ROAS = Revenue ÷ Ad Spend. The number every platform dashboard puts front and center - and the number you should trust least without context. In MarketingSync, we show platform ROAS alongside blended MER precisely because platform ROAS is almost always inflated. Google Ads and Meta both claim credit for the same conversion. A campaign showing 8x ROAS on Google while your business MER is 2x is a red flag, not a success. E-commerce healthy range: 3–5x. Lead gen with high LTV can profitably run at 1.5–2x.

MER = Total Business Revenue ÷ Total Marketing Spend across all channels. This is the metric MarketingSync was fundamentally built around. Unlike platform-reported ROAS, MER doesn't care what Google or Meta says - it uses your actual revenue. When we analyzed dozens of ad accounts, the average gap between platform-reported ROAS and actual MER was 2.3x. MER below 1.0 means you're losing money. MER above 3.0 with consistent volume is strong. Track MER weekly - it's your most honest marketing KPI.

CPA = Total Ad Spend ÷ Conversions. The workhorse metric for performance campaigns. What makes CPA tricky is that 'conversion' means something different on every platform. In MarketingSync, we normalize conversion definitions across Google, Meta, TikTok, and Taboola so you can compare apples to apples. A CPA of $50 on Google and $40 on Meta sounds like Meta wins - until you realize Meta is counting view-through conversions that would have happened anyway. Always verify what action is being counted before comparing CPA across platforms.

LTV = Total revenue from a customer over the entire relationship. The single most important number for deciding how much you can afford to spend on acquisition. If your LTV is $600 and CAC is $120, you have a 5:1 ratio - healthy growth fuel. LTV completely changes the math on CPA targets. We've seen accounts running 'unprofitable' Google campaigns at $200 CPA that were actually printing money because 12-month LTV was $1,200. MarketingSync supports LTV-adjusted ROAS calculations when you connect revenue data.

CAC = All Marketing + Sales Costs ÷ New Customers Acquired. Broader than CPA - CAC includes platform spend, agency fees, creative production, tools like MarketingSync, and any sales team costs. The LTV:CAC ratio is the fundamental health metric for any performance marketing program. Under 2:1 is trouble. 3:1 is the baseline for sustainable growth. Above 5:1 often means you're under-investing and leaving market share on the table.

Attribution = assigning conversion credit to the channels and touchpoints that influenced it. The root cause of almost every misleading ROAS report. Each platform's attribution model is designed to maximize how much credit that platform receives. Google claims the search click. Meta claims the retargeting impression. Taboola claims the awareness touchpoint. All for the same conversion. In accounts we've analyzed through MarketingSync, attribution overlap routinely inflates total reported conversions by 200–300% compared to actual orders. The fix: trust your MER, not your platform dashboards.

VTA = credit assigned to an ad that was seen but never clicked, when a conversion happens later. Meta's default 1-day view window and TikTok's 7-day view window mean that if someone saw your ad on Tuesday and bought on Wednesday after Googling your brand name, Meta claims that conversion. This is the single largest source of attribution inflation on social platforms. In MarketingSync alerts, we flag accounts where view-through conversions exceed 40% of reported Meta conversions - a threshold that typically signals inflated reporting rather than real influence.

CAPI = server-side conversion data sent directly from your server to Meta, bypassing browser tracking. Post iOS 14, the Meta pixel lost visibility into roughly 30–40% of iOS conversions. CAPI recovers most of that lost signal. Best practice: run pixel AND CAPI in parallel with deduplication enabled. Without CAPI, Meta's algorithm is optimizing on incomplete data - which explains why some campaigns that 'worked before iOS 14' started underperforming. MarketingSync detects when Meta event match quality scores are low and alerts you to implement or fix CAPI.

First-party data = data you own directly: customer emails, purchase history, CRM records, on-site behavior. With third-party cookies gone and iOS tracking restricted, first-party data is now the primary moat in digital advertising. Advertisers with strong first-party data can build better seed audiences for lookalikes, fuel CAPI with verified conversion signals, and run email retargeting that doesn't depend on platform pixels. MarketingSync connects to your CRM data to enrich campaign-level LTV calculations and identify which campaigns are actually acquiring your best customers.

PMax = Google's fully automated campaign type that serves ads across Search, Display, YouTube, Gmail, Discover, and Maps using a single budget and creative set. Google's black box. PMax campaigns frequently cannibalize branded search - meaning you're paying for clicks you would have gotten for free. We've seen accounts where removing PMax budget exclusions for branded terms cut cost by 15% with zero impact on conversions. MarketingSync tracks PMax efficiency separately and fires alerts when branded impression share drops below thresholds - a common sign of PMax cannibalization.

Smart Bidding = Google's ML-based automated bid strategies: Target CPA, Target ROAS, Maximize Conversions, Maximize Conversion Value. The algorithm needs data to work - specifically 30+ conversions per month per campaign as a baseline, 50+ to perform well. Below that threshold, Smart Bidding often overfits to noise. One pattern we see repeatedly: a campaign with 8 conversions/month on Target CPA will spend erratically, then dry up as the algorithm 'learns' false patterns. In MarketingSync, we flag campaigns on Smart Bidding with insufficient conversion volume and recommend switching to manual CPC or enhanced CPC until thresholds are reached.

