70 to 82% of Google Ads accounts still pilot on gross revenue ROAS: 2.4× apparently profitable, sometimes 0.8× in real margin — the standard gap observed between gross ROAS and margin ROAS on our panel. Understanding CPC, CPA, ROAS, ROI, LTV and knowing how to connect them is the only way to avoid this structural trap.
Open any Google Ads dashboard and you'll land on 35 columns. Impressions, clicks, CTR, CPC, CPM, CPA, ROAS, ROI, conversions, conversion value, conversion rate, impression share… Enough to lose any beginner (and quite a few experts).
Good news: 95% of PPC account management happens with 5 metrics. The others are either intermediates (CTR, conversion rate) or noise (CPM in Search, impressions without context…).
This guide lays the groundwork: precise definitions, formulas, order of magnitude by sector, and most importantly how these 5 metrics connect to drive strategic decisions.
What is CPC (Cost Per Click) and what factors push it up?
CPC is the amount you pay for each click on your ad. It's the entry metric of the funnel: everything that comes after depends on it.
CPC = Total cost ÷ Number of clicks
What pushes your CPC up:
- Competition on the keyword (e.g., "auto insurance" = $8-15 CPC in 2026)
- A low Quality Score (Google penalizes you — going from 5/10 to 8/10 cuts CPC by -24 to -36% depending on vertical)
- A too-broad match type that blows up bids
- Geographic targeting on high-demand zones (major cities)
Order of magnitude by sector (US, Search, 2026 — source: our aggregated data over a 90-day rolling period, stratified by vertical):
A low CPC isn't a good CPC if there's no conversion behind it. A $0.50 CPC with 0% conversion costs more than a $5 CPC that converts at 10%. CPC is an input, not an objective.
What is CPA (Cost Per Acquisition) and how do you lower it?
CPA is the cost to obtain one conversion (purchase, lead, signup — whatever you defined). It's the metric that 90% of advertisers track.
CPA = Total cost ÷ Number of conversions
or: CPA = CPC ÷ Conversion rate
This second formula is important: it shows you can lower your CPA in two ways:
- By lowering your CPC (better Quality Score, negatives, Smart Bidding) — see our guide The complete guide to reducing your CPA.
- By increasing your conversion rate (better landing, better message-match, reducing checkout friction).
The second is 10× more powerful than the first: doubling your conversion rate (from 2% to 4%) halves your CPA without touching bids.
With a $2 CPC and a 2% conversion rate, your CPA = $100. If you lower the CPC to $1.50 (-25%), your CPA drops to $75 (-25%). But if you raise the conversion rate to 3% without touching the CPC, your CPA falls to $66 (-34%). The conversion lever > the bid lever.
What is ROAS and what target value should you aim for?
ROAS measures how much each dollar invested in advertising earns you. More precise than CPA as soon as conversion value varies (which is the case in e-commerce).
ROAS = Conversion value ÷ Advertising cost
A ROAS of 3 means: for $1 invested, $3 of revenue generated. Often expressed as a multiplier (3x) or as a % (300%).
Target values:
You're losing money. Stop and diagnose.
Precarious balance. You cover overhead with little margin.
Comfortable. Healthy margin, reinvestment possible.
Excellent. Raise the budget as long as ROAS holds.
Watch out: break-even ROAS depends on your margin. With 30% gross margin, you must have a minimum ROAS of 3.33x (1 / 0.30) to cover your media cost. Below this threshold, you're paying to sell at a loss.
ROI · Return On Investment
ROI is the bigger sibling of ROAS: it includes all costs, not just media. For a true advertising ROI, add to the denominator:
- Media cost (= what ROAS measures)
- Creative production cost (photographer, graphic designer, videographer)
- Tool cost (SteerAds, Semrush, analytics…)
- Human cost (salary of the person running the campaigns)
ROI = (Revenue generated − Total cost) ÷ Total cost × 100%
A ROAS of 4x can yield an ROI of 150% or 20% depending on your cost structure. Internally, pilot on ROAS. In the boardroom, talk ROI.
LTV · Lifetime Value (the hidden metric that changes everything)
LTV is the total value a customer will bring you over the entire duration of their relationship with you. It's the most underestimated strategic metric by advertisers.
LTV = Average cart × Number of purchases per customer × Gross margin
The effect on your target CPA:
Imagine a SaaS at $50/month, 12-month average retention, 80% gross margin. LTV = 50 × 12 × 0.80 = $480.
With a $480 LTV, you can spend up to $200 in CPA and stay largely profitable over 12 months. Yet if you only look at CPA vs the first payment ($50), you'll cut your campaigns and kill growth.
SaaS example: Day 0 ROAS vs Month 12 ROAS
Campaign spends $10,000, acquires 100 customers at $100 CPA. First payment of $50 each = $5,000 immediate revenue.
- Day 0 ROAS = 5,000 / 10,000 = 0.5x (apparent disaster)
- Month 12 ROAS = (100 × 480) / 10,000 = 4.8x (actually excellent)
How are the 5 metrics connected to each other?
