Back to Blog
WhatsApp Chatbot Costs: Two Budget Examples for French Buyers

WhatsApp Chatbot Costs: Two Budget Examples for French Buyers

WhatsApp Chatbot Costs: Two Budget Examples for French Buyers

Business owner comparing chatbot project estimates

Budgeting for a WhatsApp chatbot typically means planning for a range from a few hundred euros a month for a SaaS or BSP-managed setup up to a high-end custom multi-agent build, according to agency pricing commonly cited in the French market. Three levers decide where you land: Meta’s conversation fees, plus the markup your provider adds on top, the implementation model you choose, and the ongoing maintenance that keeps the bot useful. The breakdown below walks through the math so you can turn these ranges into a real line-item budget.

TL;DR:

  • Meta bills conversations by category; marketing usually costs most, while service may be free, so ask vendors to itemize markup, currency conversion, and VAT.
  • Agency FAQ builds start around €2,000; integrated agents often cost €6,000 to €15,000 or more, while complex projects can take two to three months.
  • Budget for content updates, quality checks, integration repairs, hosting, and human escalation; require separate monthly maintenance and support quotes before signing.
  • For a small shop handling 2,000 utility conversations monthly, the illustrative SaaS budget can reach about €400 monthly, or €4,800 annually.
  • Require a DPA, data residency terms, consent capture, audit records, and a clear support SLA; validate volume through a proof of concept before signing.
Botiqueai
Plan a WhatsApp Chatbot That Fits
BotiqueAI creates custom chatbots and automations for business needs, helping companies improve operational efficiency and customer relations.
Explore BotiqueAI

Table of Contents

How WhatsApp conversation pricing actually works

Meta bills WhatsApp Business API usage by conversation, not by individual message. Each 24-hour conversation window falls into one of several categories, and the category determines the rate you pay, according to pricing reporting on the French market. Understanding these categories is the first step to building an accurate quote.

  • Marketing conversations: opened when you initiate contact with promotional content, generally the most expensive category.
  • Utility conversations: transactional updates like order confirmations or shipping notices, usually billed lower than marketing.
  • Authentication conversations: one-time passcodes and verification messages, priced separately from marketing and utility.
  • Service conversations: opened when a customer messages you first within a support window; many providers bill these at zero or a reduced rate.

You almost never pay Meta directly. Most businesses work through a Business Solution Provider (BSP), and that provider adds its own markup on top of Meta’s base rate. Industry analysis of the French market indicates that BSP markups over Meta’s published conversation rates commonly fall within a certain range, which means the number on a BSP’s pricing page includes their margin, not just the raw Meta cost. That comes from BSP pricing comparisons, and it is worth asking any vendor to show you the Meta rate and their markup separately rather than a single blended figure.

Currency and VAT add another layer. Meta publishes rates in US dollars for some markets, so a BSP serving French clients either converts to euros at a published exchange rate or bills in dollars directly. Either way, ask for the conversion method in writing, since a provider that locks in a favorable rate today may adjust it later. VAT treatment also varies: some BSPs include VAT in the headline price, others add it at checkout, and a quote that does not specify which is incomplete. Finally, remember that recent product changes on Meta’s side, including the global rollout of Meta’s AI agent features for WhatsApp Business, can shift what counts as a billable interaction and what a vendor includes in a standard package, so scope and pricing should be revisited whenever Meta updates its tools.

Price bands for SaaS plans, agency builds, and hybrid setups

Once you understand the per-conversation math, the next decision is which delivery model fits your business: a managed SaaS or BSP plan, a custom agency build, or something in between. Each comes with a different cost structure and a different level of control.

SaaS and BSP-managed plans bundle Meta’s fees, a platform layer, and basic support into a predictable monthly charge. Typical cost components include:

  • A one-time setup or onboarding fee, often covering WhatsApp Business verification and template approval.
  • A recurring monthly subscription or per-seat fee for the platform and support.
  • Per-conversation overage charges once you exceed an included volume, billed at the Meta rate plus the provider’s markup.

