Skip to main content
MethodMarketingPaid advertisingIntermediate

Find out why your ads are not spending or getting results

A step-by-step diagnosis for Google Ads and Meta campaigns that are not spending, stuck in learning or getting too few results, with the most likely cause, the one change to make first and when to check again.

Use this task 5 quick questions, then open it in ChatGPT, Claude or Gemini.

This is a practical, copy-paste process for a campaign has run for at least 3 days and spends well under its daily budget, has almost no impressions, or shows a limiting status. Use it when you want a clear structure, then add your own details.

Verified on 2026-10-2810 sources cited

Quick start

1. Check the boring things first: payment, approval, start and end dates, schedules and spending limits.
2. Read the status words on every level. "Limited by budget" means the budget is too small; "Cost limited" or "Bid limited" means your target is too low; "Learning limited" means too few results a week to learn.
3. Divide your daily budget by the cost of one result. Google suggests a budget of at least 2 to 3 results a day (or 10 clicks a day). Meta suggests at least 10 times the cost of one result, and about 50 results a week to finish learning. If you are well below that, the budget, not the ads, is the likely problem.
4. A target lower than what results really cost will stop your ads. Remove it or raise it.
5. On Meta, fewer ad sets learn faster. Combine similar ones.
6. Make one change, write down the date, and wait one learning window before changing anything else: about 7 days on Meta, and on Google the usual time from click to the result you judge on, at least a week (rule of thumb).

Before you start

Full method

How to use this method

  1. 1. Copy or downloadCopy the prompt below, or download the method file and attach it to your chat.
  2. 2. Paste into any AIChatGPT, Claude, Gemini or Copilot. Replace each [PLACEHOLDER] with your details, or let the AI ask you.
  3. 3. Add your materialAttach your own files, data or brief so the AI works from your facts, not guesses.

The prompt

Paste it as-is. The AI will ask for anything still in [BRACKETS].

Download .md
# Find out why your ads are not spending or getting results
Method by Deepstory Research: deepstoryresearch.com/method/ads-not-spending-diagnosis

You are a paid-ads troubleshooter who diagnoses delivery problems in Google Ads and Meta accounts. Follow the method below step by step. Goal: A step-by-step diagnosis for Google Ads and Meta campaigns that are not spending, stuck in learning or getting too few results, with the most likely cause, the one change to make first and when to check again.

## My inputs
- Platform and campaign type: [PLATFORM_AND_CAMPAIGN_TYPE]  (example: Google Search, Maximize conversions)
- How long it has been running: [HOW_LONG_IT_HAS_BEEN_RUNNING]  (example: 4 days since launch)
- Status shown: [STATUS_SHOWN]  (example: Eligible (limited), limited by budget)
- Budget and bid settings: [BUDGET_AND_BID_SETTINGS]  (example: AED 50 a day, Target CPA AED 40)
- Results so far: [RESULTS_SO_FAR]  (example: 120 impressions, 2 clicks, 0 leads)

Before you start: if any input above is still in [BRACKETS] or missing, ask me for it in one short numbered list and wait. Do not guess. If I attached files (data, briefs, Deepstory evidence files), treat them as the primary source.

## Steps to follow
1. Check the basics first: approval status, account or billing problems, dates, schedules and spending limits.
2. Read the exact delivery status on each level (campaign, ad set or ad group, ad).
3. Compare the budget with the cost of the goal: can it buy enough results to learn?
4. Check bid targets and cost controls against recent real costs.
5. Check targeting size, overlaps between campaigns, and conversion tracking.
6. Name the most likely cause and one change to make first.
7. Set a waiting time before judging the change, and the next check if it does not work.

## Deliver the output in this structure
- Basics checklist with results
- Delivery status by level
- Budget versus goal math
- Most likely cause
- First change and why
- When to check again and what next

## Quality rules
- Change one thing at a time and say how long to wait.
- Never recommend raising a budget without showing the maximum it could spend.
- Do not call learning normal if the status is "Learning limited"; explain the cause.
- Label every number as FACT (with source and date), ESTIMATE (show the calculation) or ASSUMPTION.
- If a Deepstory evidence file is attached, use its figures as the reference values and cite its source references; tag claims that are not from it as [UNVERIFIED].
- End with: the 3 assumptions that matter most, what would change the answer, and the single best next step.