Quality Score = Google's 1–10 rating of keyword + ad + landing page relevance. It directly multiplies into Ad Rank, which determines both position and effective CPC. The math: at equal bids, a QS 8 ad beats a QS 4 ad in auction AND pays less per click. We've run tests where improving Quality Score from 4 to 7 on the same keyword reduced CPC by 38% with no bid changes. Components: Expected CTR (50% weight), Ad Relevance (25%), Landing Page Experience (25%). Most Quality Score problems are landing page problems - not ad problems.

Ad Rank = the score Google uses to determine if and where your ad shows. Formula: Max Bid × Quality Score × Expected Extension Impact × Auction-time signals. What this means in practice: you can outrank a competitor bidding 2x more than you if your Quality Score is significantly higher. Ad Rank is why 'just raise the bid' is usually the wrong answer to losing impression share - and why Quality Score optimization is worth far more in the long run. MarketingSync tracks impression share lost to rank vs. lost to budget separately.

Impression Share = Impressions you received ÷ Total impressions you were eligible for. The most underused diagnostic metric in Google Ads. Lost IS (Budget) = your daily budget ran out before auctions ended. Lost IS (Rank) = your Ad Rank wasn't high enough to enter auctions. These require completely different fixes. Branded campaigns should target 90%+ IS. If you're below 70% on brand, you're gifting clicks to competitors. MarketingSync alerts fire when branded IS drops below configurable thresholds - one of the first signs of a competitor bidding on your brand name.

CTR = Clicks ÷ Impressions × 100. On Google Search, average CTR is 2–5% - but position 1 typically sees 8–12% while position 4 sees under 2%. On Meta and TikTok, CTR is more about creative quality than position. A CTR below 0.5% on Meta usually means your creative isn't stopping the scroll. A sudden CTR drop on Google Search is often the first signal of a competitor entering your auction or a quality score decline - MarketingSync fires alerts when 7-day CTR drops more than 20% versus the prior period.

CPC = Total Spend ÷ Total Clicks. Ranges from $0.10 (gaming display) to $80+ (legal/finance search). A lower CPC is not always better - it depends entirely on conversion rate and what happens after the click. The useful signal in CPC is the trend: a rising CPC with flat conversion rate means rising CPA. In MarketingSync, CPC is tracked per platform and per campaign so you can identify when a specific campaign's auction competitiveness is increasing - often a signal that a competitor is scaling.

CPM = Cost per 1,000 impressions. The fundamental unit of reach on Meta, TikTok, Taboola, and programmatic. Rising CPM is the first sign of audience saturation - you've shown your ad to everyone in the audience who is going to respond, and now you're paying premium prices for diminishing returns. On Taboola specifically, CPM spikes often precede performance drops by 3–5 days. MarketingSync tracks CPM trends per platform and audience, with alerts configured to fire when 7-day CPM rises more than 25% without a corresponding improvement in conversion rate.

CPL = Total Spend ÷ Leads Generated. The primary efficiency metric for lead gen campaigns. The trap: optimizing for low CPL without tracking lead quality almost always ends in disaster. We've seen accounts drive CPL down 60% by targeting broader audiences - while the sales team's close rate dropped 80% because the leads were unqualified. In MarketingSync, CPL is most useful when combined with a downstream conversion rate metric (lead to SQL, SQL to closed deal) to calculate true CAC.

CVR = Conversions ÷ Clicks × 100. Arguably the most impactful lever in paid media that most teams underinvest in. A 1% improvement in CVR has the same effect on CPA as a 50% reduction in CPC - but improving landing page conversion is usually faster than negotiating auction prices down. E-commerce: 1–4% is typical, 6%+ is strong. B2B lead gen: 5–15% on well-qualified landing pages. MarketingSync surfaces CVR per campaign and flags when CVR drops significantly - often caused by landing page changes, offer changes, or traffic quality degradation.

Retargeting = showing ads to users who previously visited your site, viewed a product, or engaged with your brand. The highest-intent, highest-converting audience in most accounts - typically 3–5x better CVR than cold prospecting. The mistake most teams make: treating all retargeting audiences the same. Cart abandoners (72-hour window) need urgency. Product viewers (7-day window) need social proof. Blog readers need education. Separate them. In MarketingSync, cross-platform retargeting overlap analysis shows you when the same user is being retargeted on Google, Meta, and TikTok simultaneously - which is usually wasteful.

Lookalike audiences = Meta, TikTok, or Google using ML to find users who resemble your best customers. Quality of seed data is everything. A lookalike built from 200 email addresses performs far worse than one built from 5,000 verified purchasers. 1% lookalikes on Meta are most similar (smallest, most targeted). 5–10% are broader but offer more scale. Post iOS 14, lookalike quality has degraded because the underlying pixel data is less complete - which is why CAPI implementation and first-party data quality have become so important for maintaining lookalike performance.

The funnel = the journey from first impression to paying customer. TOFU (Top of Funnel): awareness via YouTube, TikTok, Taboola - optimize for reach and CPM. MOFU (Middle): consideration via Meta retargeting, Google Display - optimize for engagement and CPL. BOFU (Bottom): conversion via Google Search, branded campaigns - optimize for CPA and ROAS. The common mistake: measuring all three stages with the same metrics. MarketingSync separates funnel stage performance so you can see that TikTok awareness campaigns (TOFU) are driving better downstream BOFU conversion rates even when they show poor direct ROAS.