If you are a B2B industry and your “CRM” in practice is a mix of spreadsheet + salesperson’s WhatsApp + lost email, the fastest way to unlock predictability is design a funnel that respects the industrial cycle (diagnosis, engineering, proposal, validations, purchase, implementation) and force the process to happen inside the CRM. It’s not about changing tools; it’s about implementing a method with data, owners, and triggers.
I am Luiz Otávio Gonçalves, a mechanical engineer who migrated to Digital Engineering. What I see every month in industry is the same: the team “even has CRM,” but the project stalled because they tried to copy an e-commerce/infoproduct funnel. Industry doesn’t sell like that. Here I will show you an angle that almost no one mentions: the industrial funnel needs to have technical stages and risk checkpoints, otherwise you just digitize chaos.
Who it’s for (and who it’s not for)
Serves for B2B industries that:
- sell projects, machines, recurring industrial services, maintenance, parts, or contracts;
- have sales cycles longer than 15 days (usually 30–180+);
- need to involve engineering, procurement, finance, and sometimes post-sales;
- want Predictability and governance (without “each salesperson doing it their own way”).
Is not useful (still) if:
- you can’t define a standard “next step” by sales type;
- the owner wants the CRM to “solve” the lack of product, price, deadline, and decent proposal;
- the team refuses to log activity (no data, no management).
The #1 mistake in industry: funnel without an engineering stage (and no risk owner)
In industry, sales die for three recurring reasons:
- Bad lead (outside the ICP, no real pain, no budget);
- Slow proposal (engineering becomes a bottleneck and no one notices);
- Inconsistent follow-up (the opportunity disappears and reappears “out of nowhere”).
When the funnel lacks a technical stage, the salesperson pushes everything to “Proposal sent” and starts praying. That’s the opposite of a process. I prefer a funnel where the CRM exposes bottlenecks and demands action.
The industrial funnel model I implement (and why it works)
This is a framework I adapt by vertical (metallurgy, automation, machining, boiler making, maintenance, integrator, etc.). Notice it has commercial stages and technical, with entry/exit criteria.
| Stage in the CRM | Objective | Criteria to advance (DATA) | Owner |
|---|---|---|---|
| 1) Qualification (ICP) | Filter what’s viable | Segment + application + capacity to serve + desired deadline | SDR/Sales |
| 2) Technical diagnosis | Understand scope and constraints | Checklist completed (process, layout, standard, capacity, input) | Sales + Engineering |
| 3) Feasibility / Pre-engineering | Avoid impossible proposals | Risk classified (low/medium/high) + estimated deadline + assumptions | engineering |
| 4) Proposal in progress | Turn scope into price and deadline | Items and alternatives defined (option A/B) + base cost | Engineering + Sales |
| 5) Proposal Sent | Ensure traceable sending | Proposal attached + date/time sent + expected decision | Sales |
| 6) Negotiation / Adjustments | Converge on terms | Objection map recorded + next action scheduled | Sales |
| 7) Approval / Purchasing | Navigate procurement | Requested documents + approval status | Sales + Admin |
| 8) Closed won | Generate perfect handoff | Order/contract + conditions + kickoff scheduled | Sales |
| 9) Closed lost | Learn and adjust ICP | Standardized reason + competitor + lesson | Management |
The trick here is simple: each stage has a verifiable criterion. If there’s no criterion, it becomes guesswork and you go back to the spreadsheet.
What really changes when you stop depending on the salesperson’s WhatsApp
I’m not against WhatsApp. I’m against WhatsApp becoming the company’s database. When the conversation stays in the individual’s hands:
- if the salesperson leaves, the history leaves with them;
- no one audits cadence and follow-up;
- there’s no reliable forecast;
- engineering receives demand without context and without standards.
The CRM becomes the “central nervous system” when you implement 3 things:
- Automatic interaction logging (email, WhatsApp API when applicable, calls);
- Mandatory tasks per stage (minimum cadence);
- Standardized handoff for engineering and operations.
Field checklist (without this you can’t optimize)
Industry needs fields that generic CRM doesn’t provide out of the box. I implement at least:
- Application (where it fits in the client’s process);
- Capacity / volume (number);
- Materials (critical input);
- Standards (NR, ISO, client requirements);
- desired deadline and possible deadline;
- Competitor and “why they win”;
- Target margin (range) and discount limits;
- Technical risk (low/medium/high) with justification;
- Next action with date (mandatory).
This is what allows turning CRM into management: filter, compare, learn, repeat.
Industrial cadence: follow-up that doesn’t annoy or disappear
Cadence is not spam. Cadence is control of next step. In industrial B2B, I like to work with windows (adjust according to ticket and urgency):
- Diagnosis to proposal: follow-up every 2–3 business days, because engineering is usually the internal bottleneck.
- Proposal sent until decision: weekly follow-up + event triggers (e.g., “when purchasing, request document”).
- Negotiation: follow-up according to milestones (meeting, counterproposal, technical visit, pilot).