What you need to give the AI

InputPlaceholderExample
Platform and campaign type[PLATFORM_AND_CAMPAIGN_TYPE]Google Search, Maximize conversions
How long it has been running[HOW_LONG_IT_HAS_BEEN_RUNNING]4 days since launch
Status shown[STATUS_SHOWN]Eligible (limited), limited by budget
Budget and bid settings[BUDGET_AND_BID_SETTINGS]AED 50 a day, Target CPA AED 40
Results so far[RESULTS_SO_FAR]120 impressions, 2 clicks, 0 leads

Use it when

  • A campaign has run for at least 3 days and spends well under its daily budget, has almost no impressions, or shows a limiting status.
  • A new campaign has run for days with no or very few impressions.
  • Spend is far below the daily budget.
  • Meta shows "Learning limited", "Bid limited" or "Creative fatigue", or Google shows "Limited".

Best with: Any chat AI; paste the status text and a screenshot or export of the campaign, ad set and ad rows for the last 7 days.

What you get

  • Basics checklist with results
  • Delivery status by level
  • Budget versus goal math
  • Most likely cause
  • First change and why
  • When to check again and what next

The AI will ask you first

Answer these before the AI advises you; each one changes what it recommends.

  1. 1

    Has the campaign spent nothing at all, or does it spend but get few results?

    The checks differ: spending nothing points to blockers, review, dates and limits; spending with few results points to learning, bids, budget size and creative.

    If unknown: Ask for spend and results for the last 7 days.

  2. 2

    Which platform and campaign type, and how many days since launch or the last big change?

    Google reviews new or edited ads within about 1 to 2 days, and Meta can take up to 24 hours to start delivery; judging earlier leads to wrong fixes.

    If unknown: Ask for the launch date and the date of the last edit.

  3. 3

    What does the Status or Delivery column say at campaign, ad set or ad group, and ad level?

    Each status points to a different cause: disapproval, budget, bid limit, learning limited, creative fatigue.

    If unknown: Ask the user to copy the status words from each level.

  4. 4

    What is the daily budget, and is there a target CPA or ROAS, cost per result goal or bid cap? On Meta, is the budget set on the campaign (Advantage+ campaign budget) or on each ad set?

    A target below recent real costs, or a budget too small for the goal, stops delivery; the budget level decides the per-ad-set maths.

    If unknown: Ask the user to open the bid strategy settings and copy the strategy name and any target, goal or cap; do not assume there is none.

  5. 5

    What does one result cost, or what do you expect it to cost?

    Budget versus cost decides whether the campaign can get enough results to learn.

    If unknown: Use the cost of the chosen goal from the last 30 days, or ask for a rough estimate.

  6. 6

    Are there billing problems, account warnings, spending limits or schedules set?

    These stop all spend regardless of anything else.

    If unknown: Ask the user to check billing and account notices first.

  7. 7

    Is conversion tracking recording real leads or sales?

    Conversion bidding needs conversion data; without it campaigns serve very little.

    If unknown: Treat tracking as unconfirmed and check it.

  8. 8

    How big is the audience or location, and do other campaigns target the same people or keywords?

    Narrow targeting and overlap between your own campaigns limit delivery.

    If unknown: Ask for the location, audience size estimate and number of active campaigns.

The procedure, step by step

Read this before you run it, so you can check the AI actually followed the method.

  1. 1

    Rule out blockers

    Approval status, billing and payment, account warnings, start and end dates, schedules, account spending limit, Google account budget, campaign and ad set spending limits.

  2. 2

    Read every status

    Google: Status column at campaign, ad group, ad and keyword level, plus Recommendations. Meta: Delivery column at campaign, ad set and ad level.

  3. 3

    Do the budget math

    Daily budget divided by cost per result = results a day. Compare with Google (10x CPC or 2 to 3x Target CPA) and Meta (10x goal cost; about 50 events a week to exit learning).

  4. 4

    Check bids and targets

    Compare any Target CPA, ROAS, cost per result goal or bid cap with the last 30 days of real costs.

  5. 5

    Check reach

    Match types and negatives, audience size, location, overlap between your own campaigns or ad sets.

  6. 6

    Check tracking

    Is the conversion goal recording real leads or sales? If not, fix tracking before changing bids.

  7. 7

    Make one change

    The change that fits the most likely cause. State the maximum it could spend: Google up to 2x the daily budget on a day and 30.4x in a month (or the total budget, if the campaign uses one); Meta up to 1.75x on a day and 7x in a week.

  8. 8

    Set the next check

    Date to review, what to compare, and the second most likely cause to test if this fails.

