7 Meta Ads Metrics Founders Should Track Every Week

The Meta Ads metrics that matter for a founder are not the 40 columns Ads Manager shows you. They are seven numbers, each tied to one decision. This post is that shortlist: what to track every week, the threshold that turns it from a number into a signal, and the exact action it should trigger. The goal is a 10-minute Monday read, not a spreadsheet you keep open all day.

Why founders need a shortlist, not a dashboard

A founder running Meta in-house has a job that isn't "media buyer." You don't have time to read every column, and most of them don't change a decision anyway. Ads Manager is built for analysts who live in it. You need the opposite: the fewest metrics that tell you whether to leave the account alone, refresh creative, or stop the bleeding.

The 2026 changes made the shortlist harder to read, not easier. Meta narrowed what counts as a click, removed two attribution windows, and made creative the primary signal its delivery system uses. If you're tracking the same metrics the same way you did in 2024, two of your seven numbers now mean something different. The list below is built for the current rules.

Read this first: which conversion number you're even looking at

Before any of the seven metrics, confirm you're reading conversions on the 2026 attribution stack. Otherwise ROAS and CPA — two of the seven — are wrong at the source.

Two changes are responsible:

• On January 12, 2026, Meta removed the 7-day-view and 28-day-view attribution windows from the Ads Insights API, confirmed by PPC Land's coverage of the developer notice.
• In March 2026, Meta narrowed click-through to link clicks only and moved likes, comments, shares, saves, and 5-second-plus video views into a separate "engage-through" bucket. This is in Meta's own announcement on the click-attribution update and Search Engine Land's writeup of the rollout.

The practical result: the default attribution setting is now 7-day click + 1-day engage-through + 1-day view. Engage-through is its own column, not folded into click-through. If your weekly view still reads click-through alone against a pre-March baseline, your conversions look like they fell off a cliff when they mostly just moved columns. The full diagnostic for that is in why your weekly Meta Ads report shows fewer clicks; the backstory on the metric change itself is in the engage-through attribution primer. Set the stack once, then read the seven.

The 7 Meta Ads metrics that matter, at a glance

| # | Metric | What it answers | Weekly watch threshold | |---|---|---|---| | 1 | ROAS (trend) | Are we profitable, and which way is it moving? | Below your breakeven ROAS two weeks running | | 2 | CPA / CAC | Does each new customer cost less than they're worth? | CPA above your contribution-margin ceiling | | 3 | CPM | Is it getting more expensive to reach people? | Week-over-week spike of 15–20%+ | | 4 | Link CTR | Is the creative still landing? | Decline of 10–15% week over week | | 5 | Frequency | Are we burning the same people out? | Above ~2.5 on cold audiences | | 6 | Conversion rate | Is the problem the ad or the page? | CVR falling while CTR holds | | 7 | Spend pacing | Is the money flowing to winners? | Budget on sub-target ads creeping up |

For category context, Triple Whale's 2025 benchmark across roughly 35,000 ecommerce brands (updated April 7, 2026) gives median anchors: ROAS 1.86, CPA $38.19, CPM $14.19, CTR 2.19%, and conversion rate around 1.6%. Treat these as a rough sanity check, not a target — the vertical spread is wide, and your breakeven is the number that actually matters.

1. ROAS — read the trend, not the absolute

Return on ad spend is the headline, but the absolute number lies to founders constantly. A 1.86 median tells you almost nothing about your account, because the only ROAS that matters is your breakeven: 1 divided by your contribution margin. A brand at 30% margin breaks even at roughly 3.3x; a brand at 60% margin breaks even at 1.7x. Same 2.0 ROAS, opposite verdict.

Track two things weekly: ROAS versus your breakeven, and the week-over-week direction.

• Act today — ROAS below breakeven two weeks running. You're buying customers at a loss and it isn't a blip.
• This week — ROAS above breakeven but trending down three weeks straight. Something is decaying; find it before it crosses the line.
• Monitor — single-week dip inside normal variance. Don't touch it.

2. CPA / CAC — against your ceiling, not a benchmark

Cost per acquisition only means something next to your CAC ceiling: the contribution margin a new customer delivers (factor in repeat rate if you know it). The median CPA of $38.19 is irrelevant if your ceiling is $25.

