If you are researching Kommo + WhatsApp API, the direct answer is: Kommo handles CRM and customer service flow very well, but the real cost to operate WhatsApp at scale is not just the Kommo license. What determines ROI is the sum: Kommo license + official WhatsApp Business Platform (WABA) provider + WhatsApp conversations (charged by Meta) + process discipline. And this is where most fail, blow the budget, and conclude that “CRM doesn’t work.”
I am Luiz Otávio Gonçalves, a Mechanical Engineer who became a pioneer in Digital Engineering in Brazil. I implement Kommo and sales automation in practice, with data and execution. In this article, I will break down the numbers and show what changes in ROI when you move from the “owner’s WhatsApp” to a sales/customer service operation with multiple agents, queue, SLA, and traceability.
What you really pay to use WhatsApp on Kommo (without romanticizing)
Let's separate what is CRM cost from what is WhatsApp cost. Kommo is the “engine” of the funnel, automation, tasks, and history. Official WhatsApp is another world: it has a provider and charges per conversation.
- 1) Kommo license (per user): it gives you funnel, cards, automations, reports, and team operation.
- 2) Official provider (BSP) of the WhatsApp Business Platform: it’s who connects your number to official WhatsApp via API and charges a Monthly Fee and/or fee per message (depends on the BSP).
- 3) Meta’s charge per conversation: Meta charges for 24-hour conversations (marketing, utility, authentication, service). This cost varies by country and category and changes over time, so I can’t make up a number here. You validate it in Meta’s official pricing and your BSP contract.
If you want to understand Kommo’s licensing in depth (and avoid the trap of a “cheap” plan that becomes expensive), I’ve already detailed it here: Kommo CRM licenses: Which is the best for your company?.
The real fact I can state: Kommo is a predictable cost; WhatsApp is a variable cost
Kommo tends to be a more “fixed” cost (licenses per user). Official WhatsApp has a variable part (conversations). This creates a typical behavior in operations I help unlock:
- The team starts sending too many messages (without criteria).
- The lead replies off-hours, reopens the window, and you pay for the conversation again.
- Marketing sends templates to a cold base and you pay to “bother.”
- Customer service turns into “endless chat” instead of a decision process.
Practical conclusion: the numbers don’t add up because of the tool. They add up because of process + automation + governance. And this is engineering: method > improvisation.
Table: where ROI dies when you scale WhatsApp (and how I fix it)
| Problem (what happens) | Impact on cost | Practical correction (actionable) |
|---|---|---|
| Every lead becomes human customer service | Time + unnecessarily open conversation | Automatic screening (contact reason, minimum qualification, queue routing) |
| No SLA and no lead owner | Lead reopens conversation, becomes ping-pong | Distribution rules + tasks with deadlines + automatic follow-up “pull” |
| Long and aimless messages | Low conversion, more time per lead | Short scripts + decision questions + mandatory next step in pipeline |
| Marketing sends template to wrong base | You pay for the conversation and still damage reputation | Segmentation by stage + consent + reengagement window with clear offer |
| No funnel indicators | You don’t know where money is leaking | Rates per stage + response time + no-show rate + reopening rate |
If you want to go deeper into classic follow-up mistake (which is where many operations waste money), I recommend this article: Automatic sales follow-up: what it is, why almost everyone does it wrong, and how to set it up.
When Kommo + WhatsApp API is worth it (and when it’s NOT)
I’ll be direct because this saves your time and mine.
Worth it when you have at least one of these scenarios:
- a team (2+ agents) and you need queue, distribution, tracking, and management.
- Volume of leads that already make you lose opportunities due to delay/response.
- Consultative selling where history and context change the closing rate.
- Repurchase/recurrence (clinics, services, maintenance, contracts) and you need to reactivate the portfolio.
it's not worth it (or it becomes overkill) when:
- You are alone, have few leads, and have no process (will buy a tool to compensate for lack of routine).
- Your model depends on “mass blasting” without segmentation (this will give you high variable cost and operational risk).
- You don’t accept operating officially and want a “workaround” (risk of blocking and losing your number; I don’t implement this kind of thing).
If you’re in this doubt zone, there’s a very honest guide about where Kommo wins and where it loses on WhatsApp: Kommo and WhatsApp: official integration (API) vs unofficial and why this decides if you get banned.
Implementation playbook I use to close the numbers (without guesswork)
The goal is not to “have WhatsApp in the CRM.” The goal is to reduce cost per sale and increase conversion rate. I implement in layers, so you can deploy quickly and optimize with data.
Layer 1 — Funnel and rules (without this, automation just speeds up chaos)
- Pipeline by intention: pre-sale, qualification, proposal, negotiation, won/lost (and reasons).
- Mandatory fields to prevent the agent from “guessing” (origin, product, estimated ticket, deadline).
- Distribution rules: round-robin, by specialty, by region, by schedule.
- Response SLA by stage: new lead can’t wait 3 hours.
Layer 2 — Routine automation (what I automate first because it gives quick ROI)
- Automatic task creation (follow-up with realistic deadline).
- Confirmation messages (schedule, documents, next steps) with objective language.
- Internal alerts (VIP lead, lead without response, lead stuck in critical stage).
- Data hygiene: standardize name, phone, source, and avoid duplicates.
Layer 3 — WhatsApp Governance (where variable cost is controlled)
- Template policy: what is allowed, when it is allowed, and for whom.
- Segmentation by stage: templates only for those it makes sense for (not for a “dead” base).
- Goals per agent: it’s not about “sending messages,” it’s about moving the stage with quality.
- Metrics dashboard: time to first response, contact rate, stage advancement rate, closing rate, reasons for loss.
What few people measure (and what I use to prove ROI)
If you want to know if the operation is improving, I focus on five numbers that don’t lie:
- AHT (average time to first response) by channel and by agent.
- Contact rate: leads who responded / leads worked.
- Advancement rate: how many move from “new” to “qualified.”
- Cycle time: how much time from first contact to closing.
- Revenue per conversation (when it can be linked to source and sale).
If you want a more complete ROI calculation model (to justify investment and stop deciding by gut feeling), I recommend this one: How to measure the ROI of business automations.
“But Luiz, can you predict the monthly cost in R$?”
Partly yes, partly no. I can predict with high confidence:
- Kommo licenses (fixed per user)
- BSP monthly fee (if included in the contract)
- Implementation hours (project closed by scope)
What I does not I bet without looking at your data:
- Conversations charged by Meta (depends on volume, category, and team behavior)
- Window reopening due to process failure (this is 100% operational)
In practice, I use your history (last 60–90 days) to estimate volume and create a contact policy that controls variability. The goal is simple: more conversion with less useless conversation.
Decision checklist: are you ready to scale Kommo + official WhatsApp?
- You have a clear funnel and a mandatory “next step” at each stage?
- Do you agree to operate Official WhatsApp (and not a workaround)?
- You have responsible for operation (not just marketing)?
- You want tracking and predictability (and not just “more messages”)?
- Are you willing to train the team and enforce routine?
If you answered “no” to 3 or more, I don’t recommend starting with advanced automation. Start with process and discipline, then we automate and scale.
What I would do today if I were starting from scratch (without losing 6 months)
- Week 1: pipeline + mandatory fields + basic distribution rules.
- Week 2: standard messages (short) + automatic tasks + response SLA.
- Week 3: integration with official WhatsApp + queues + permissions + audit.
- Week 4: metrics dashboard + script adjustments + weekly improvement routine.
This is the shortest path to go from “tool installed” to operation running with ROI.
If you want me to implement this in your operation (with funnel, automations, governance, and metrics to prove return), the next step is clear: request a project.