Funnel stages and lifecycle definitions for B2B SaaS
A B2B SaaS funnel usually runs through five stages: lead, marketing-qualified lead (MQL), sales-qualified lead (SQL), opportunity and customer. Each stage needs a written entry rule and one team that owns it. In First Page Sage's benchmarks for B2B SaaS, between 26% and 51% of MQLs become SQLs depending on channel, so the stage rules decide most of what a funnel report shows.
Lead, marketing-qualified lead, sales-qualified lead, opportunity and customer, each with one written rule and one owner.
MQL to SQL conversion runs from 26% for paid search to 51% for SEO in First Page Sage's B2B SaaS benchmarks.
Stages that mean different things to marketing and sales. Every report built on them inherits the gap.
Write down the entry rule for each stage and check ten recent records against it.
What are the funnel stages in B2B SaaS?
Five working stages sit between a first contact and a paying customer. HubSpot's default lifecycle model is a common reference point, and its definitions are quoted below.
| Stage | HubSpot's definition | Usual owner |
|---|---|---|
| Lead | A contact "that has converted on your website" or through another interaction beyond a subscription | Marketing |
| Marketing-qualified lead (MQL) | A contact "that your marketing team has qualified as ready for the sales team" | Marketing |
| Sales-qualified lead (SQL) | A contact "that your sales team has qualified as a potential customer" | Sales |
| Opportunity | A contact "that is associated with a deal" | Sales |
| Customer | A contact "with at least one closed deal" | Customer success |
HubSpot's full model has eight default stages. It adds Subscriber before Lead, Evangelist after Customer for customers who have advocated for the company, and Other for records that fit none of the stages. The same stages apply to contacts and companies.
Why does each stage need a written rule?
Because a stage without a rule means whatever the person updating it thinks it means. An MQL might be a demo request to one marketer and a whitepaper download to another. Every conversion rate, forecast and cost per stage built on top of it then carries that difference.
A usable rule has three parts.
- The trigger. The action or score that moves a record into the stage, such as a demo request or a lead score above a set number.
- The owner. The team that can move a record in or out of the stage.
- The exit. What happens next, and within what time, such as sales accepting or rejecting an MQL.
The gap is easy to measure. In the 2026 audit of 104 Google Ads accounts, 78% of form fills never became an SQL. A funnel that counts form fills as leads, then reports on leads, is mostly counting people who will never buy.
What conversion rates should each stage have?
It depends heavily on the channel. First Page Sage publishes stage-by-stage rates for B2B SaaS from more than 50 clients over a decade, mostly companies with $10M to $100M in revenue.
| Channel | Visitor to lead | Lead to MQL | MQL to SQL | SQL to opportunity | Opportunity to close |
|---|---|---|---|---|---|
| SEO | 2.1% | 41% | 51% | 49% | 36% |
| 1.3% | 43% | 46% | 48% | 32% | |
| Webinar | 0.9% | 44% | 39% | 42% | 40% |
| 2.2% | 38% | 30% | 41% | 39% | |
| Paid search (PPC) | 0.7% | 36% | 26% | 38% | 35% |
The widest spread in percentage points is at MQL to SQL, where SEO converts at almost twice the rate of paid search. Later stages are closer together. That pattern points to where definitions matter most: the handoff from marketing to sales.
A second source gives a lower range for paid leads. SaaS Google Ads Benchmarks 2026 (GrowthSpree, April 2026), whose publisher states 300+ accounts but no sample per row, puts median MQL to SQL for paid search leads at 13% to 18%. The two sources use different samples and definitions, so the figures are not averaged. The full breakdown is in B2B SaaS funnel conversion benchmarks.
Where does the handoff from marketing to sales break?
At the MQL. It is the only stage that one team defines and another team acts on, so it is where the two teams' numbers drift apart.
Three signs show the handoff is broken.
- Sales ignores MQLs. Leads sit in the queue without a status change, and sales builds its own pipeline from outbound.
- Marketing hits its MQL target while pipeline falls. The MQL rule has been loosened to meet a number.
- Nobody records why a lead was rejected. Without a reason, the MQL rule cannot be fixed.
HubSpot stores the steps inside a sales-qualified lead in a separate field, Lead Status, so sales can record where a lead stands without changing the lifecycle stage. A rejection reason in that field is the cheapest way to see what the MQL rule is letting through. The difference between the two stages is covered in MQL vs SQL in B2B SaaS.
Can a record move backward through the stages?
Not automatically in HubSpot. HubSpot's documentation says its default automatic updates "will only move the stage forward." A record can be set to an earlier stage by hand, and tools can move it backward only if the stage value is cleared first.
That design keeps the funnel history clean, but it has a cost. A lead that went cold after becoming an SQL stays an SQL in reports unless someone resets it. Two habits keep the counts honest.
- Use Lead Status for "disqualified" and "nurture". The lifecycle stage then shows the furthest point a record reached, and Lead Status shows where it is now.
- Report on dated stage changes. A count of records that entered the SQL stage this month is more useful than a count of records sitting in it today.
How should funnel stages connect to ad platforms?
By sending each stage change back to the ad platform as its own conversion. Without that, bidding learns from form fills, and the 78% figure above shows why that is a poor signal.
Google's guidance is to create "a separate conversion action for each stage of the funnel" and to bid to one stage. HubSpot can send lifecycle stage changes to Google, LinkedIn, Meta, Microsoft and ChatGPT ads as ad conversion events. The setup is on the ad spend to revenue page and in how to send CRM stages to ad platforms.
Frequently asked questions
What is the difference between a lead and an MQL?
A lead is anyone who has shared contact details, for example by filling in a form. An MQL is a lead that marketing has judged ready for sales, against a written rule. In First Page Sage's B2B SaaS benchmarks, between 36% and 44% of leads become MQLs, depending on the channel.
Who should own the definition of an MQL?
Marketing and sales together, with RevOps writing it down. Marketing applies the rule, but sales is the team that acts on it, so a rule that sales does not accept produces leads nobody works. The definition should be reviewed whenever sales rejects a large share of MQLs.
How many funnel stages should a B2B SaaS company track?
Five working stages are enough for most companies: lead, MQL, SQL, opportunity and customer. More stages add detail but also more places for definitions to drift. HubSpot's default model has eight, including Subscriber, Evangelist and Other, and its stages can be customized or extended.
What is a good MQL to SQL conversion rate?
It depends on the channel. First Page Sage's B2B SaaS benchmarks put it at 51% for SEO, 46% for email, 39% for webinars, 30% for LinkedIn and 26% for paid search. A rate far below those for the same channel usually points to a loose MQL rule.
Should the funnel be tracked on contacts or companies?
Both, when the sale involves several people. HubSpot applies lifecycle stages to contacts and to companies. A company view shows how many accounts are in each stage, which matters when several contacts from one company would otherwise be counted as several separate leads.
Figures on this page come from published studies and platform documentation. Where the numbers come from
Talk your funnel through with a RevOps expert
A free consultation. Bring your funnel, CRM or reporting question and brainstorm the next step.