Custom agency builds trade predictability for capability. A basic FAQ bot that answers common questions and routes complex queries to a human typically starts around €2,000 as a floor price in agency packages seen in the French market. A more capable agent that connects to your order system, CRM, or booking calendar and handles multi-step workflows generally runs €6,000 to €15,000 or more, depending on integration complexity. Multi-agent architectures, where separate AI agents handle distinct functions like sales qualification, support escalation, and order tracking, tend to start at €15,000 and scale from there. These figures reflect common agency package ranges rather than a single universal price list, so treat them as planning anchors and confirm scope against your own requirements before comparing quotes.

Delivery timelines track complexity: a basic FAQ bot can often launch within a few weeks, while a multi-agent system with deep backend integration can take two to three months from kickoff to production.

Hybrid approaches split the difference. Some businesses use a low-code automation layer to connect a SaaS chatbot to internal systems without a full custom build, which keeps upfront costs lower but adds a recurring fee for the automation tool itself. Others pair an agency-built bot with a managed hosting and monitoring service, paying more upfront for a custom agent but avoiding the ongoing engineering overhead of running it in-house.

Price bands for SaaS plans, agency builds, and hybrid setups — overview diagram

Hidden costs that change your total cost of ownership

The sticker price of a chatbot project rarely reflects what you actually spend over a year. Several recurring costs tend to catch buyers off guard, and each deserves its own line in your budget.

  1. Content and knowledge-base upkeep: someone needs to update answers, add new products or policies, and retire outdated responses, which typically runs a few hours a month whether handled internally or by a vendor retainer.
  2. Quality assurance and retraining: reviewing conversation logs to catch misunderstood intents and adjusting the bot’s responses is an ongoing task, not a one-time setup step.
  3. Integration maintenance: connections to a CRM, e-commerce platform, or analytics tool can break when those systems update, and middleware tools like n8n or Make often carry their own subscription fee on top of the integration work itself.
  4. Support and escalation infrastructure: routing unresolved conversations to a human agent, plus log retention and backup storage, adds a small but persistent operational cost.

BSPs and platform vendors also tend to layer feature add-ons onto their base pricing: premium templates, advanced analytics dashboards, and service-level agreement guarantees are frequently sold as separate line items rather than included by default. Reducing the support burden a chatbot creates is itself a documented reason businesses invest in automation in the first place, so weigh these recurring costs against the staff time they replace, not just against the sticker price.

Pro Tip: Ask every vendor to quote maintenance and support as a separate monthly line item, not bundled into “implementation,” so you can compare true ongoing cost across proposals.

Building your budget: formula and two worked examples

A reproducible formula keeps your budget from becoming a guessing game. Add together these components:

  1. Base platform fee or one-time project cost (SaaS subscription or agency build price).
  2. Estimated monthly conversations multiplied by the Meta rate for each category, multiplied by (1 plus your BSP’s markup).
  3. Seats and support costs for your team.
  4. Integration costs for any connected systems.
  5. Monthly maintenance (content updates, QA, retraining).
  6. A compliance allowance for GDPR-related setup and review.

Worked example A, small e-commerce site: say a shop sends mostly transactional utility messages, roughly 2,000 conversations a month, on a SaaS plan with a €150 monthly base fee. Adding estimated per-conversation charges, one support seat, and basic maintenance might bring the monthly total to somewhere near €400, or roughly €4,800 annually. These figures are illustrative inputs for the formula, not published market rates.

Worked example B, mid-market service business: say a company runs more marketing and open-window conversations, around 8,000 a month, on a custom agent built for €10,000 upfront with a €600 monthly maintenance and hosting retainer. Because marketing conversations cost more per unit, this business is far more sensitive to volume swings than Example A, and a slow month can meaningfully shift the monthly total even though the upfront investment stays fixed.

What GDPR and CNIL compliance actually costs

Compliance is not a box to check after launch. It is a set of concrete line items that belong in your initial quote, and skipping them creates financial risk later.

