Quick answer
Use the free trial to confirm that the functions you need exist, understand how they work and complete a small controlled workflow from beginning to end. When you need to test real business volume, use a suitable paid plan and coordinate the test with WAB2C in advance.
Guide information
Best for
New evaluators, owners, project leads, IT and procurement
Time required
7 days with 30–90 focused minutes per day
Trial objective
Confirm fit and prepare a paid pilot decision
Not designed for
Production, bulk messaging, load, stress or SLA tests
01 — Start with the purpose
A free trial answers a fit question—not a scale question
The trial is a guided product evaluation. Use it to see whether one agent can receive and organise a conversation, build a simple flow, hand work to a human, connect an approved AI provider and exchange a few controlled test events.
A free trial answers “Does this workflow fit our business?” A paid, coordinated capacity test answers “How will the configured system behave at our expected volume?”
“Full platform access” means exploring available functions within the trial allowances. It does not mean unlimited contacts, seats, sessions, messages, API calls, AI tokens, onboarding work or support capacity. Faster or “Skip Waiting List” access changes when you can start—not how much capacity you receive.
Mark as done when: Your team agrees that the trial is for functional evaluation, not production benchmarking.
02 — Know the boundaries
Read the limits before inviting anyone
50 contacts
Use synthetic, internal or consented test contacts—not your full CRM.
1 agent
Nominate one operator; other reviewers can observe or review notes.
50 sessions
Run complete journeys; a session is not an individual message.
1 bot flow
Build one representative journey with fallback and handover.
Your AI key
Use a restricted test key and a low provider budget.
Separate costs
Messaging, provider and AI charges can sit outside the platform offer.
A session is not the same as a message. Treat it as a customer-service window containing a conversation. Do not interpret a session allowance as permission for unlimited message loops, concurrent calls or automated retries.
Mark as done when: You recorded the limits shown in your own dashboard and selected a test that stays comfortably inside them.
03 — Name the test
Choose the test that matches the question
Mark as done when: “Feature preview” or “functional test” is written at the top of your trial plan.
04 — Before Day 1
Prepare one journey, one operator and safe data
Choose a useful, repeatable journey: order status, a sales enquiry, support handover, COD confirmation or one webhook event. “Test everything” is not a usable objective.
Assign responsibilities
- Trial owner: owns the plan and decision
- Trial operator: uses the single seat
- Technical reviewer: checks API, webhook or AI
- Business reviewer: confirms process fit
Create a safe test pack
- 5–10 internal or consented contacts
- Redacted sample enquiries or orders
- One flow sketched on paper
- Five expected test conversations
- A restricted AI key and simple issue log
Never paste passwords, payment credentials, identity details, confidential payloads or unrestricted production keys into trial conversations or issue reports.
Mark as done when: One journey, one operator, safe test data and expected results are ready.
05 — The working plan
Seven days, seven clear acceptance checks
Learn the workspace and its limits
Find usage, subscription, Team Inbox, contacts, labels, flow builder, integrations, analytics and settings. Record what is active and what depends on approval.
The operator can locate the core features and explain every trial limit.
Test the Team Inbox
Use two or three test contacts. Receive a conversation, open the contact, reply, label it, search it and inspect its history. This is workflow evaluation—not a call-centre simulation.
One enquiry can be received, organised and completed by the trial operator.
Build one useful bot flow
Create a narrow journey such as order status: request a reference, handle a valid and invalid response, then offer human handover. Avoid a giant menu with dozens of branches.
The same flow produces the expected outcome in repeated low-volume tests.
Review templates, consent and costs
Use properly consented test recipients, distinguish service replies from approved outbound templates and confirm any Meta or provider cost before sending.
The team understands consent, approval dependencies and separate cost layers.
Test one integration or AI step
Use a development-style setup, one request at a time, small known payloads and a restricted AI key with a low provider budget. Log timestamps, status and identifiers.
One integration or AI-assisted step works and its dependencies are documented.
Run controlled UAT
Combine the journey into a written script: trigger, input, handover, operator reply, label and analytics check. Run it a few times at normal human speed.
Business and technical reviewers agree whether the journey fits or needs work.
Review evidence and decide
Classify each requirement as confirmed, configuration required, paid-plan validation required, provider-dependent, unresolved defect or out of scope.
The next step is based on recorded results—not impressions or an overload event.