The decision is binary. CPA under your ceiling: acquisition is profitable, scale-eligible. CPA over your ceiling: every new customer loses money, regardless of how good the ROAS chart looks on blended numbers. When CPA crosses the ceiling and stays there, that's Act today — pause the offenders and find where the cost is coming from with a 30-minute account audit.

3. CPM — the cost of being in the auction

CPM is what it costs to reach 1,000 people, and it's the metric founders most often misread as a creative problem. Meta's 2025 median sat at $14.19 and the direction has been up. A rising CPM alone isn't an emergency — it's the weather. A sudden week-over-week spike is the signal.

When CPM jumps 15–20% or more in a week, check two things before you blame anything: frequency (metric 5) and audience size. A CPM spike usually means the auction got more expensive or your audience got narrow and saturated — not that the ad is bad. Cutting creative in response to a CPM spike is a common and expensive mistake.

4. Link CTR — the earliest warning you get

Link click-through rate is the leading indicator for the whole account. It moves before ROAS does, because falling resonance shows up in clicks days before it shows up in revenue. The 2025 median was 2.19%, but the level matters less than the slope.

A week-over-week CTR decline of 10–15% is your early warning. On its own, queue a creative refresh for This week. Paired with rising frequency, it's the textbook fatigue pattern — covered in depth in what creative fatigue actually is. Watch CTR and you rarely get surprised by a ROAS drop.

5. Frequency — how fast you're burning the audience

Frequency is average impressions per person over your window. It's the fatigue metric, and it matters more in 2026 than it used to. Triple Whale's creative fatigue framework (April 2026) flags frequency above roughly 2.5 on cold, prospecting audiences as the point where performance typically starts to decline. Many media buyers treat anything climbing past 3 as a hard refresh trigger.

Why faster now: with Meta's Andromeda retrieval engine, creative became the primary signal the system uses to match ads to users. Search Engine Land's breakdown of Andromeda (January 2026) notes that as a result, "creative fatigue accelerated." The fatigue cycle is shorter than it was two years ago — the Andromeda creative fatigue explainer covers the thresholds in detail.

• Act today — frequency above 2.5 and CTR falling. That's confirmed fatigue. Cut the tired creative, ship the next.
• This week — frequency climbing toward 2.5 with CTR still flat. Queue replacements now so you're not scrambling.
• Monitor — frequency under 2 and stable. Fine.

6. Conversion rate — is it the ad or the page?

Conversion rate (clicks to purchase) is the metric that isolates where a problem lives. The 2025 ecommerce median was around 1.6%. The diagnostic power is in pairing it with CTR:

• CTR holding, CVR falling → the ad is doing its job and the landing page or offer is the problem. Don't touch creative. Fix the page, price, or promo.
• CTR falling, CVR holding → the opposite. It's a creative problem, not a site problem.

Most founders react to a soft week by changing ads. Half the time the ad was fine and the landing page quietly broke, slowed down, or went out of stock. CVR tells you which half you're in.

7. Spend pacing — where the money is actually going

The last metric isn't a rate, it's a flow check: is your budget reaching the ads that work? Every account leaks spend into ad sets that quietly underperform while the winners get starved. Weekly, look at how much of total spend is sitting in ads below your target ROAS and whether that share is creeping up.

When sub-target ads are eating a growing slice of the budget, that's the cheapest fix on this list: cap or kill the losers, let the winners breathe. This is where most "wasted spend" hides, and it's a core part of any account audit.

A founder's Monday read, worked through

Take a hypothetical $25K/month DTC brand at 40% contribution margin (breakeven ROAS 2.5). The Monday numbers:

• ROAS 2.3, down from 2.7 and 2.9 the prior two weeks. Below breakeven, trending down. Flag.
• CPA $46 against a $40 ceiling. Just over. Flag.
• CPM $16.80, up 4% week over week. Normal drift, not a spike. Fine.
• Link CTR 1.7%, down from 2.1% — a 19% drop. Flag.
• Frequency 2.9 on the main prospecting audience, up from 2.4. Over threshold. Flag.
• CVR 1.6%, flat. The page is fine.