Here's the mental model to remember. The 5 metrics form a chain that starts at the click and ends at profitability:
CPC × (1 / Conversion rate) = CPA
CPA < LTV × Margin → Profitable
Cart value / CPA = ROAS
(Revenue − Total cost) / Total cost = ROI
Golden rule #1: never optimize CPC alone. Always look at it in ratio with conversion rate.
Golden rule #2: always compare CPA to LTV, not to first-sale price. Otherwise you kill growth.
The minimum dashboard to monitor
For healthy management, 7 KPIs are enough. Add them to your favorites in Google Ads or your monitoring tool — and configure an alert on each (see our alerts setup doc).
Need to see these 7 KPIs quickly calculated on your own account? Our free audit outputs this dashboard in 3 minutes after OAuth connection, plus preconfigured alerts if a KPI drifts from its target.
Margin ROAS vs gross ROAS: why 70 to 82% of accounts get this wrong
This is trap #1 of Google Ads dashboards in 2026. The default ROAS shown is a gross-revenue ROAS — it divides conversion value (= revenue generated) by ad spend without accounting for your real margin. Yet 70 to 82% of audited accounts make their budget calls based on this gross ROAS thinking it reflects profitability, when it's nowhere close.
Full worked example: a fashion e-commerce sells an average cart at $100. Of that $100, COGS (cost of goods sold) is $35, so gross margin is $65. Logistics (picking, shipping, returns) costs an extra $18 per order. Stripe payment fees: 1.4% + $0.25 = $1.65. Marketing cost (the Google Ads click that drove the sale): $20. Average customer service: $6 per order.
- Gross-revenue ROAS shown by Google Ads: 5x ($100 revenue / $20 spend) — looks profitable.
- Contribution-margin ROAS: ($100 − $35 − $18 − $1.65 − $6) / $20 = $39.35 / $20 = 1.97x.
- Final net-margin ROAS (after pro-rata share of fixed costs): $4 / $20 = 0.2x — outright loss.
Total flip: the same "ROAS 5x" campaign in the interface is in fact unprofitable at the operational level. Across 1,000 orders, the advertiser spent $20,000 to generate $4,000 of net margin — a $16,000 hole nobody sees inside Google Ads.
How to avoid the trap: across all your dashboards, replace gross-revenue ROAS with a contribution-margin ROAS, computed via Looker Studio blending or a Sheets tab connected to BigQuery. Formula: (Revenue × contribution margin %) / spend, where contribution margin includes COGS + logistics + payment + proportional CS — but not fixed costs (rent, non-marketing salaries). It's the metric that speaks to the CFO, not to the traffic manager.
Concrete rollout step: build a single Looker Studio data source that joins the Google Ads export (spend, conversion value) with a sheet holding a contribution-margin multiplier per product category. That multiplier, refreshed quarterly by finance or controlling, varies: 0.42 for fashion, 0.18 for electronics, 0.68 for SaaS. The dashboard then displays both gross ROAS and margin ROAS — the latter becoming the single source of truth for budget calls. Teams that rolled out this dual-view report -22 to -34% savings on inefficient campaigns within 60 days.
a gross ROAS of 4x can hide a margin ROAS between 1.2x and 2.8x depending on the vertical. On the accounts we audit, the median gap observed between gross ROAS and margin ROAS is 2.4x vs 1.1x. Before any budget decision, compute margin ROAS — it's the only metric that tells you whether the marginal spend is profitable.
CAC vs CPA: the nuance that changes everything
CAC and CPA are often used interchangeably — wrongly. The difference is structural and shapes the read-out of long-run acquisition profitability.
- CPA (Cost Per Acquisition) = media cost to obtain a conversion. Numerator: Google Ads spend (or Meta, or TikTok). Denominator: tracked conversions.
- CAC (Customer Acquisition Cost) = total cost to acquire a customer. Numerator: media spend + acquisition salaries + tools + agency + creative production fees. Denominator: net new customers (not conversions, unique customers).
Worked example on a US B2B SaaS: $12,000 of Google Ads spend, $20,000 of Meta spend, $4,800 of SDR salary (pro-rated to the acquisition share), $1,500 of tools (HubSpot, Semrush, Looker Studio Pro), $3,000 of agency. Total: $41,300. 95 new customers acquired in the window. CAC = 41,300 / 95 = $435 per customer. Meanwhile, the Google Ads CPA reads $12,000 / 60 conversions = $200. CAC is 2.2× the CPA — a structural gap most advertisers ignore in their reporting.
When to use CPA vs CAC: pilot on CPA for daily tactical optimizations (bid adjustments, negatives, budgets). Pilot on CAC for annual strategic decisions (SDR hiring, agency choice, tooling investment). Mixing the two leads to bad calls — a stable CPA does not guarantee a stable CAC if you grow operating costs in parallel.
Target ROAS and Target CPA: how to set a business-realistic target value
Once the definitions are nailed down, the practical question remains: what target value should you aim for on ROAS or CPA? Too high, you choke volume; too low, you lose money. The formula that works in 80% of cases starts from your gross margin and your net-margin objective.
