Sales funnel

A 35-person cybersecurity company can have 500 contacts in its CRM and still have almost no real funnel.

That happens when the team treats every form fill, webinar attendee, and target account as “top of funnel.” The same email sequence then goes to a compliance manager comparing vendors, a student downloading a checklist, and a founder who glanced at one blog post. The CRM looks busy. Buyers get irrelevant messages.

A sales funnel is useful only when it helps the team answer three questions: what does this account know, what has it done, and what should happen next?

Salesforce describes a funnel as the buyer’s journey from awareness to purchase. Salesforce's explanation is a reasonable starting point, but the tidy inverted pyramid can create bad habits. Buyers do not move neatly from awareness to interest to purchase, and a seller cannot force them through numbered boxes.

Start with the market, not the database

Before defining stages, decide who belongs in the funnel.

For the cybersecurity company above, the ideal customer profile might be a SaaS business with 100 to 1,000 employees, a security or compliance hire, and a sales motion that depends on SOC 2 or ISO 27001 evidence. A five-person design studio may read the company’s audit-readiness guide, but that does not make it a worthwhile sales target.

Separate account fit from buyer activity.

Fit includes company size, industry, geography, systems, and role. Activity includes asking for security documentation, replying to cold outreach, requesting a demo, or asking how implementation works. Intent data can show that an account is researching a subject, but it cannot tell you why. Three visits to an audit-readiness page could mean an upcoming audit, casual research, or homework.

My view is that teams overvalue activity because it is easier to count than fit. A low-fit account with ten page views is still a low-fit account. Fit determines whether the account deserves attention. Behavior determines what kind of attention it deserves.

Do not confuse the funnel with the pipeline

The funnel describes the buyer’s position. The pipeline describes the seller’s commercial process.

An account can be deep in evaluation while the opportunity remains in an early pipeline stage because the account executive has not confirmed budget, decision authority, or access to the buying committee. That is not necessarily a reporting error. The buyer may have done more work than the seller has verified.

Pipeline stages usually represent events the sales team can manage: discovery completed, technical review scheduled, proposal sent, procurement approved. Funnel stages represent changes in the buyer’s knowledge and commitment.

This distinction matters when marketing reports that 500 accounts are “in consideration.” Those accounts may have visited product pages, attended a webinar, or matched a technographic profile. They are not 500 qualified opportunities. Treating them that way creates inflated forecasts and annoyed sales teams.

Give stages observable thresholds

Stage names are not useful unless someone can point to the evidence behind them.

At awareness, the buyer has noticed a problem or encountered the category. For example, an operations leader at a payments company starts searching for reconciliation controls after an audit finding. A page view is weak evidence. A question about whether the product supports that company’s payment processor is much stronger.

At interest, the buyer is trying to understand possible approaches. They may subscribe to research, attend a webinar, or ask for a comparison. The seller should clarify the problem and the available options, not immediately push for a meeting.

At consideration, the questions become specific. Does the product integrate with NetSuite? Can it support regional data storage? How long will implementation take? The account is testing fit, so technical documentation and relevant customer examples matter more than broad claims.

Evaluation begins when the buyer takes an action that carries cost or risk. They request a demo, invite a security reviewer, start a trial, or agree to a proof of concept. A sales development representative should not treat all of these signals as equal. A trial from a target account that just hired a VP of Finance is different from a trial created with a personal email address and no product activity.

Purchase is the contract event, not proof that the customer has achieved anything. For a B2B product with a six-month implementation, onboarding and adoption belong in the broader customer journey. A signed contract can still turn into a poor sale.

Set movement rules that sales can verify

A cloud infrastructure company might define movement this way:

• An account moves from awareness to interest when a relevant person engages with infrastructure content and identifies a use case.
• It moves from interest to consideration when someone asks about architecture, pricing, or fit with the current stack.
• It moves from consideration to evaluation when the account agrees to a technical review, trial, or proof of concept.
• It moves from evaluation to purchase when the buying committee confirms requirements, commercial terms, and its decision process.

The exact rules will vary, but they should be based on behavior rather than an arbitrary score. “Lead score above 70” hides the evidence. A score may combine a perfect account profile with meaningless email opens, or reward a poor-fit account for downloading several assets.

Record why the stage changed. A useful CRM note says, “Director of Engineering replied, named the current vendor, described a migration deadline, and requested a technical review.” It does not say, “MQL created.”

This also keeps handoffs honest. Someone who downloads a guide may belong in nurture. A director who describes the current vendor, names a deadline, and asks about migration effort has earned a direct sales conversation.

Find the leak by segment and trigger

A funnel shows where momentum stops. It does not explain the cause.

Suppose a 120-person software company gets strong engagement from target accounts, but very few reach evaluation. The issue could be weak discovery, unclear technical requirements, or a failure to involve security and finance stakeholders. More top-of-funnel activity will not fix any of those problems.

Compare movement by account segment, source, role, and trigger. If accounts with a newly hired compliance executive progress while accounts without a clear trigger stall, the team may need trigger-based prospecting. If opportunities reach technical evaluation and then stop at procurement, the sales process may be missing security documents, data processing terms, or a credible implementation plan.

Look for the specific break:

• Which target accounts stop responding?
• Which role enters first, and which roles appear later?
• How long does each stage take?
• Which business triggers appear in deals that move?
• Are account executives rejecting the handoffs, or are buyers rejecting the message?

A funnel metric without this context produces bad corrections. More contacts can hide poor fit. More meetings can hide weak qualification. More pipeline can hide a buying committee that has never agreed on the problem.

Let buyers move backward

A buyer might visit a comparison page, request a demo, return to implementation documentation, involve procurement, and then ask a basic question about the category. That is normal. Salesforce also notes that buyers can skip stages, revisit earlier questions, or move backward.

Stage changes should not be treated as permanent progress. If a buyer asks for pricing before confirming the business problem, answer the pricing question without pretending the account is ready to buy. If an opportunity goes quiet after a proposal, the next move may be to rebuild the business case rather than send another generic follow-up.

The funnel should tell the seller what uncertainty remains. Early conversations create relevance. Middle-stage conversations test fit. Late-stage conversations help the buying committee make and defend a decision.