What is a marketing funnel?

Most frameworks describe three to five stages. The simplest version uses three layers: Top of Funnel (TOFU / Awareness), Middle of Funnel (MOFU / Consideration), and Bottom of Funnel (BOFU / Conversion). Extended models add Loyalty and Advocacy stages to capture post-purchase value and referral loops.

At TOFU, the goal is reach — attracting strangers who match your ideal customer profile through SEO content, paid ads, social media, and events. At MOFU, the goal is relevance — nurturing engaged prospects with case studies, webinars, and personalized email sequences that position your solution against alternatives. At BOFU, the goal is friction removal — demos, trials, social proof, and pricing pages that make it easy for a ready buyer to say yes.

In B2B, each stage maps to a lead status that serves as a handoff gate: a raw lead becomes a Marketing Qualified Lead (MQL) when MOFU engagement crosses a scoring threshold; an MQL becomes a Sales Qualified Lead (SQL) when sales validates the fit; and an SQL becomes an Opportunity when a deal is created in the CRM. The precision of these definitions — and the SLAs governing each handoff — determines how much pipeline actually flows through to revenue.

A funnel works by combining content, channels, and automation calibrated to where a prospect is in their decision process. TOFU content is educational and non-promotional — it earns attention. MOFU content is comparative and proof-heavy — it earns consideration. BOFU content is commercial — it earns the decision.

In a modern B2B setup, marketing automation platforms (HubSpot, Marketo, Pardot) track engagement across touchpoints and score leads automatically. When a prospect hits a score threshold — say, downloading a buyer's guide plus visiting the pricing page — the system routes them to a sales sequence or rep queue without manual review.

Conversion rates between stages are the key operational metric. Industry benchmarks reported by Marketful (sourcing First Page Sage data) show B2B averages of roughly 2.3% for visitor-to-lead, 31–41% for lead-to-MQL, 13–21% for MQL-to-SQL, and 22–30% for opportunity-to-close. One of the biggest levers on conversion is speed-to-lead: a 2007 MIT and InsideSales.com study covering 15,000+ leads found that responding to inbound within five minutes makes a team 21 times more likely to qualify a prospect than waiting 30 minutes — a finding that has held up across numerous subsequent replications.

The terms are often used interchangeably, but they refer to different ownership and scope. The marketing funnel spans the full journey from stranger to qualified lead — it is marketing-owned and covers awareness, interest, and consideration. The sales funnel picks up from the SQL stage and tracks the deal from opportunity creation through to closed-won — it is sales-owned.

In practice, the two overlap at the MQL-to-SQL handoff, which is one of the most contested boundaries in B2B revenue organizations. Marketing defines what qualifies as an MQL; sales decides what qualifies as an SQL. Misalignment here — different definitions, slow handoff SLAs, lack of feedback loops — is a common cause of pipeline leakage and cross-functional friction.

A Revenue Operations (RevOps) function typically owns the full combined funnel — setting shared definitions, SLAs for lead response, and unified reporting that neither marketing nor sales can manipulate independently. In organizations with mature RevOps, the combined view is often called a revenue funnel or demand waterfall, and it spans from first anonymous touch to closed-won and beyond into expansion revenue.

The linear funnel has been widely declared dead, and the critique is valid: real buyers do not move in a tidy sequence from awareness to purchase. They research independently, loop back to earlier stages, consult peer communities, and often arrive at a vendor's pricing page having already made their shortlist through channels no CRM ever tracked — the so-called dark funnel.

6sense's 2025 Buyer Experience Report found that buyers contact vendors when they are, on average, 61% through their buying process — and the vendor they prefer before engaging still wins 80% of deals. BCG's January 2025 analysis, "It's Time for Marketers to Move Beyond the Linear Funnel," argued that digital transformation and AI have fractured consumer journeys into non-linear, unpredictable patterns that a static stage-gate model cannot capture, and proposed replacing the funnel with dynamic "influence maps" built from real behavioral data.

The funnel remains useful as a planning and measurement scaffold — it just cannot be taken literally as a predictive model of how any individual buyer behaves. Treating it as a directional framework for resource allocation (how much to invest at each stage) and a stage-gate system for lead management preserves its value while discarding its false precision. The right posture: design programs that meet buyers where they actually are (including dark-funnel channels), while using funnel metrics to track aggregate conversion health over time.

The dark funnel refers to the portion of the B2B buyer journey that happens outside trackable marketing channels — private Slack communities, LinkedIn DMs, G2 reviews, analyst briefings, podcast consumption, and peer word-of-mouth. These interactions leave no cookie, no UTM parameter, and no CRM record, but they are often the most influential touchpoints in shaping a buyer's shortlist.

Forrester estimates that B2B buyers complete 70–80% of their research in these untrackable channels before contacting sales. This means a traditional funnel with a clean "source: paid search" attribution model is systematically undercounting the influence of brand, community, and content investments that operate in the dark.

Practically, this means marketing funnels should be designed with the dark funnel in mind: invest in brand presence, analyst relations, review-site profiles, and community participation as TOFU plays even when their ROI is difficult to measure directly. The signal that those investments are working is often found in pipeline velocity and win rates, not in first-touch attribution reports.

Komo is designed for the moment where modern funnels break down most often: the gap between a prospect showing intent and a rep actually sending a relevant, timely message. Most teams detect a buying signal — a job change, a pricing page visit, a G2 review — and then lose hours or days to manual research and drafting before outreach goes out. By then the window has closed.

Komo monitors buying signals across your accounts continuously, researches the account and contact the moment a signal fires, and drafts the outreach and follow-up sequence — the repetitive work that lives between your CRM and your inbox. This is MOFU and BOFU acceleration: turning signal detection into personalized outreach in minutes rather than days.

The key design principle is human-in-the-loop: Komo handles detection, research, and drafting, but keeps a human on every send that matters. That preserves the quality and deliverability that fully automated systems sacrifice, while delivering the speed advantage that determines whether a signal becomes a meeting or a missed opportunity.