Target ROAS formula: Target ROAS = 1 / (gross margin % − net margin objective %). Fashion e-commerce example with 35% gross margin and an 8% net-margin objective: Target ROAS = 1 / (0.35 − 0.08) = 1 / 0.27 = 3.7x minimum. Below 3.7x gross ROAS, the campaign does not deliver the targeted net margin.
Target CPA formula: Target CPA = LTV × gross margin × (1 − share allocated to other costs). B2B SaaS example with 24-month LTV × ARPU $200 = $4,800, 70% gross margin: cumulative contribution margin = $3,360. If you allocate 25% to acquisition (the rest = R&D, support, growth content), Target CPA = 3,360 × 0.25 = $840. That's your ceiling — beyond it, the cohort is no longer profitable at the reference LTV.
Healthy range by vertical (to be indexed on your real margin, these numbers are reference points):
Classic mistake to avoid: setting a Target CPA 30 to 40% below the historical observed CPA, hoping Smart Bidding will "figure it out." Documented result across 2,000+ accounts: -55 to -72% of volume in 14 days, the algo cuts delivery rather than going under the threshold. Best practice: start Target CPA at +10 to +15% above the historical CPA, then step down by 10% every 14 days as long as volume holds.
MER (Marketing Efficiency Ratio): the KPI rising in 2026
The MER (Marketing Efficiency Ratio) is the KPI that rose the most in CMO and head-of-growth circles in 2025-2026. The logic: aggregate every marketing expense (Google Ads, Meta, TikTok, LinkedIn, influence, SEO tools, agency, salaries) and divide by total revenue. Simple formula: MER = Total revenue / Total marketing spend.
Why MER is rising in 2026: it neutralizes attribution biases. Google Ads ROAS over-attributes brand (last-click), Meta ROAS under-attributes cross-device conversions, PMax over-attributes by +32% vs the +11% real incrementality observed. When you stack the ROAS from every channel, you often hit an absurd 8x to 12x total — while your P&L says real marketing efficiency is 2.5x to 4x. MER, on the other hand, is agnostic: it doesn't ask "who made the sale" but "how much each global marketing dollar returns."
Healthy thresholds observed across 2,000+ accounts:
- Mainstream B2C e-commerce: MER ≥ 3 (typically 3.5 to 5 on the best accounts).
- B2B SaaS: MER ≥ 5 on a 12-month rolling window (higher product margins).
- Omnichannel retail with physical presence: MER ≥ 2.5 if store traffic is measured.
- DTC luxury: MER ≥ 4 (higher cart compensates a tighter reach).
Use case: arbitrating between Google Ads and Meta Ads at a constant total budget. If the global MER drops while Google Ads ROAS climbs, you're probably looking at cannibalization — Google captures sales that Meta would have driven in the upper funnel. MER lets you settle that debate without sliding into a last-click ROAS war.
Recurring trap: forgetting to include human and tool costs in the denominator. A "media-only" MER misleads as much as a gross-revenue ROAS. Agency, freelance and in-house staff costs can represent 15-25% of the real media budget — ignoring them artificially inflates MER by 20-30%.
Recap: the 5 metrics that pilot a profitable account in 2026
If you only remember one thing: the gross ROAS shown by Google Ads lies. Not maliciously, by construction. It divides revenue by media spend without integrating margin, other variable costs, or the gap between Google attribution and real incrementality. To run an account healthily in 2026, switch to modern KPIs:
- Contribution-margin ROAS rather than gross-revenue ROAS — the only metric that speaks to the CFO.
- CAC vs CPA — separate media cost (CPA) from total acquisition cost (CAC).
- LTV:CAC ≥ 3 — the ratio that tells you whether the acquisition model is sustainable long-term.
- Target CPA = LTV × margin × acquisition allocation — a business-realistic formula rather than copy-paste benchmarks.
- MER ≥ 3 (e-com) or ≥ 5 (B2B SaaS) — the aggregated KPI that neutralizes attribution biases.
Sources
Official sources consulted for this guide:
FAQ
Should I optimize for CPA or ROAS?
If the value of each conversion is the same (e.g., B2B leads with uniform sales process), pilot on CPA. If value varies (e-commerce with carts from $30 to $500), pilot on ROAS. In SaaS, pilot on CPA with an LTV-aware target (CPA < 1/3 of the 12-month LTV).
Why does my Google Ads ROAS differ from my Google Analytics ROAS?
Attribution differences. Google Ads counts conversions it can attribute to an ad click (default last-click model or data-driven). GA4 measures conversions via its own model (cross-device, cross-domain). It's normal to have a 10-25% gap. Beyond that, check tracking (UTMs, Enhanced Conversions enabled).
How do I calculate my LTV if I just launched?
With no history, use estimated LTV = average cart × estimated sector retention × margin. For SaaS, take 12 months of retention at launch, to revalidate after 6 months of real data. For e-commerce, typical retention is 1.3 to 2.5 purchases per customer over 12 months.
What should I do if my CPC suddenly explodes?
3 frequent causes in order: a competitor raised their bids, your Quality Score dropped (landing + ad audit), or a match type change widened your targeting. Check in this order. See our guide to the 10 Google Ads mistakes for more details.