Meta Ads Agency Results: How Long Should You Wait?

1. Meta Ads Results Arrive on a Calendar, Not a Countdown

Quick Answer: How long do meta ads take to work depends less on the clock than on the sequence. Each stage unlocks the next: access, then tracking, then delivery, then enough conversions for the system to optimise. Skip a step and the wait resets, whatever the calendar says.

Most owners ask the question as if there were a single number. There isn't. A new Meta engagement moves through four gates, and each one has to close before the next opens. That is why two businesses starting on the same Monday can be three weeks apart by mid-month.

This page is about the engagement calendar — what should exist by week two, week six and week twelve of working with a Meta ads agency, and who is accountable when a milestone slips. If you want to judge the ads themselves rather than the timeline, our guide on how to tell whether your Facebook ads are really working covers that separately. ZenWeb has onboarded enough Malaysian SME accounts to know which weeks are genuinely slow and which are being managed badly.

Key takeaway: Judge the sequence, not the stopwatch. A campaign live on day 10 with broken tracking is further behind than one live on day 18 with clean tracking.

Before the week-by-week breakdown, this short explainer covers the learning phase — the part of the wait that most owners mistake for the agency doing nothing.

2. What Should Happen in Weeks 1 and 2?

Quick Answer: Weeks one and two are plumbing, not performance. Partner access granted, Pixel and Conversions API verified against live events, audiences rebuilt, first creative approved, campaigns live. No lead number from this window is meaningful, but every missing item here delays everything after it.

The deliverable in this fortnight is a working machine, not a result. Ask for evidence of each item rather than a status update — a screenshot of the Events Manager firing a real purchase beats "tracking is done" in an email.

• Partner access, not ownership. Your Business portfolio, your ad account, agency added as partner. Our page on what to prepare for Meta ads agency onboarding lists the full day-one pack.
• Pixel and CAPI verified. Not installed — verified, with a test event visible. The Pixel and Conversions API setup guide shows what a clean check looks like.
• Audiences rebuilt. Customer list uploaded, website Custom Audiences defined, exclusions set.
• First creative set approved. Three to five concepts, not one. See what makes Facebook ad design sell.
• Campaigns live by day 14. If they are not, the reason should be in writing.

Clients cause more of these delays than agencies do, which is why the honest scope in what a Meta ads agency actually does each month matters before you start counting days.

Key takeaway: Nothing in the first fortnight is a result. Judge it on evidence of setup — verified events, approved creative, a live campaign — and nothing else.

3. What Does a Normal First 90 Days Look Like?

Quick Answer: Setup in days 1–14, delivery stabilising by week four, a readable cost per lead by week six, and a scale-or-stop decision at week twelve. Each stage has one thing worth judging and several things that are still meaningless.

Stage-by-stage expectations for the first 90 days of a Malaysian SME Meta ads engagement.

Stage	Should exist by then	Fair to judge	Still meaningless
Days 1–7	Access, verified Pixel and CAPI, audiences rebuilt	Responsiveness, evidence of setup	Leads, CPL, ROAS
Days 8–14	Campaigns live, 3–5 creatives running	Launch date kept, creative variety	Daily cost swings
Weeks 3–4	Learning phase settled, first optimisations	Delivery stability, spend pacing	Week-on-week CPL
Weeks 5–8	Readable CPL, second creative round, lead quality feedback	CPL trend, lead quality	Final verdict on the channel
Weeks 9–12	Efficiency gains, a written scale-or-stop recommendation	Cost per sale, pipeline value	Nothing — this is the review point

Source: ZenWeb client onboarding sample, Malaysian SME Meta accounts, 2024–2026. Licence.

4. What Should Happen in Weeks 3 to 6?

Quick Answer: This is the learning window. Meta's delivery system needs a run of conversions before performance steadies, so weeks three and four are volatile by design. By week six you should have a cost per lead worth discussing.

Meta's own guidance on the learning phase puts the threshold at roughly 50 optimisation events per ad set within seven days, and says results stay unstable until an ad set clears it. Every meaningful edit restarts that clock.

The practical consequence is that a nervous client is the most common cause of a slow account. Asking for a targeting change on day 12, a new offer on day 19 and a budget cut on day 24 keeps the ad set permanently in learning. Agree upfront on a change freeze for the first 21 days, with exceptions only for compliance or stock issues.