  • Data Processing Agreement (DPA) review: legal review and negotiation of a DPA with your BSP or chatbot vendor, since you remain responsible for how customer data is handled even when a third party processes it.
  • Consent capture implementation: building non-prechecked opt-in checkboxes, double opt-in flows, or QR-code based consent journeys, as guidance for WhatsApp business messaging in France recommends, each of which adds development hours to your project.
  • Audit logging and records retention: timestamping consent, storing opt-in sources, and maintaining records that demonstrate compliance if questioned.
  • Policy updates and periodic review: revisiting your privacy notices and data transfer disclosures as regulations or vendor terms change.

The stakes for skipping these steps are real. Regulatory scrutiny of WhatsApp’s data-sharing practices has previously drawn threatened fines in France, and compliance guidance for WhatsApp Business in France notes that CNIL enforcement has produced significant fines in past cycles. Treating compliance as a line item rather than an afterthought is cheaper than remediation after the fact.

How working with an agency changes your budget lines

Engaging an agency for a WhatsApp project replaces a pile of uncertain hourly estimates with two predictable budget lines: a project fee for the build and a monthly subscription for the production tool afterward. We handle chatbot development, WhatsApp integration, and the n8n and Make automations that connect a bot to your CRM or order system, which means compliance review, vendor management, and integration maintenance are folded into the engagement rather than left for your team to coordinate separately. For a deeper look at how these costs compound over several years, our chatbot TCO playbook walks through the full budgeting template.

Our take on procurement sequencing

The biggest mistake we see is buyers pricing the chatbot and the compliance work separately. Start with a proof of concept, measure real conversation volume before committing to a full build, and insist on a DPA, itemized per-conversation pricing, data residency terms, and a clear support SLA before signing anything.

— Botiqueai

Get a line-item quote before you commit to a platform

Most WhatsApp chatbot quotes bundle Meta fees, markup, and support into one opaque number, which makes real comparison nearly impossible. We start every engagement with a free initial audit that separates those costs so you see exactly what you are paying for and why, then build toward a working proof of concept rather than a long contract you cannot easily unwind.

Botiqueai

  • A line-item breakdown covering Meta conversation fees, BSP markup, platform cost, and integration work.
  • A compliance plan that accounts for DPA review and consent capture from day one.
  • A production path from proof of concept to a scaled deployment without renegotiating scope mid-project.

If you are ready to see what a WhatsApp deployment would actually cost for your business, reach out through our WhatsApp Business integration page and we will walk you through a quote built around your real conversation volume.

This article is general information, not a substitute for advice from a qualified lawyer. Consult a qualified legal professional about your own circumstances before acting on anything here.

FAQ

What is the typical price range for a chatbot?

A basic FAQ chatbot commonly starts around €2,000 as a one-time agency build, while a custom agent with backend integrations typically runs €6,000 to €15,000 or more depending on complexity. Ongoing monthly costs for maintenance, hosting, and conversation fees come on top of the initial build.

How much does WhatsApp business messaging cost?

WhatsApp Business API costs are billed per conversation, split into marketing, utility, authentication, and service categories, with service conversations often billed at zero or a reduced rate.

Can you get a chatbot for free?

Some basic chatbot builders offer free tiers for very limited volume or testing, but a production WhatsApp deployment with verified business messaging, templates, and support almost always carries platform or conversation fees. Free tools typically lack the compliance tooling, integrations, and support needed for a real customer-facing deployment.

What should I look for when comparing vendors?

Ask for a quote that separates Meta’s conversation fees, the provider’s markup, platform subscription, and integration work into distinct line items rather than one bundled number. Also confirm the vendor provides a Data Processing Agreement and clear terms on data residency before signing, since compliance gaps create financial risk later.

How long does a WhatsApp chatbot project take to launch?

A basic FAQ bot can often go live within a few weeks of kickoff, covering setup, template approval, and testing. A custom agent with multiple integrations or a multi-agent architecture typically takes two to three months from the start of development to a production launch.

Sources

© 2026 BotiqueAI — Reproduction prohibited without attribution.