Can I review and edit whacl’s customer reply before it sends?

You do not have to let whacl send every AI-generated message automatically. In review mode, your team sees the suggested reply, edits it when needed, and approves it before the customer receives anything.

whacl still reads the conversation and uses your catalog and policies to prepare the response. A person makes the final decision about what gets sent.

How review mode works

A customer sends a message through WhatsApp, Instagram, or Messenger. whacl reads the conversation and uses the business information you provided—such as products, prices, delivery rules, opening hours, returns, and booking policies—to draft a reply.

Instead of sending the reply immediately, whacl places it in your shared inbox for review. Your team can read the customer’s message, check the suggested response, change the wording or details, and send the edited version from the same conversation.

The customer does not see the draft while it is waiting for approval. They receive only the reply your team approves and sends.

What you can edit before sending

Review mode is useful when whacl has the right general information but the answer needs a decision about this customer’s situation. For example, whacl may draft a reply saying that a product is in stock and can be delivered tomorrow. Your team can change the delivery window if the route is full, offer a different product, or ask for an address before sending.

You can edit the wording and business details directly in the conversation. That is useful when the reply includes a delivery date, discount, replacement option, quoted price, or appointment time that needs confirmation.

You can also reject the draft and reply from scratch. The AI suggestion may still show what whacl understood from the conversation and your business information, while your team remains in control of the final message.

Review mode and auto-send are separate choices

whacl has two ways to handle AI replies: review mode and auto-send mode. Review mode holds the draft for approval. Auto-send sends the generated reply without waiting for someone to approve it.

You can start with review mode while checking how whacl uses your catalog and policies. Once your team is comfortable with the answers, you can use auto-send for routine questions and keep review for conversations where the wording or decision needs attention.

For example, a business might use auto-send for questions such as “What time do you close?”, “Do you deliver to this area?”, or “Is the blue size medium available?” when the answers are present in its current information. The team can keep review mode for refunds, complaints, custom orders, unusual delivery requests, or prices that need confirmation.

You do not have to choose between full automation and manual replies for the entire inbox. whacl lets your team review AI-prepared answers when they need control and send predictable answers automatically when they do not.

Your replies come from your information

Reviewing a draft is useful only when the draft is based on information your business can stand behind. whacl answers from the catalog and policies you provide, rather than requiring your team to write a separate script for every question.

That information can include product names, descriptions, prices, variants, stock information, delivery areas, delivery fees, opening hours, return rules, payment instructions, and frequently asked questions.

Your team can check whether the draft matches the current situation before sending it. If your delivery policy says orders normally arrive in two days but a particular area is delayed, review mode gives someone the chance to correct the answer.

If reviewers repeatedly correct the same price, qualification rule, or delivery condition, update the catalog or policy so future drafts start with the right information.

The same review process works across three channels

whacl brings WhatsApp, Instagram, and Messenger conversations into one inbox. Review mode therefore also applies to an Instagram direct message or Messenger conversation, not only to WhatsApp messages.

Your team can see which channel the message came from, read the conversation, edit the suggested reply, and send it from the shared whacl inbox.

This gives your team one place to review product questions from Instagram, delivery questions from WhatsApp, and older customer conversations from Messenger instead of checking three separate apps.

A human can take over with the full thread

Some messages need more than a small edit. A customer may be upset, ask for an exception, explain a complicated order, or need a decision that is not covered by your policies. In those cases, your team can take over the conversation directly.

The person taking over receives the full thread context, including what the customer asked, what whacl already answered, and which details have been discussed. The customer does not need to repeat the problem.

Reviewers can approve straightforward answers, correct details that depend on the specific order, and hand over conversations that go beyond the catalog and policies.

You can choose the model behind the reply

whacl lets you choose Claude, GPT, or Gemini from one setting. You do not need to rebuild your catalog or reconnect your channels to change the model preparing the reply.

The review process stays the same: whacl reads the thread and your business information, creates a draft, and either waits for approval or sends automatically according to the mode you selected.

You can compare how the models handle your actual customer messages. One may produce wording your team prefers, while another may handle a particular product question more clearly.

Changing the model does not replace the need for accurate business information. Claude, GPT, and Gemini can prepare a reliable answer only when the catalog and policies contain the details needed to answer the question. Review mode lets your team catch an incomplete or unsuitable answer before sending it.

You can see what replies are costing

whacl tracks cost per reply, so your team can see how much AI-assisted messaging is costing as conversations move through the inbox.

A reviewed reply still uses AI to prepare the draft, even though a person approves it. Cost-per-reply tracking lets you measure that activity instead of treating it as an untracked monthly expense.

The cost information can help you compare where AI is being used. If routine delivery questions make up a large share of replies, auto-send may be suitable for those answers. If a category regularly needs corrections or handoffs, review mode may be the better fit while you update the relevant policy or catalog information.

A practical way to start

If you are unsure about automatic replies, begin with review mode across your connected channels. Let whacl draft answers from your catalog and policies, then have a team member approve, edit, or replace them.

• Use review mode while checking product prices, stock answers, delivery rules, and return information.
• Edit drafts when the general policy is right but the customer’s specific order needs a different answer.
• Hand the conversation to a human when it involves a complaint, exception, sensitive issue, or decision outside your policies.
• Move repetitive, low-risk questions to auto-send after your team has reviewed enough drafts to trust the result.
• Use cost-per-reply tracking to see how much AI-assisted messaging each workflow uses.

You can change the level of automation as your business changes. A seasonal delivery delay may mean more replies need approval for a few weeks. After you update the catalog, you may be able to use auto-send again for routine questions.

The short answer

Yes—whacl can show your team the AI’s suggested customer reply before it sends. You can review it, edit it, approve it, or take over the conversation with the full thread visible.

You can also use auto-send for predictable, low-risk replies, choose Claude, GPT, or Gemini from one setting, and track cost per reply across your shared WhatsApp, Instagram, and Messenger inbox.