Short answer: B2B lead generation metrics should run from cost per lead down to revenue: cost per SQL, lead-to-SQL rate, win rate, pipeline coverage, customer acquisition cost (CAC), CAC payback and LTV:CAC. Calculate ROI as (gross profit from new customers − total acquisition cost) ÷ total acquisition cost, and judge each lead source on cohorts that have had a full sales cycle to mature. As a guide, an LTV:CAC above 3:1 and CAC payback under 12 months indicate healthy unit economics.
Which B2B lead generation metrics matter, and how do you calculate them?
The metrics form a ladder. Each rung answers a different question, and a cheap lead at the top only matters if it survives to the rungs below. Write these definitions into your CRM and dashboard notes before arguing about any number.
| Metric | Formula | What it tells you |
|---|---|---|
| Cost per lead (CPL) | Programme cost ÷ leads captured | Top-of-funnel efficiency only; easy to game |
| Lead-to-MQL and MQL-to-SQL rate | MQLs ÷ leads; SQLs ÷ MQLs | Targeting and offer quality; whether sales accepts what marketing sends |
| Cost per SQL | Programme cost ÷ sales-qualified leads | The first quality-adjusted cost; the number to compare channels on |
| SQL-to-opportunity rate | Opportunities ÷ SQLs | Whether discovery calls find real projects with budget |
| Win rate | Won ÷ (won + lost) opportunities in the cohort | Sales effectiveness and fit; exclude deals still open |
| Average contract value (ACV) | First-year contract value ÷ new customers | How much each win is worth |
| Sales cycle | Median days from lead created (or SQL) to won | How long to wait before judging a channel |
| Pipeline coverage | Open pipeline value ÷ revenue target for the period | Whether there is enough pipeline to hit target |
| Pipeline velocity | (Opportunities × win rate × ACV) ÷ sales-cycle days | Revenue the pipeline produces per day |
| CAC (fully loaded) | (Marketing + sales cost in period) ÷ new customers | What a customer really costs |
| CAC payback (months) | CAC ÷ (monthly revenue per customer × gross margin) | How long until a customer repays its acquisition cost |
| Customer lifetime value (LTV) | Annual revenue per customer × gross margin × expected customer years | Gross profit a customer generates over its life |
| LTV:CAC | LTV ÷ CAC | Whether growth creates or destroys value |
| Marketing ROI | (Gross profit attributed − cost) ÷ cost | Return per rupee spent, on a stated attribution rule |
Three definitional choices change every result, so state them on the report. Which costs are in "programme cost": media only, or media plus agency, tools, content and salaries? Which stage definitions you use: HubSpot's lifecycle stage definitions describe an MQL as a record "marketing has qualified as ready for the sales team" and an SQL as one "your sales team has qualified as a potential customer"; write the observable criteria using our lead qualification framework. Revenue or gross profit: ROI on revenue overstates return for any business with real delivery costs.
Operational metrics such as speed to first contact, contact rate and data-quality checks belong on the weekly dashboard; their definitions are in our leads-to-revenue dashboard guide.
How do you work out B2B funnel maths? A worked example in ₹
Example (all figures are assumptions to replace with your CRM history): a Noida-based SaaS company sells HR software to Indian mid-market firms. Average contract value is ₹6,00,000 a year at 75% gross margin. The target is ₹60 lakh of new contract value next quarter, which means 10 new customers. Last year's CRM shows these stage rates:
- Lead to MQL: 30%
- MQL to SQL: 40%
- SQL to opportunity: 50%
- Opportunity to won: 25%
- Median sales cycle: 120 days
Work backwards from the target, dividing by each rate in turn:
| Stage | Calculation | Needed in the quarter | Cost per unit (₹15 lakh marketing) |
|---|---|---|---|
| Customers | ₹60 lakh ÷ ₹6 lakh | 10 | ₹1,50,000 (marketing CAC) |
| Opportunities | 10 ÷ 25% | 40 | ₹37,500 |
| SQLs | 40 ÷ 50% | 80 | ₹18,750 |
| MQLs | 80 ÷ 40% | 200 | ₹7,500 |
| Leads | 200 ÷ 30% | 667 | ₹2,249 |
Assume marketing spends ₹15 lakh in the quarter (media, content, tools and the marketing team) and the sales team costs another ₹15 lakh (salaries, incentives and their share of overheads). Then:
- CPL = ₹15,00,000 ÷ 667 = ₹2,249
- Cost per SQL = ₹15,00,000 ÷ 80 = ₹18,750
- Fully loaded CAC = (₹15,00,000 + ₹15,00,000) ÷ 10 = ₹3,00,000
- Year-one ROI = (10 × ₹6,00,000 × 75% − ₹30,00,000) ÷ ₹30,00,000 = (₹45 lakh − ₹30 lakh) ÷ ₹30 lakh = 50%
The point of the exercise: a ₹2,249 CPL means nothing on its own. If a new channel offers ₹1,200 leads but only 10% become MQLs instead of 30%, its real cost per MQL is ₹12,000, worse than the current ₹7,500. Compare channels on cost per SQL, then confirm on cost per opportunity once a cycle has passed. For typical Indian CPL ranges by channel, see our cost-per-lead benchmarks, and to run your own funnel, use the lead funnel calculator.