06 — Keep it representative
Test each feature at the right scale
Contacts
Test creation, search, labels and groups with a small representative sample.
Team Inbox
Receive, read, label, search, respond and close with the single operator.
Flow Builder
Test logic, invalid input, fallback, handover and completion in one flow.
API
A successful transaction proves feasibility—not production throughput.
AI
Test selected redacted examples and decide when AI answers or hands over.
Analytics
Confirm useful events are captured; do not infer KPIs from a tiny sample.
07 — Interpret the result
A boundary is not automatically a platform defect
A trial may enforce limits with disabled actions, warnings, queues, throttling, temporary restrictions or quota errors. The 51st contact failing to import points to the contact cap; a second agent failing to join points to the seat entitlement.
However, a normal permitted action that repeatedly fails with valid input may be a genuine functional or configuration issue. Reproduce it once at controlled volume, capture evidence and report it. Do not create panic from a sudden, out-of-scope burst—and do not use limits to dismiss a reproducible defect.
08 — Make problems reproducible
Report a real issue with evidence—not only urgency
A useful report includes the workspace identifier, date and timezone, feature, exact steps, expected and actual results, redacted screenshots, relevant conversation or request IDs, browser or device, approximate test size and any provider error.
Classify the likely source
Severity should follow validated impact. Trial assistance is not the same as a contractual response-time SLA unless the applicable plan or written agreement says so.
09 — When volume matters
Plan capacity testing as an engineering exercise
If capacity is central to the buying decision, start with a suitable paid configuration and agree the test in advance. Payment alone does not authorise an unlimited stress test.
Send a written brief
State the business objective, journey, date, duration, contacts, agents, normal and peak rates, concurrency, ramp, payload sizes, integrations, monitoring, success measures, stop conditions, rollback and cleanup.
10 — Decide deliberately
Graduate from trial to a paid pilot—or stop with clarity
Document your expected contacts, seats, teams, shifts, conversation volume, flows, campaign needs, API traffic, AI tasks, reporting, onboarding, integrations, SLA needs and growth. Then choose the plan that fits the operating model—not merely the lowest headline price.
Stop: the core journey is not a fit
Pilot: configure normal business use
Capacity test: coordinate a measured paid test
Trial exit checklist
- One defined workflow tested end to end
- Trial limits respected
- Configuration gaps separated from defects
- External dependencies identified
- Seats, contacts, sessions and flows estimated
- Normal and peak usage defined
- Support and SLA requirement identified
- No unnecessary production data or secrets stored
Prove you understand the trial before extending it
Complete all eight questions correctly. A private 3-day extension coupon is revealed only after a 100% score, so you can use the extra time for purposeful testing rather than repeating the same uncertainty.
0%
0 of 8 answered
Questions answered
WAB2C free-trial FAQ
Is the WAB2C trial really free for seven days?
The current pricing page lists a 7-day platform trial with defined limits and a one-time setup and onboarding fee collected at signup. WhatsApp messaging, provider charges and third-party AI usage can be separate. Check the current offer and your dashboard before starting.
Does full platform access mean unlimited usage?
No. It means access to explore the available platform functions within the published contact, agent, session, flow and integration allowances.
Can I add 15 agents and four teams during the trial?
No. The current trial lists one agent. A larger multi-team evaluation needs an appropriately configured paid plan or written pilot arrangement.
Can I use the trial for real customer support?
Use it for controlled evaluation with internal or properly consented test contacts. Do not depend on the evaluation account as a production help desk.
Does the trial include unlimited API calls or AI tokens?
No. A token grants authorised access; it does not promise unlimited capacity. AI usage remains subject to the selected provider's quota, billing and policies.
Why did requests slow down when we sent many at once?
You may have reached a trial safeguard, account quota, provider limit or retry loop. Stop the burst, reproduce the core action at normal volume and report it if the controlled action still fails.
Does a free trial include a two-hour support SLA?
Do not assume so. A response-time SLA applies only when stated in the applicable plan or written agreement.
Can WAB2C guarantee Meta or provider approval and uninterrupted delivery?
No. Platform access, policy, pricing, approvals and delivery include Meta and provider dependencies. Review the WAB2C Provider Disclosure before launch.
Explore small. Learn clearly. Scale deliberately.
A successful trial is not the one that generates the most activity. It is the one that removes the most uncertainty and makes the next decision easier to defend.