The First 30 Days on a WhatsApp CRM, in Order

Most WhatsApp CRM failures are decided in week one. A business gets access, imports twenty thousand contacts, sends a campaign, and by day three the number is restricted and the team has gone back to their personal phones. The tool gets blamed for a sequencing mistake.

This is the order that works, with the reasoning, so you can adapt it rather than follow it blindly.

Week 1 — make the inbox real

Goal: every incoming message reaches the right person, and the number builds a history of being welcome. No campaigns this week.

• Complete the business profile. Display name, description, address, website, logo, hours. A half-finished profile gets reported more often, and reports are what damage a new number.
• One login per person. Never a shared "sales" account — you lose the audit trail and cannot remove one person's access later.
• Decide who owns an unassigned chat. One named person, with a backup. This single decision prevents the most common failure, which is chats nobody claims.
• Write ten quick replies for the questions you answer most: price, availability, location, hours, delivery, payment.
• Put the WhatsApp link on your website and your Google Business Profile, with a prefilled message. Inbound messages are the cleanest possible start, because the customer messaged first.
• Send utility messages only — order updates, appointment reminders, payment confirmations — to customers from the last thirty days. Fifty to a hundred a day is plenty.

Import your full contact list if you like, but do not message it. The warm-up logic behind this is in the WhatsApp number warm-up guide.

Week 2 — structure, and the first conversations

Goal: the data starts being useful, and the number starts holding two-way conversations.

• Agree five tags and no more. New lead, in discussion, customer, repeat, not interested. Teams that start with thirty tags end up with none being used.
• Add the two custom fields you will actually filter on. Usually city and service interest. Resist the urge to recreate your whole spreadsheet.
• Tag the source on every new contact — which form, which QR, which ad. Without this you can report totals in month three and explain none of them.
• Send one message type that invites a reply: a service follow-up, a feedback request, a question about a recent order. Replies are the strongest positive signal a number can collect.
• Submit your first templates for approval. They take time, and you want them ready before you need them. Category first, wording second — the common rejections are listed in the template approval guide.

Week 3 — automate the four questions, then your first campaign

Goal: stop typing the same answers, and run one small, well-aimed campaign.

• Automate the lookups, not the conversations. Price, availability, location and order status cover most inbound volume. Keep the automation shallow — a two-question flow that hands over to a person beats a ten-step tree nobody finishes.
• Set a specific after-hours message. "We reply between 10 am and 7 pm, Monday to Saturday" produces far fewer repeat messages than "we will get back to you soon".
• Run your first campaign small. One segment, people who have interacted in the last two weeks, an offer that genuinely fits them. Keep it under half your current messaging limit.
• Then stop and read the numbers for a full day. Delivered, read, replied, blocked. If blocks are up, the list or the offer was wrong, and you have learned it on hundreds of people rather than thousands.

Week 4 — measure, and decide what month two looks like

Goal: know whether this is working, from your own data rather than impression.

• Median first response time. Median, not average — one message ignored overnight ruins an average and hides the real picture.
• Replies per hundred messages sent. The best single health number there is.
• Percentage of conversations with a recorded outcome. If this is low, your pipeline is a guess.
• Messages actually sent. Unglamorous, and it catches the real failure mode: a tool nobody uses.
• Quality rating. Should still be green. If it moved to yellow, stop marketing and read what to do when a number is restricted.

Then write down one number you want to change in month two, and change one thing to move it. A business that tries to improve five metrics at once improves none of them. The full calculation of whether it pays back is in how to calculate WhatsApp CRM ROI.

What to deliberately leave alone in month one

Everything below is useful later and a distraction now:

• A deep chatbot. Build it once you know what people actually ask, which takes a month of real conversations.
• Full CRM integration. Tempting, slow, and it blocks the work that matters. Export a CSV for now.
• Importing years of history. Nobody reads it, and it makes your contact list harder to use.
• Elaborate reporting dashboards. Four numbers, weekly, is more than most teams look at.
• Multiple numbers. One number, done properly, first.

The two mistakes that end the trial

1. The day-one bulk campaign. An old list, a festival offer, twenty thousand messages. The number is at medium quality by morning and the business concludes WhatsApp marketing does not work. What actually happened is that a brand-new number with no history sent promotional messages to people who did not expect them.

2. Nobody owns the inbox. Three people have logins, chats arrive unassigned, everyone assumes someone else is handling it. Replies get slower than they were on one person's phone, and within a fortnight the team drifts back to personal numbers. The fix is a named owner and a backup, decided in week one.

The short version

Week one: profile, one login per person, a named owner for unassigned chats, ten quick replies, and utility messages only to recent customers. Week two: five tags, two custom fields, source tagging, and messages that invite replies. Week three: automate the four lookup questions, then one small campaign to your warmest segment. Week four: measure four numbers and pick one to improve. Leave chatbots, integrations and extra numbers for month two.

Who does what, in writing, before day one

The single most common reason month one goes badly is that nobody is actually responsible. Four roles, and one person can hold more than one of them, but each must have a name against it:

• Inbox owner. Checks every morning that nothing is sitting unassigned, and is the fallback when an agent is away. Without this role, chats arrive and wait.
• Template owner. Writes and submits templates, tracks what is approved, and keeps a short list so nobody resubmits something that already exists.
• Numbers owner. Reads the four metrics once a week and says out loud whether they moved. Thirty minutes a week.
• Account owner. Holds the Meta Business Manager admin rights. This should be the business owner, not an agency and not an employee who might leave.

Write the four names on the same page as your response-time commitment and pin it where the team can see it. A tool with no owner becomes a tool nobody opens, and that failure looks exactly like the software not working.

What to do at the end of month one

Hold a thirty-minute review with the four numbers in front of you and answer three questions honestly:

1. Is the inbox actually being used, or has the team drifted back to personal phones? Messages sent per agent answers this, and it is the one that decides everything else.
2. Which question do customers ask most? That is your first chatbot flow in month two, chosen from data rather than guesswork.
3. Where did the enquiries come from? If the answer is "we do not know", fix source tagging before you spend anything on campaigns.

Then pick one number to improve and one change to make. Teams that leave month one with five initiatives finish month two with none of them done.

Frequently Asked Questions

What should I set up first on a new WhatsApp CRM?

The business profile, one agent login per person, and utility messages to recent customers. Not broadcasts and not a chatbot. The first week is about proving the inbox works and giving the number a history of welcome messages, which is what protects your ability to send later.

Why do businesses stop using a WhatsApp CRM after a month?

Two reasons, both avoidable. Either they started with a bulk campaign to an old list and the number was restricted in the first week, or nobody owned the inbox so chats sat unanswered and the team drifted back to personal phones. Both are decisions made in week one, not software faults.

Should I import all my contacts on day one?

Import them, but do not message them. Start by sending only to customers from the last thirty days, and only utility messages. An old list messaged on day one is the single most common cause of a new number losing its quality rating.

When should I build the chatbot?

Month two, after you have a month of real conversations to read. Build flows for the questions people actually asked, keep each flow short, and hand over to a person quickly. A chatbot designed before you have the data ends up answering questions nobody asks while missing the ones they do.

How many people should have access in the first month?

Everyone who will actually reply to customers, each with their own login, plus one owner. Keeping it to one person "until we settle in" usually backfires, because the rest of the team keeps using their personal numbers and you end up with two systems instead of one.