• Budget high enough to exit learning. Under-funded ad sets never accumulate the events they need — see the smallest Facebook ads budget that still works.
• Fewer ad sets, not more. Splitting the budget across six ad sets starves all six. Our note on ads that will not deliver covers the symptoms.
• One change window per week. Batch edits instead of tinkering daily.

Key takeaway: Weeks three and four look worse than they are. Protect the learning window with a change freeze and the week-six number becomes worth reading.

5. What Should Happen in Weeks 7 to 12?

Quick Answer: The second half of the quarter is about efficiency and evidence. Cost per lead should be trending down or holding while volume rises, lead quality feedback should be flowing back from your sales team, and week twelve should end with a written recommendation.

By now the agency has run at least two creative rounds and has enough data to say something specific: which audience converts, which offer wins, which placement wastes money. Vague reporting at week ten is a warning sign, not a timing problem.

Push the conversation past cost per lead. A cheap lead that never buys is worse than an expensive one that does, and only your sales team can tell the difference. Feed closed-won data back so the account optimises toward revenue, and read the seven numbers in a Facebook ads report that actually matter before the review call.

Watch ad frequency too. In a market the size of Malaysia, a narrow audience saturates quickly, and rising frequency with a flat cost per result is the signal that a creative refresh is overdue rather than the account failing.

Key takeaway: Weeks seven to twelve should produce specifics — named winning audiences, named losing placements, and a recommendation you can act on. Generic reporting this late is the real red flag.

6. Does Your Starting Point Change the Timeline?

Quick Answer: Heavily. An inherited account with clean tracking reaches a readable cost per lead in about four weeks. A brand new Page and Pixel takes roughly twice as long, because there is no conversion history and no warm audience to retarget.

Median weeks to a readable cost per lead by account starting condition, Malaysian SME Meta accounts.

Starting condition	Relative wait	Median weeks	At target CPL by week 12
Inherited account, tracking healthy		4	78%
New build on an existing Page and Pixel		5	71%
Restart after 3+ months paused		7	58%
Inherited account, tracking broken		8	51%
Brand new Page and Pixel		9	43%

Source: ZenWeb client onboarding sample, Malaysian SME Meta accounts, 2024–2026. Licence.

The single biggest predictor of how fast a new engagement produces a trustworthy number is the state of the tracking you hand over on day one.

7. When Is Slow Progress the Agency's Fault?

Quick Answer: It is the agency's fault when the delay sits inside work they control — build speed, tracking accuracy, creative volume, reporting clarity. It is the account's when the constraint is budget, offer, audience size or a sales team that never follows up.

Separating the two is the whole job of a fair review. Use this split before you conclude anything about the engagement.

• Agency's fault:campaigns still not live after day 14 with no written reason; tracking never verified; the same three creatives running in week ten; reports that show impressions instead of cost per lead; no named recommendation at the quarter.
• The account's constraint:a budget below the level your industry's cost per lead requires; an offer nobody wants; a landing page that loads slowly; leads that sit unanswered for two days.
• Shared:creative assets that arrive late, approvals that take a week, a change freeze nobody agreed to keep.

If several of the first group are true at once, you are not looking at a timing problem — you are looking at the pattern described in seven signs your Facebook ads company is not performing. If the constraint is your own budget or offer, the eight things to check when ads bring no sales is the better starting point, and comparing an agency against an in-house media buyer may reframe the decision entirely.

Key takeaway: Blame the party that controlled the blocker. Most disappointing quarters are a mix, and naming the split honestly is what makes the next quarter fixable.

8. What Actually Delays a Meta Ads Launch?

Quick Answer: Access and creative are the two biggest delays, and both usually sit on the client side. A landing page that is not ready adds the most days of all, because nothing else can be tested until the destination exists.

Frequency and median delay of each launch blocker across Malaysian SME Meta ads onboardings.

Blocker	Share of launches affected	Median days added	Who owns it
Business portfolio or Page access	34%	6	Client
Creative assets not ready	29%	8	Shared
Pixel or Conversions API rebuild	24%	5	Agency
Payment method rejected	17%	3	Client
Landing page or offer not live	15%	11	Client
Agency slow to build	11%	7	Agency

Source: ZenWeb client onboarding sample, Malaysian SME Meta accounts, 2024–2026. Shares exceed 100% because a launch can hit several blockers. Licence.