Platform settings(show)
PlatformPathSettingNote
Google AdsCampaigns > RecommendationsRecommendation types such as "Campaign hasn’t started or has ended", "All ads disapproved", "Adjust your budgets"
Google AdsCampaigns > Status columnEligible, Eligible (limited), Limited by budget, bid strategy status
Meta Ads ManagerCampaigns, Ad sets, Ads > Delivery columnLearning, Learning limited, Bid limited, Cost limited, Creative limited, Creative fatigue
Meta Ads ManagerColumns > Customize columns > Optimization eventsOptimization events columnShows whether an ad set is getting enough events to exit learning.
Meta Ads ManagerCampaign or Ad set > Budget & scheduleSpending limits, lifetime budget, scheduleChanges usually take effect within 15 minutes of publishing, but delivery can take a few hours to improve; do not re-edit in that window.

Worked example

The figures below are illustrative. Use your own numbers and facts.

Scenario: An Indian B2B SaaS company’s new Google Search campaign and Meta Leads campaign, 5 days after launch.

Input: Google Search: Maximize conversions with Target CPA INR 1,500, daily budget INR 2,000, exact match only, 0 conversions recorded so far, 140 impressions. Meta Leads: ad set budgets totalling INR 1,500 a day, cost per lead about INR 600, 3 ad sets with similar audiences, all "Learning limited".

Output excerpt: Blockers: billing and approvals fine. Google: 140 impressions in 5 days. The INR 1,500 Target CPA is a guess (0 conversions, no history), and the budget is about 1.3 times that target, below Google’s guideline of 2 to 3 times (INR 3,000 to 4,500). Step 0, checks rather than changes: read the keyword statuses (an exact-match campaign with 140 impressions may be "Low search volume"), and submit a test form to confirm a conversion records within a day. If the keyword statuses are "Eligible" (ASSUMPTION here), the most likely cause is a conversion strategy with a target and no data. First change: switch to Maximize Clicks for 5 to 7 days (Google’s advice for new campaigns with limited data). Maximum spend after the change: the full INR 2,000 a day, up to INR 4,000 on a single day and no more than INR 60,800 in a month. Check again in 7 days; if impressions are still low, look at keyword statuses (for example "Low search volume") and test phrase match next. Meta: INR 1,500 a day buys about 2.5 leads a day, about 17 to 18 a week, across 3 ad sets with overlapping audiences, so each ad set gets about 6 a week, far below about 50. Meta’s rule of thumb is a daily budget of 10 times the goal cost (INR 6,000); about 50 leads a week would need about INR 4,300 a day. First change: combine the 3 ad sets into one with all the ads (the new ad set starts learning). Even combined, 17 to 18 a week is below 50, so "Learning limited" may stay; judge on cost per qualified lead, and do not edit again for 7 days. Maximum spend after combining (one ad set budget of INR 1,500, so ad set budget sharing no longer applies): up to INR 2,625 on a day and no more than INR 10,500 a week.

Check your result

Before you trust the AI's output, confirm each of these.

  • ☐ Blockers (billing, approval, dates, limits) were checked first.
  • ☐ The status at every level was read, not guessed.
  • ☐ Budget versus goal-cost math is shown.
  • ☐ Exactly one change is recommended first, with a waiting time.
  • ☐ The maximum possible spend is stated for any budget increase.
What you'll get(show template)
## Delivery diagnosis: [CAMPAIGN]

Platform: [Google / Meta] | Running since: [date] | Last edit: [date]

### Blockers checked
Billing: [ok/issue/not checked] | Approval: [ok/issue/not checked] | Dates and schedule: [ok/issue/not checked] | Spending limits: [ok/issue/not checked] | Tracking: [records real results / counts [what] / not confirmed]

### Not known yet
[answers the user did not know; mark the diagnosis provisional where they matter]

### Status by level
Campaign: [status] | Ad set / ad group: [status] | Ads: [status]

### Budget versus goal
Daily budget [x] / cost per result [y] = [z] a day, [w] a week. Platform guideline: [Google 10x CPC or 2 to 3x tCPA / Meta 10x goal cost, about 50 a week]

### Most likely cause
[cause] because [evidence]

### First change
[one change] | Maximum spend after change: [amount a day / week / month]

### Check again
[date]: compare [metric] | If no better: test [second cause]

Quality checks built into the prompt

  • • Change one thing at a time and say how long to wait.
  • • Never recommend raising a budget without showing the maximum it could spend.
  • • Do not call learning normal if the status is "Learning limited"; explain the cause.
  • • Every number labelled FACT, ESTIMATE or ASSUMPTION; claims outside attached evidence tagged [UNVERIFIED].

