When a company starts scaling, the problem isn't "getting more leads," it's having **a system** that predictably transforms a stream of prospects into qualified appointments, then into clients. A sales funnel serves precisely this purpose.
When a company starts scaling, the problem isn't "getting more leads," it's having a system that predictably transforms a stream of prospects into qualified appointments, then into clients. A sales funnel serves precisely this purpose.
If you are looking for a sales funnel example with a ready-to-copy model, you are in the right place: below, you will find a complete template (steps, content, messages, KPIs, automations) that you can adapt to your B2B context (services, agency, SaaS, firm).
If you need a refresher on the concept, you can read our article: Sales funnel definition: simple explanation.
A good funnel is not a succession of tools. It is a sequence of logical engagements for a specific target client.
You cannot "optimize" a funnel if you don't know who should enter it. Define an ICP with simple criteria: sector, size, maturity, stack, intent signals.
To frame this properly: ICP (Ideal Customer Profile).
Your funnel will be much more effective if you propose an entry offer that is easy to accept (diagnostic, audit, workshop, template, scoped consultation). It serves to build trust and qualify, not to "sell everything" at once.
Typical B2B examples:
Audit (e.g., AI audit, RevOps audit, SEO audit)
Workshop (e.g., 2h scoping, process mapping)
Demo + action plan (if you sell a product)
The prospect rarely buys "because you posted on LinkedIn." They buy when a trigger appears: recruitment, new KPI to reach, tool migration, rising CAC, churn, board pressure.
You can link this subject to your steering metrics like CAC (Customer Acquisition Cost).
The model below is designed for an SME or scale-up selling: recurring services, projects, consulting, or SaaS with a non-immediate sales cycle.
Step | Goal | Typical Channels | Asset to produce | CTA (expected action) | KPI to track |
|---|---|---|---|---|---|
1. Discovery | Attract the right ICP | SEO, LinkedIn, partnerships, ads | 3 to 5 "problem" content pieces + 1 offer page | Read / subscribe / visit | Qualified sessions, CTR, % ICP |
2. Capture | Transform visitor into lead | Landing page, form, chatbot, newsletter | Lead magnet or contact request | Download / subscribe / ask for info | Visitor→lead conversion rate |
3. Nurturing | Progress trust + intent | Email, retargeting, content, events | Email sequence + 1 proof (case study, demo, ROI) | Reply / book a call | Open rate, click, reply, engagement |
4. Qualification | Identify real opportunities | SDR, qualifying form, short call | Questionnaire + scoring + MQL/SQL rules | Book qualified appointment | MQL→SQL, no-show, response time |
5. Closing |
You can deepen the qualification concepts with:
Copy-paste this block, then replace the fields in brackets.
Short promise: {measurable result} for {ICP} without {main pain point}.
Step 1, Discovery
Problems addressed: {P1}, {P2}, {P3}
Channels: {SEO / LinkedIn / Ads / Partners}
Content: {3 articles} + {5 posts} + {1 offer page}
KPI: {ICP sessions} + {CTR} + {engagement}
Step 2, Capture
Landing: {URL / page}
Capture offer: {checklist / guide / express audit}
CTA: {download / subscribe}
KPI: {visit→lead}
Step 3, Nurturing
Sequence: {5 emails over 10 days} or {weekly newsletter}
Proofs: {case study} + {before/after} + {methodology}
CTA: {15 min meeting} or {audit request}
KPI: {reply rate} + {meetings}
Step 4, Qualification
MQL Rules: {size} + {signal} + {score}
SQL Rules: {budget/urgency} + {use case} + {decision maker}
Owner: {SDR / Sales}
KPI: {MQL→SQL} + {no-show}
Step 5, Closing
Process: {discovery} → {proposal} → {validation}
Key objections: {O1}, {O2}, {O3}
KPI: {win rate} + {cycle} + {average value}
Step 6, Onboarding
Plan: {30-60-90 days}
KPI: {activation} + {satisfaction} + {NRR if applicable}
Let's take a realistic case for a company that wants to structure AI adoption (automation, integration, internal platforms):
Entry offer: AI opportunity audit (scoping, risks, roadmap)
Core offer: Development and integration (automation, platform, agents, etc.)
ICP: Executive, Ops, CIO, Head of Product in an SME or scale-up
Content that works well in B2B:
"How to reduce time wasted on {process}" (productivity angle)
"Checklist to scope {subject} without risk" (governance angle)
"Frequent mistakes before deploying {tech}" (mastery angle)
Example of related content on Impulse Lab:
Effective capture doesn't need to be complex. The most important thing: consistency between the consumed content and the offered asset.
Two simple options:
Lead magnet: "AI Audit Checklist in 30 minutes (10 questions)".
Appointment booking: "20-minute scoping session (free): evaluate 1 process to automate".
If your traffic is heterogeneous, a smart landing page with contextual chatbot can help guide towards the right action (without promising miracles, just by qualifying better).
Useful nurturing is mini-education guided towards a decision. Example of a 5-email sequence:
Email 1: "Here is the checklist + how to use it in 15 minutes"
Email 2: "3 signals that AI can improve {process} (and 2 signals where you shouldn't go there)"
Email 3: "Example of a 4-week roadmap: scoping → prototype → measurement → decision"
Email 4: "Frequent risks: data, security, adoption (and how to limit them)"
Email 5: "Proposal: we evaluate a use case together, then you leave with an action plan"
The goal isn't to write a lot. The goal is to address objections before the call.