How much pipeline coverage do you need?
Pipeline coverage is open pipeline value divided by the revenue target for the same period. The common rule of thumb is 3x, but that number only fits one situation. Salesforce's pipeline coverage guide notes that the 3x rule "works only when your win rate is exactly 33% and your sales cycle is exactly one year," and gives a per-rep formula:
Pipeline coverage = annual quota ÷ win rate ÷ (365 ÷ sales-cycle days)
In the example, each of five account executives carries an annual quota of ₹1.2 crore with a 25% win rate and a 120-day cycle:
- ₹1.2 crore ÷ 0.25 = ₹4.8 crore of pipeline needed across the year
- 365 ÷ 120 = 3.04 cycles a year
- ₹4.8 crore ÷ 3.04 = about ₹1.58 crore of open pipeline per rep at any time
For the quarter, the simpler check is coverage = 1 ÷ win rate: at 25% you need 4x, so 40 opportunities worth ₹2.4 crore against a ₹60 lakh target. A team winning 40% needs only 2.5x. Dave Kellogg makes the same point in "The Self-Fulfilling 3x Pipeline Coverage Prophecy": a fixed 3x target gets met because reps are pressured to hit it, not because it predicts bookings. He recommends targets derived from each team's historical conversion. Two practical rules follow:
- Set coverage per segment from your own win rate. Enterprise deals with a 15% win rate need about 6.7x; SMB deals at 35% need under 3x.
- Clean the pipeline before measuring coverage. Close out deals with no activity for two full cycles; stale deals inflate coverage and hide a shortfall.
Pipeline velocity puts the same inputs into one number: 40 opportunities × 25% × ₹6,00,000 ÷ 120 days = ₹50,000 of new contract value a day. Raising any of the first three or shortening the cycle raises velocity, which makes it a useful way to compare improvement ideas.
What are good CAC payback and LTV:CAC ratios for B2B?
CAC payback asks how many months of gross profit a customer needs to repay its acquisition cost. In the example, monthly revenue per customer is ₹6,00,000 ÷ 12 = ₹50,000; at 75% margin that is ₹37,500 of gross profit a month.
CAC payback = ₹3,00,000 ÷ ₹37,500 = 8 months
LTV needs a retention assumption. If 30% of customers leave each year, the average customer stays about 1 ÷ 0.30 = 3.3 years:
LTV = ₹6,00,000 × 75% × 3.33 = ₹15,00,000, so LTV:CAC = ₹15,00,000 ÷ ₹3,00,000 = 5:1
The most widely cited guidelines come from David Skok's SaaS Metrics 2.0: the best SaaS businesses have an LTV:CAC "higher than 3, sometimes as high as 7 or 8"; many recover CAC "in 5-7 months"; and profitability looks "anemic" when payback runs beyond 12 months. Skok stresses that these are guidelines and that early-stage companies often miss them while improving. Treat them as a sanity check, not a target.
| Reading | What it usually means | Action |
|---|---|---|
| LTV:CAC below 1:1 | Each customer destroys value | Stop scaling spend; fix pricing, retention or the channel |
| 1:1 to 3:1 | Marginal; little room for error | Improve conversion rates before adding budget |
| 3:1 to 5:1 | Healthy | Scale the channels with the lowest cost per SQL |
| Above 5:1 with payback under 6 months | Possibly under-investing | Test more spend while capacity holds |
Two cautions. For services businesses or one-off projects, use the gross profit of the first project plus realistic repeat business rather than a subscription lifetime. And cap expected customer years (we use three to five) so a low churn assumption does not inflate LTV.