Decision rules

If/then rules the AI should apply. Heuristic rules are a practitioner's rule of thumb, not a platform rule.

IfThenType
A Google campaign is new or ads were just edited.New or edited ads go through a 1 to 2 day policy review, and automated bidding needs time to learn. If there are still no impressions after 3 business days, work through the common causes.Platform rule
A Google campaign shows no impressions.Check in this order: account suspended or billing issue; campaign dates; paused or disapproved ad groups, ads or assets; low Ad Strength; bid targets too low; budget too low; targeting too narrow; overlap with your other campaigns; conversion tracking; auction competition.Platform rule
A Google campaign is held back because the Target CPA is too low.Google suggests raising the Target CPA by 10% to 20% (or lowering the Target ROAS), then waiting at least 1 to 2 conversion cycles before judging.Platform rule
A Google campaign is "Limited by budget" or barely serving on a small budget.Google’s guideline is a daily budget of at least 10 times the expected cost per click, or 2 to 3 times the Target CPA; a modest increase (about 20%) can help impression volume.Platform rule
A new Google campaign has little or no conversion data and will not serve on conversion bidding.Google suggests Maximize Clicks or Manual CPC for 5 to 7 days to gather data, then a conversion strategy once there is enough data (for example 30 or more conversions in 30 days). See "Set ad budgets and choose a bid strategy" (https://deepstoryresearch.com/method/ad-budget-and-bid-strategy) for what to switch to once data exists.Platform rule
A Google campaign has only exact match keywords, many negatives, or audiences set to "Targeting".Targeting may be too narrow: add phrase or broad match with negatives, review negatives that block real searches, set new Search audiences to "Observation", and consider "Presence or Interest" location targeting where it fits.Platform rule
A Meta campaign has spent nothing.Allow up to 24 hours; "Preparing ad delivery" means it passed review. Then check campaign and ad set spending limits, lifetime budget, budget schedules, restrictive hours, bid cap or cost per result goal, the performance goal’s cost, and estimated audience size.Platform rule
A Meta daily budget is small compared with the cost of the performance goal.Meta’s rule of thumb is a daily budget of at least 10 times the average cost of the performance goal; otherwise choose a goal higher in the funnel (for example landing page views instead of conversions), but only if it still predicts sales; otherwise accept "Learning limited" and judge on cost per qualified result.Platform rule
A Meta ad set shows "Learning limited".It is unlikely to get about 50 optimisation events in the week after the last significant edit. Causes: small audience, low budget, low bid or cost control, high auction overlap, an infrequent event, or too many ads. Fixes: combine ad sets and campaigns, expand the audience, raise the budget or cost control, or choose a more frequent event.Platform rule
A Meta ad set shows "Learning".Wait before editing; it usually ends after about 50 optimisation events in the week after the last significant edit. Edit only if it is not getting enough events.Platform rule
A Meta ad set shows "Bid limited" or "Cost limited".The bid control or cost per result goal is holding delivery back: switch to highest volume, or raise the goal.Platform rule
A Meta ad or ad set shows "Creative limited" or "Creative fatigue", or frequency is high.Meta shows "Creative limited" when cost per result is above your past ads but under twice as much, and "Creative fatigue" when it is twice or more; the recommendation is shown only for ad sets with one creative. Add a new, materially different image or video and keep the original running, or expand the audience. Adding an ad restarts learning.Platform rule
You plan an edit on Meta.Targeting, creative, optimisation event, a new ad, a bid strategy change, or pausing for 7 days or more restart learning. Budget, bid or cost goal amounts and ad set spending limits may or may not, depending on size. With Advantage+ campaign budget, a campaign budget or bid strategy change can send several ad sets back into learning. Batch changes.Platform rule
A Meta ad has been published for at least 48 hours and no check fixes it.Meta recommends deleting and recreating that ad; if more than one ad in the campaign is not delivering, contact Meta support.Platform rule
Some Meta ads in an ad set get little delivery while others spend.That is normal: the system favours the ads it predicts will do best. Use an A/B test if you need an equal split.Platform rule
You state the maximum spend after a change on Google.A campaign can spend up to 2 times its average daily budget on one day, but no more than 30.4 times it in a month. Since 1 June 2026, campaigns whose ad schedule switches off whole days still pace towards 30.4 times the daily budget a month, so they spend more on the days they run. A campaign on a campaign total budget has no daily cap: it can spend more than 2 times its daily average on a day, but is never billed more than the total.Threshold: 2x a day; 30.4x a month; total budget if one is setPlatform rule
You state the maximum spend after a change on Meta.Meta may spend up to 75% over a daily budget on a day (175%) but no more than 7 times the daily budget in a Sunday to Saturday week; with ad set budget sharing on, up to 210% a day and 8.4 times a week.Threshold: 175% a day; 7x a weekPlatform rule
Google keywords show "Low search volume".Those keywords are inactive until searches pick up; broaden the match type or use more common wording.Platform rule
Google shows "Account budget exhausted".Nothing can spend until the account budget is raised or its end date is extended.Platform rule
The primary conversion counts every form fill and most leads are not qualified.This is not a delivery problem: bidding is learning to find cheap form fills. Do not change bids, budgets or match types first; change what is counted (for example import qualified leads), and judge on cost per qualified lead.Heuristic
Status, spend per campaign or tracking cannot be read yet.Make no change. First do a read-only check: for the last 30 days, export each campaign’s status, bid strategy and target, cost and conversions, and confirm the main conversion action is recording. Diagnose after that.Heuristic
Several fixes seem possible.Make the single change that addresses the most likely cause, write down the date, and wait one full learning window (about 7 days on Meta; 1 to 2 conversion cycles on Google, meaning the usual time from click to the conversion you judge on) before the next change.Heuristic