The pattern reads itself: CPM is stable and CVR is flat, so this isn't an auction problem or a landing-page problem. CTR is down sharply and frequency crossed 2.5 — that's creative fatigue, and it's dragging ROAS below breakeven and CPA over the ceiling. The action is Act today: cut the fatigued creative on the prospecting audience and ship the next batch. Not a budget cut. Not a landing-page rebuild. One move, pointed at the actual cause.

That's the whole point of the shortlist. Six of the seven metrics existed only to rule out the wrong fixes, so the seventh decision is obvious.

Common mistakes reading weekly Meta Ads metrics

• Judging ROAS against a benchmark instead of your breakeven. A "good" ROAS that's below your margin line still loses money. The median is not your target.
• Reading conversions on the old attribution stack. If click-through still excludes engage-through against a pre-March baseline, ROAS and CPA are understated at the source. Set the 7-day-click + 1-day-engage-through + 1-day-view stack first.
• Cutting creative on a CPM spike. CPM is an auction-and-audience signal. Check frequency and audience size before touching the ads.
• Changing ads when CVR is the problem. If CTR is holding and conversion rate dropped, the page or offer broke, not the creative. Diagnose before you act.
• Tracking 20 metrics so none of them trigger anything. A metric you don't have a threshold and an action for is decoration. Seven with rules beats forty without.

FAQ

What Meta Ads metrics should a founder track weekly? Seven: ROAS (as a trend against your breakeven), CPA versus your CAC ceiling, CPM, link CTR, frequency, conversion rate, and spend pacing. Each one should have a threshold and a pre-decided action, so the weekly read produces a decision instead of more questions.

What's a good ROAS for Meta Ads in 2026? There's no universal number. Triple Whale's 2025 benchmark put the ecommerce median at 1.86, but the only ROAS that matters is your breakeven — 1 divided by your contribution margin. A 2.0 ROAS is profitable at 60% margin and a loss at 30%.

What frequency is too high on Meta Ads? On cold prospecting audiences, frequency above roughly 2.5 is where performance typically starts to decline, per Triple Whale's 2026 creative fatigue framework. Watch frequency together with CTR — rising frequency plus falling CTR is the confirmed fatigue signal that warrants cutting creative.

Why do my Meta Ads conversions look lower in 2026? Two attribution changes. Meta removed the 7-day-view and 28-day-view windows in January 2026, and in March it moved engagement-based conversions into a separate engage-through column. If you compare current click-through to a pre-March baseline, conversions look down even when they aren't. Read the fewer-clicks diagnostic to confirm artifact versus real decline.

How is CTR different from CVR, and why track both? CTR (click-through rate) measures whether the ad earns the click; CVR (conversion rate) measures whether the click turns into a purchase. Tracking both lets you locate a problem: falling CTR points to the creative, while falling CVR with steady CTR points to the landing page or offer.

How a pre-diagnosed action list handles this

The seven metrics are simple to name and tedious to run every Monday — pull each one, rebuild the baseline on the right attribution stack, compare against the threshold, and cross-reference CTR with frequency and CVR to find the actual cause. That's analysis. It's exactly the work most founders don't have time for, which is why the account drifts.

Good Morning does that read for you. Every Monday it scores the same seven metrics, flags which thresholds were crossed, and outputs a single urgency-tiered action list — Act today, This week, Monitor — with the cause already isolated. Action items, not analysis. Zero columns to interpret. The Account Health Score compresses all seven into one 0–100 number for the fast executive read, and the creative fatigue tool watches frequency and CTR together so the fatigue call is made before ROAS drops.

If you're a founder running Meta in-house, the in-house team and DTC brand flavors cover the same seven metrics from each operating-model angle. Comparing tools? Good Morning vs Motion and Good Morning vs Triple Whale lay out how an action-list reporter differs from a creative analytics platform and an ecommerce intelligence suite.

The one-line takeaway

Track seven Meta Ads metrics, each with a threshold and a pre-decided action — then the weekly read takes ten minutes and ends in a decision, not a dashboard.

Want those seven scored for you every Monday, with the action already decided? See how Good Morning works →. Pricing is flat at $50/month — see the pricing page, and the Meta Ads glossary has plain-language definitions for any term above.