Which conversion rate moves B2B ROI the most?
Re-run the example with one change at a time, holding CPL at ₹2,249 and sales cost at ₹15 lakh:
| Scenario | Leads needed | Marketing cost | CAC | Payback | LTV:CAC |
|---|---|---|---|---|---|
| Base case | 667 | ₹15.0 lakh | ₹3.00 lakh | 8.0 months | 5.0 |
| Win rate falls from 25% to 18% | 926 | ₹20.8 lakh | ₹3.58 lakh | 9.6 months | 4.2 |
| MQL-to-SQL rises from 40% to 50% | 533 | ₹12.0 lakh | ₹2.70 lakh | 7.2 months | 5.6 |
| Both changes | 741 | ₹16.7 lakh | ₹3.17 lakh | 8.4 months | 4.7 |
A seven-point fall in win rate costs more than a ten-point gain in MQL-to-SQL saves. That is why we review lost-deal reasons with sales before we touch ad budgets. In practice, MQL-to-SQL often improves fastest by calling leads sooner and tightening forms; win rate improves through better ICP fit and fewer deals with no decision-maker involved.
How do attribution and cohorts change B2B ROI?
Use cohorts, not calendar months. With a 120-day cycle, the deals closed in March came mostly from leads created in October to December. Compare a channel's spend in a month with the SQLs, opportunities and wins from leads created in that month, and label cohorts younger than one sales cycle as immature. Until a cohort matures, judge it on cost per SQL and cost per opportunity.
Choose one operational attribution rule and disclose it. GA4 now offers only data-driven and last-click models; Google's GA4 attribution help confirms first-click, linear, time-decay and position-based models were removed in November 2023. B2B journeys with several contacts and offline meetings are rarely visible in GA4 anyway, so keep the original source and the latest source on each CRM contact and opportunity, as set out in our CRM revenue attribution guide. Report:
- Sourced pipeline: opportunities whose first recorded source was the channel.
- Influenced pipeline: opportunities where any contact touched the channel before the opportunity was created. Never add sourced and influenced together; they double-count.
- Self-reported source: a "How did you hear about us?" field on demo forms. It often credits podcasts, LinkedIn posts and referrals that tracking misses.
Feed qualified stages back to ad platforms. Bidding on form fills optimises for cheap leads. Importing SQL or opportunity stages lets Google and Meta optimise toward the rungs that drive ROI; our guide to tracking qualified leads from Google Ads to CRM covers the set-up.
What should a monthly B2B ROI review decide?
- Confirm the cost boundary. Same cost lines as last month; GST and agency fees handled consistently.
- Check data completeness. Share of leads with a source, share of lost deals with a reason. Below 90% on either, fix the data before drawing conclusions.
- Review cost per SQL by channel for the latest cohort and cost per opportunity for cohorts one cycle old.
- Review coverage by segment against 1 ÷ win rate, after closing out stale deals.
- Update CAC, payback and LTV:CAC quarterly, not monthly; small monthly samples swing too much.
- Make one decision per channel: scale (cost per SQL below target for two cohorts and sales has capacity), fix (one stage clearly below its historical rate), or stop (below target after a full cycle with adequate spend and follow-up).
Do not pause a channel because one week's CPL rose, and do not scale one because lead volume jumped. To see what each input costs in India, read our B2B lead generation cost guide; if you want this measurement built into your CRM, see our B2B lead generation services.
Sources
- Salesforce: pipeline coverage: definition, per-rep formula and why fixed 3x fails.
- Kellblog: The Self-Fulfilling 3x Pipeline Coverage Prophecy: case for coverage targets derived from historical conversion.
- David Skok, SaaS Metrics 2.0: LTV:CAC above 3 and CAC recovery guidelines.
- HubSpot KB: lifecycle stages: MQL, SQL and opportunity definitions.
- Google Analytics Help: attribution models: GA4 data-driven and last-click models; others removed November 2023.