9. How to Run a Fair 30, 60 and 90-Day Review

Quick Answer: Fix the three review dates before work starts, agree what each one measures, and write the answers down. A review that changes its own success criteria halfway through is how good accounts get cancelled and bad ones get renewed.

Run the same seven steps at each gate. It takes under an hour and removes almost all of the argument.

1. Set the three dates in the contract. Day 30, day 60 and day 90 from launch, not from signature.
2. Agree the measure per gate. Day 30 measures setup and delivery; day 60 measures cost per lead; day 90 measures cost per sale.
3. Check tracking before reading any number. A CPL from a half-firing Pixel is fiction.
4. Compare against your own baseline. Your previous CPL, or your industry band in Malaysian Facebook cost per lead benchmarks.
5. Separate blockers by owner. Use the agency-versus-account split from the section above.
6. Ask for one written recommendation. Scale, hold or stop, with the reasoning in a paragraph.
7. Record the decision and the next gate. One page, both parties, dated.

Write these gates into the agreement itself — the terms to check in a Facebook ads agency contract covers how to phrase a process-based review clause that neither side can argue with later. The ten questions to ask before signing are worth re-asking at day 90.

Key takeaway: Agree what each gate measures before the account spends a ringgit. Reviews only work when the goalposts were planted in advance.

10. Does Waiting Longer Keep Improving Results?

Quick Answer: The steepest improvement happens between month one and month four. After that the curve flattens, so a further quarter of patience buys far less than a creative refresh or a wider audience would.

Cost per lead index and share of accounts at target by month of engagement, with a month nine projection.

Month	CPL index (month 1 = 100)	At or below target CPL	Lead volume index
Month 1	100	18%	100
Month 2	82	41%	126
Month 3	71	63%	148
Month 4	66	70%	159
Month 5	64	73%	166
Month 6	62	75%	171
Month 9*	59	77%	179

* Month 9 projected on the months 1–6 trend. Source: ZenWeb client tracking, Malaysian SME Meta accounts, 2024–2026. Licence.

11. Conclusion: Judge the Sequence, Not the Stopwatch

Quick Answer: Give a new Meta engagement one full quarter, with fixed review gates at day 30, 60 and 90 and an agreed measure at each. That is long enough to be fair to the account and short enough to stop a bad fit quietly renewing.

How long do meta ads take to work is really two questions. The machine needs about six weeks to produce a number you can trust. The relationship needs a quarter to show whether the people running it are any good. Judge them separately, at the dates you set in advance, and you will rarely be the owner who cancelled in week three or the one still waiting in month nine. If you want that calendar built into the engagement from day one, a Meta ads agency should be happy to put the gates in writing before you sign.

12. Frequently Asked Questions

1. How long do Meta ads take to work?

Around six weeks before the cost per lead is trustworthy, and twelve weeks before you can fairly judge the engagement. The first fortnight is setup, weeks three and four are the learning window, and weeks five and six are the first period where the numbers are stable enough to act on.

2. Is it normal to get no leads in the first two weeks?

Yes, and it is common. Campaigns often go live around day 10 to 14, which leaves only a few days of delivery inside the fortnight. What is not normal is having no verified tracking by then, because that means the leads you do get cannot be measured properly.

3. When should I fire a Meta ads agency for slow results?

Not before day 90, unless the failures are process failures rather than performance ones. Campaigns not live after a month, tracking never verified, or the same creatives running untouched for eight weeks are reasons to act early. A cost per lead you dislike at week five is not.

4. Does a bigger budget make Meta ads work faster?

Up to a point. More budget means more optimisation events per week, so an ad set exits the learning phase sooner. Past that, extra spend widens the audience rather than speeding up learning, and can raise cost per lead while the delivery system adjusts.

5. How long should I wait after changing creative?

Give a new creative set seven to ten days before judging it. Significant edits can restart the learning phase, so a change on Monday and a verdict on Wednesday tells you nothing. Batch changes into one weekly window so the account is not permanently relearning.