Failure modes

Common mistakes with this method, how to spot them, and how to fix them.

MistakeSymptomFix
Changing several settings at once.Impossible to tell what worked; Meta keeps re-entering learning.One change, a date, and a waiting window.
Setting a target below real costs.Google "Limited" or Meta "Cost limited"; very low spend.Remove the target or set it near recent actual cost.
Splitting a small budget across many ad sets or campaigns.Every ad set "Learning limited".Combine ad sets and campaigns.
Judging a campaign on day 1 or 2.Good campaigns paused before review and learning finish.Wait 3 business days on Google and up to 24 hours plus learning on Meta.
Ignoring billing, spending limits and schedules.No spend at all, everything looks "Active".Check account notices, payment, limits and schedules first.
Conversion bidding with broken tracking.Almost no impressions on Maximize conversions.Fix tracking, or use a click-based strategy until data exists.
Pausing a Meta ad set for a week "to reset".Learning restarts on unpause.Lower the budget instead of pausing, or make a planned significant edit.
Panicking at a high-spend day after a fix.One day spends well over the daily budget and the owner cuts the budget.Judge spend by the week on Meta and the month on Google; that is normal budget flexibility.
Deleting and relaunching campaigns or ad sets to "reset" delivery.Every relaunch starts learning again; no history builds up.Only delete and recreate a single Meta ad after 48 hours with no fix, as Meta advises; otherwise fix the cause.
Overlapping campaigns or ad sets competing for the same people.High auction overlap; one campaign starves the other.Merge them or separate keywords and audiences.

Agent mode: if an AI agent runs this unattended

May do without asking

  • • Read statuses, delivery columns, recommendations and settings in a read-only session.
  • • Calculate budget versus goal cost and results a week from exported data.
  • • Draft a change log and the recommended change for the user to apply.

Must stop before (the user decides and does these)

  • • Changing any budget, bid, target or bid strategy, because it changes spend.
  • • Changing keywords, match types, negatives, locations, audiences or any other targeting, because it changes who sees the ads and what they spend.
  • • Editing, adding or uploading any ad, creative or asset, because on Meta it restarts learning and on Google it triggers a new review.
  • • Changing conversion actions, primary goals, pixel or tracking settings, because bidding depends on them.
  • • Pausing, enabling, deleting or recreating any campaign, ad set or ad.
  • • Applying recommendations, or turning on auto-apply.
  • • Publishing any new ad or ad set.
  • • Changing spending limits, billing or payment settings, or accepting terms.

Evidence for this method

This method runs on your own material: your files, brief or code. If your task needs market or economic numbers, attach a sourced context file from the evidence library alongside the prompt.

Browse evidence

Run next

Methods that chain well after this one.

Deepstory methods are process guidance for AI tools. Review AI output before relying on it; legal, tax, medical and investment decisions need a qualified professional.