Before you start
- Access to Meta Ads Manager with reporting permission
- A defined conversion event firing in your Pixel + Conversions API
- Clarity on which attribution window you're reporting against (1-day click vs 7-day click are different numbers)
- A spreadsheet to do the math outside Ads Manager when you want different definitions
The playbook
7 steps
Pick the numerator: which conversion event counts
The numerator is the count of the conversion event you care about. **Pick ONE event** - usually Purchase for e-commerce, Lead/Complete Registration for B2B SaaS, or a custom event for specialized funnels. Mixing multiple events in the numerator produces a meaningless blended rate. If you have multiple key conversions, calculate them separately.
Expected outcome
Single conversion event chosen as the numerator (Purchase / Lead / etc).
Pick the denominator: clicks, landing-page views, or impressions
Three valid choices, each measures something different. **Clicks (CTR-style)**: Conversions ÷ Link Clicks. This is the 'classic' ad conversion rate. **Landing-page views (LPV-style)**: Conversions ÷ Landing Page Views. Filters out users who clicked but bounced before the page rendered (more accurate post-iOS 14). **Impressions (CVR-style)**: Conversions ÷ Impressions. Useful for view-through analysis. **For most cases, use Landing Page Views** as the denominator - it's the cleanest signal in 2026.
# Three valid conversion rate formulas: # Click-based: Conversions ÷ Link Clicks × 100 # Landing-page-based: Conversions ÷ Landing Page Views × 100 # Impression-based: Conversions ÷ Impressions × 100 # Default: Landing Page Views denominator in 2026Expected outcome
Denominator chosen consciously, not by default; documented in your reporting standard.
Set the attribution window explicitly
Meta defaults to **7-day click + 1-day view** in most accounts in 2026 (down from 28-day click pre-iOS 14). The attribution window dramatically changes both numerator and timing. **For internal reporting**, pick one window and stick to it - typically 7-day click. **For cross-team reporting**, footnote which window you used or all the numbers will be incomparable.
Expected outcome
Attribution window locked (e.g. 7-day click) and footnoted in reports.
Pull the numbers from Ads Manager with the right columns
Open Ads Manager → your campaign/ad set/ad. Add columns: **Link Clicks, Landing Page Views, [your conversion event], Cost per result**. Set the date range to a meaningful window (last 7 days minimum, 30 days preferred). Don't calculate conversion rate from day-1 data - it's noise.
Expected outcome
Ads Manager view configured with the right columns over a meaningful date range.
Calculate the rate
Conversion rate = (Conversions ÷ Denominator) × 100. **Example**: 47 Purchases ÷ 1,200 Landing Page Views × 100 = 3.92%. Round to 2 decimals for reporting. Don't quote 3.9234567% - the false precision masks the underlying signal-to-noise ratio.
# Worked example (e-commerce): # Conversions (Purchases): 47 # Denominator (Landing Page Views): 1,200 # Calculation: 47 / 1200 = 0.0392 # As percentage: 3.92% # Reported: "Purchase CVR: 3.9% (7d click, LPV-denominator)"Expected outcome
Conversion rate calculated, rounded to 2 decimals, footnoted with attribution + denominator.
Benchmark against your account's history, not industry averages
Industry-average conversion rates ('2-4% for e-commerce!') are mostly useless because they bundle wildly different products, audiences, and attribution windows. **Benchmark against your own account's rolling 90-day median** instead. Movement against your own baseline is the signal that matters; movement against an industry average is mostly noise.
Expected outcome
Internal 90-day rolling baseline computed; campaign rates measured against your baseline, not industry.
Diagnose conversion-rate movement systematically
Conversion rate moves for 4 reasons. **(1) Traffic quality changed**: audience expansion brought lower-intent visitors. **(2) Creative changed**: new ad attracts different scrollers. **(3) Landing page changed**: a deploy broke or improved the page. **(4) Offer changed**: price, promo, urgency shifted. When CVR moves >15% week over week, walk through these 4 in order and identify which lever changed. Without diagnosis you're guessing.
TipAnnotate creative launches, page deploys, and offer changes in a shared changelog. When CVR moves, the changelog tells you the answer in 30 seconds instead of 3 hours of forensics.Expected outcome
CVR movements diagnosed against the 4-cause framework; root cause documented for each significant movement.
Shuttergen
Move CVR through creative, not spreadsheet math.
