search terms can be added straight from the report as a negative keyword. This isn't a tracking fix; it's a check on whether the budget is even being spent against the right searches in the first place.
"Use the search terms report to know how your ads performed when triggered by actual searches within the Search Network" (Google Ads Help, About the search terms report). "If a search term isn't relevant enough to the products or services you offer, add it as a negative keyword" (ibid.). Negative keywords stop your ad from being served against those terms — though they're "not as precise on the Display Network as they are on the Search Network," and for Display and Video campaigns Google considers "a maximum of 1,000 negative keywords" (Google Ads Help, Add negative keywords to campaigns). No Google Help page officially folds this check into a "why aren't I converting" flow — it lives in separate keyword documentation. That's exactly why it gets skipped in practice, even though it's a fast way to see whether budget is burning on irrelevant queries before a conversion was ever possible.
Performance Max: its own failure points
New or edited ads need 24 to 48 hours of review. For automated bid strategies, Google recommends waiting 7 to 14 days for the system to train. If there are still no impressions after 3 business days, Google points to its own troubleshooting list — and a fluctuating daily budget isn't, by itself, a fault.
"It takes 24-48 hours for newly created or edited ads to be reviewed to ensure they meet all policy requirements" — and "it's recommended to wait 7-14 days to allow the system to ramp up and start to perform towards your set goal" (Google Ads Help, Fix Performance Max campaign not running or low traffic). "If your ad still doesn't receive impressions after 3 business days, check the reasons listed in this article" (ibid.). For interpreting budget swings: "Google Ads may spend up to 2 times your average daily budget on any given day. However, you'll never be charged more than your monthly charging limit" (ibid.) — an expensive single day is, on its own, not a red flag. Per the benchmark behind this guide, only 2 of 22 competitor articles reviewed cover Performance Max at all, despite it being the default recommendation for many new accounts.
What order should you work through this?
Technical status first, then measurement window, then volume, then content — in that order, because each later step is only meaningful once the one before it checks out. Optimizing ads or bids against broken numbers wastes time and budget twice over.
- Check the tracking status. Open the troubleshooter, check each conversion action individually — fix "Misconfigured" first.
- Check the conversion window. Does it match the real purchase cycle? At least 7 days, as Google itself recommends.
- Check consent mode and consent rate. Is the account reaching the 700-click threshold per country/domain? If not, the gap stays unmodeled.
- Check conversion volume against your bid strategy's minimums. Target CPA runs without history; Target ROAS doesn't.
- Review the search terms report and plug the budget leak. Exclude irrelevant search terms before spending more.
- For Performance Max: respect the timelines before touching assets or goals — 24-48 hours of ad review, 7-14 days of training time.
Our free account audit as the next step
In our free account audit, we walk through this exact order with you: tracking status, conversion windows, consent mode coverage, conversion volume per bid strategy. We're a small, founder-led remote team; our Google Ads management starts at €349/month, month-to-month, no setup fee — full details on the pricing page. If the honest answer is "the account needs more time or budget first," we say that as plainly as the opposite. For realistic budget sizing, see our guide to what Google Ads actually costs.
FAQ
How do I know if my Google Ads tracking is broken? Open the status of each conversion action in the troubleshooter. "Misconfigured" means conversions have stopped recording entirely — a genuine technical problem. "Awaiting conversions" on an action under 48 hours old is, by contrast, normal.
Why does my account show "no conversions in the last 7 days" despite clicks? That's the standard "Awaiting conversions" status — per Google, typical for new actions under 48 hours old, paused campaigns, or too little traffic, not necessarily a fault.
What does consent mode have to do with missing conversions? Without consent, part of your real conversions are missing from reports. Google models that gap above 700 ad clicks per 7 days, country, and domain — below that threshold, it stays unmodeled.
How many conversions does Google Ads need to bid automatically? Depends on the strategy: Target CPA runs with no history at all, Target ROAS needs 15 to 50 conversions in a defined period depending on campaign type.
What do you check in the free account audit? We check tracking status, conversion windows, consent mode coverage, and conversion volume against your bid strategy's requirements — and say plainly what, if anything, needs fixing first. Book a call.