If you want to structure this with automation, I have already written about the basics of the topic in Automatic sales follow-up. Here the difference is: in industry, automation must respect the technical cycle and not seem robotic.
How I measure CRM ROI in industry (without self-deception)
CRM delivers ROI when it becomes a lever for 3 numbers:
- Response time (lead → first contact; diagnosis → proposal);
- Advancement rate between stages (qualification → diagnosis; proposal → negotiation);
- Win rate and Cycle time.
I don’t “guess” ROI. I calculate it based on what changed in these indicators. If you want a direct calculation guide, I detailed it in how to measure the ROI of business automations (applies perfectly to CRM + sales automation).
Tool: Kommo, GoHighLevel or other? I start with the process
I implement Kommo and GoHighLevel, but my opinion is unpopular: industry shouldn’t start with the tool. Start with funnel design + fields + cadence + minimal integration with your stack (email, telephony, WhatsApp when it makes sense, ERP if necessary).
To avoid it becoming a religious debate, I use an execution rule:
- If the operation is very WhatsApp-centered and needs a fast funnel, with strong team adoption, Kommo tends to fit well.
- If you are an operation that needs to consolidate marketing + broader automations and has maturity to manage more layers, GHL may make sense.
If you are at the moment of choosing a tool, I recommend reading first how to choose the ideal automation tool — because the common mistake is buying a “cannon” and not having ammunition (process + data).
Real price (objective data) that impacts the decision
Without making up numbers: CRM has license cost and implementation cost. For Kommo, the public and stable data is that there is 14-day trial (this already allows validating funnel, fields and team usage before paying). I have a specific post explaining how to use this with criteria: Kommo 14-day trial it’s official and you can test what really matters in your scenario.
Regarding project investment (implementation + automation + integrations), I prefer to treat it as engineering: it depends on scope, number of funnels, level of automation and integrations (especially with ERP). I detail ranges and what is included in the project here: How much it costs to implement Kommo (implementation service).
Implementation in 10 business days (executable roadmap)
If you want to start from zero and stop “tinkering with CRM” without finishing anything, here is a roadmap I use in quick projects:
- Day 1–2: map ICP, sales types, stages and criteria (one page, no novel).
- Day 3: build funnel in CRM + mandatory fields + loss reasons.
- Day 4: create email/WhatsApp templates (short, objective messages) and tasks per stage.
- Day 5: set up simple automations: task creation, alerts, proposal SLA, stage owner.
- Day 6–7: integrate the minimum: email, forms/landing, lead source, and if applicable WhatsApp API.
- Day 8: train team with real cases (5 real opportunities become the “pilot”).
- Day 9: review bottlenecks, adjust stages, lock critical fields (without this data gets dirty).
- Day 10: publish simple dashboard (pipeline by stage, aging, time to proposal, win rate).
The focus is: Run. Then you optimize. Method > improvisation.
What I don’t automate (because it goes wrong)
Automation is a lever, not a crutch. I don’t automate:
- deadline promise without engineering validation;
- “automatic” proposal when the scope changes all the time (becomes an expensive error);
- aggressive message in follow-up (industry is relationship and reputation);
- lead scoring without minimum data history (score without data is decoration).
I automate what is repeatable, auditable and with clear rules: lead distribution, tasks, alerts, cadence, reminders, stage update by event.
Conclusion: the right industrial funnel gives you predictability — and forces the process to exist
The goal is not to “have a CRM”. The goal is to build a commercial system that consistently closes contracts, even with a long cycle, multiple decision-makers and engineering bottleneck. When the funnel has technical stages, criteria and risk owner, CRM stops being software and becomes operation.
If you want me to design and implement this funnel in your industry (with automations, governance and integration with what you already use), the next step is simple: request a project.
FAQ — questions I get asked the most
1) What is the best CRM for B2B industry?
It depends on your process and channels. I choose based on funnel adherence, ease of team adoption and minimal integrations (email, telephony, WhatsApp when applicable, and sometimes ERP). If you start with the tool without process, you will get stuck with any of them.
2) How long does it take to see results with CRM in industry?
When the funnel is well defined and the team logs activity, you can first see results in Response time and proposal time (weeks). Win rate and predictability usually improve in complete cycles (30–90+ days), depending on your ticket and cycle.
3) Do I need to integrate with ERP for it to work?
Not to start. I start with the minimum viable and integrate ERP when there is clear gain: pulling customer registration, order status, billing, or avoiding rework of typing. Poorly done integration becomes cost and confusion.
4) How to prevent CRM from becoming “just another place to fill out”?
With 3 things: mandatory fields only for what is decisive, tasks per stage (cadence) and weekly management looking at numbers (aging, advancement rate, proposal time). Without management, CRM becomes decoration.
5) Is it worth using WhatsApp in industry or is it better to remove it from the process?
It’s worth using, but with governance. WhatsApp can be a quick contact channel; the mistake is it becoming the repository of history and decision. History and next steps need to be in the CRM for the company (not the individual) to control the sale.