Once you know your real CVR, the lever to move it is creative quality and volume. Shuttergen generates the variants Meta needs to find higher-intent buyers and lift conversion rate without lifting spend.
Pitfalls
What goes wrong
Mixing multiple conversion events in the numerator
Bundling Lead + Trial + Purchase into one 'conversion' produces a meaningless blended rate. Calculate each conversion separately.
Inconsistent denominators across reports
Using Link Clicks in one report and Landing Page Views in another creates non-comparable numbers. Pick a standard and footnote it.
Ignoring the attribution window
7-day click conversion rate is a different number than 1-day click. Without the window in the footnote, comparisons across reports are corrupt.
Benchmarking against industry averages
Industry averages bundle wildly different inputs. Your own 90-day rolling baseline is the only useful benchmark.
False precision in reporting
Quoting 3.9234567% implies a precision the data doesn't support. 2 decimals (3.92%) is the maximum honest precision for most accounts.
Limits
When this playbook won't work
- Brand-new campaigns with <100 conversions - any rate calculated is noise-dominated
- Accounts without Conversions API - the denominator (LPV) is underreported by 15-30%
- View-through-heavy campaigns where click-based CVR misses most of the conversion signal
- Cross-platform funnels where the conversion happens days later via email/sales call (single-touch attribution doesn't capture it)
Why most Facebook conversion-rate numbers people quote are wrong
Three corruption sources. (1) Attribution window not specified - 1-day click and 7-day click can differ by 2-3x on the same account. (2) Denominator not specified - clicks vs landing-page views vs impressions produce dramatically different rates. (3) Conversion event ambiguous - 'conversion' meaning Purchase vs Lead vs PageView corrupts comparisons.
The fix is footnoting, not better tools. Every conversion-rate number you report should be one sentence: '3.9% (Purchase, 7-day click, LPV denominator, 30-day window).' This sentence makes the number comparable across teams, reports, and time periods. Without it, you're producing pretty graphs of meaningless numbers.
Comparison to industry averages is mostly performance theater. 'Industry average e-commerce CVR is 2-4%' bundles $20 AOV impulse buys with $4000 considered purchases - they should have different rates. Use your own 90-day rolling median as the benchmark and treat industry averages as orientation, not target.
Internal: best-practices-for-facebook-ads, average-ctr-for-linkedin-ads.
Move CVR through creative, not spreadsheet math. Once you know your real CVR, the lever to move it is creative quality and volume. Shuttergen generates the variants Meta needs to find higher-intent buyers and lift conversion rate without lifting spend.
What changed in 2026 that breaks legacy CVR thinking
Attribution windows shrank. iOS 14 + GA4 + privacy regulations collapsed the default click window from 28 days to 7 days (often 1 day for view-through). This means a CVR you measured in 2021 vs the same CVR measured in 2026 is structurally different - apples to a different fruit.
Conversions API recovered some of the missing signal. CAPI adds 15-30% to measured conversions in most accounts. Without it, your CVR denominator is underreported and your real conversion rate is higher than Ads Manager shows. With it, you have a cleaner read.
View-through attribution matters more. With shorter click windows, view-through conversions represent a larger share of the picture. Pure click-based CVR misses this. For most accounts, the 7-day click + 1-day view default is the right balance; for high-AOV B2B, longer windows still earn their weight where the platform allows.
Internal: hyros, hyros-vs-triple-whale.
FAQ
Frequently asked
How do I calculate Facebook ad conversion rate?
What's a good Facebook ad conversion rate?
Should I use clicks or landing-page views as the denominator?
What attribution window should I use for Facebook ad conversion rate?
Why is my Facebook ad conversion rate dropping?
Does Conversions API change conversion rate calculations?
How many conversions do I need before the rate is reliable?
Related
Keep reading
Resource
Best practices for facebook ads
Broader best-practices guide.
Resource
How long does it take for facebook ads to optimize
Optimization timing guide.
Resource
Hyros
Server-side attribution deep dive.
Research
Triplewhale Vs Hyros
Attribution platform comparison.
Research
Triplewhale Deep Dive
Multi-touch attribution research.
Move CVR through creative, not spreadsheet math.
Once you know your real CVR, the lever to move it is creative quality and volume. Shuttergen generates the variants Meta needs to find higher-intent buyers and lift conversion rate without lifting spend.