At this stage, you want to sort:
"Curious" prospects
"Potential" prospects
"Ready now" prospects
Put light qualification in:
A form (size, tools in place, targeted process, urgency level)
Scoring
A mini-call (10 to 15 minutes) if you have a role like SDR
To industrialize, link your qualification to a CRM with MQL/SQL statuses and handoff rules (marketing → sales).
In service and project sales, closing often depends on three elements:
A clear scope (what is included, what is not)
A delivery plan (who does what, when, how we validate)
Risk reduction (pilot, milestones, weekly deliverables)
If you structure your execution with RevOps rituals, you gain predictability. See: RevOps (Revenue Operations).
The classic trap: tracking 25 metrics... and making no decisions.
Here is a minimal set of KPIs per step.
Step | Main KPI | Question to answer | If the KPI is bad |
|---|---|---|---|
Discovery | Qualified sessions (ICP) | Are we attracting the right people? | Review topics, targeting, distribution |
Capture | Visit→lead conversion | Is the asset promise clear? | Review landing, entry offer, friction |
Nurturing | Reply rate or Meeting/lead | Are we creating enough trust? | Add proofs, clarify objections |
Qualification | MQL→SQL | Are we sending the right leads to sales? | Adjust scoring + thresholds + questions |
Closing | Win rate + cycle | Is the sales process solid? | Review proposal, differentiation, follow-up |
To stay consistent with your costs, keep an eye on CAC, even if you are still small. As soon as you scale, it becomes a guardrail.
Automation serves two purposes: reducing response time and ensuring execution consistency. It does not replace strategy.
Some useful automations:
When a lead downloads {lead magnet}, create contact in CRM + tag {theme}
If the lead visits {offer page} 2 times in 7 days, trigger an email "Need help scoping?"
If the form contains {size} + {urgency}, create an opportunity and notify the person in charge
After a call, automatically send a recap + next steps + scheduling link
To frame the fundamentals: Automation.
Without minimal instrumentation (conversion, delays, steps), you don't know where to act. Measure little, but measure well.
This is almost always a problem of MQL/SQL definition, scoring, or internal SLA. The pages MQL and SQL can help standardize.
If your funnel contains 12 steps, you have likely created an internal process, not a customer journey. Simplify.
If you use cookies/trackers, ensure you respect consent rules. In France, the practical reference is the CNIL (cookies and other trackers).
What is the best sales funnel for a B2B SME? A short and instrumented funnel: Discovery → Capture → Nurturing → Qualification → Closing (then onboarding). The "best" depends mainly on your ICP, your entry offer, and your sales cycle.
What is the difference between a marketing funnel and a sales funnel? The marketing funnel aims primarily to attract, capture, and mature demand. The sales funnel transforms an opportunity into a client (qualification, proposal, closing). In practice, you must align both via MQL/SQL definitions.
How many steps should be in a sales funnel? In B2B, 5 to 6 steps are sufficient in most cases. If you exceed 7 steps, you risk creating friction and losing clarity.
Which KPIs to track to optimize a funnel? Prioritize: visit→lead conversion, Meeting/lead (or reply rate), MQL→SQL, win rate, sales cycle. Then add CAC and average value if you invest in acquisition.
How do I know if my leads are of good quality? Measure the MQL→SQL rate, the no-show rate, and the win rate by source. If a source generates volume but few SQLs, you have a targeting, messaging, or qualification problem.
Can AI improve a sales funnel? Yes, especially to accelerate qualification, enrich data, personalize sequences, and automate repetitive tasks. But AI does not compensate for a vague offer or a poorly defined ICP.
If you want to move from a "funnel on a slide" to a system that runs, Impulse Lab can help you frame priorities, choose the right use cases, and build reliable execution (audit, automation, integration with your tools, and adoption training).
To quickly identify opportunities and risks: Strategic AI Audit
To discuss your context (stack, acquisition, process, objectives): contact Impulse Lab
Got questions? We've got answers.
Leonard
Co-founder
Our team of experts will respond promptly to understand your needs and recommend the best solution.
Convert into client
Sales call, proposal, follow-up |
Deck, proposal, deployment plan |
Sign / pay / launch |
Win rate, sales cycle, pipeline velocity |
6. Onboarding (often forgotten) | Ensure adoption + expansions | Onboarding emails, CSM, training | 30-60-90 day plan + rituals | Activation / delivery / usage | Activation, satisfaction, upsell |
Continue reading with these articles
In 2026, the question is no longer "which AI is trendy?", but "**which AIs are actually useful for an SME** (and in what order to deploy them)". Between writing assistants, autonomous agents, RAG, automation, multimodal AI, and specialized tools, it quickly feels like...
Choosing an **artificial intelligence platform** is no longer just a "tool" decision; it's a decision about architecture, compliance, and internal adoption. By 2026, most organizations have already tested generative assistants, but many struggle to move from the "wow" factor to a reliable system...
In 2026, an SME "testing AI" without a method risks impactless POCs or rushed deployments. The good news is that an **SME AI lab** can be launched and set in motion in **90 days**, provided the lab is treated as a product (objectives, backlog, iterations), not as an innovation showcase.
