Warm-up via browser extension: architecture, logic, setup
AndySendy academy

AndroidSender Browser Extension: 14-Day Warm-Up Plan and Setup

Warm-up does not guarantee protection from blocking

This is important to understand before you start.

Warm-up is the foundation. But list quality and message content affect results more than warm-up itself. Good warm-up will not save a bad campaign. It only gives the account a reserve of trust — how quickly that reserve is spent depends on what you write and to whom.

Important: warm-up and list quality work as a pair. Warm up the account and clean the list at the same time — not one after the other.

Glossary

Before reading further, learn these four terms — they appear in every section.

Response Ratio — the share of conversations in which the other party replied to your message. This is the main trust signal for the WhatsApp algorithm: the higher the reply rate, the “healthier” the account looks to the system.

Unanswered Counter — a running count of messages without a reply over 30 days. The WhatsApp algorithm tracks messages that go unanswered for 48 hours — a direct spam signal. The higher the counter, the higher the risk of blocking.

Trust pool — a set of real warmed accounts that are guaranteed to reply to incoming messages. In this course — the author's accounts, used as the “receiving side” during warm-up.

Spintax — syntax for variable templates. Lets you generate thousands of unique messages from one template. Example:

{Hi|Hello|Good day}, {how are you|how's it going|what's new}?

Each time, a random option is picked from each set of curly braces. Identical text across thousands of messages is one of the fastest spam-filter triggers.

Where the numbers in this plan come from

The plan rests on one principle: smooth growth without sharp jumps.

The WhatsApp algorithm tracks activity dynamics. A sudden spike — for example, from 20 messages to 100 in one day — is detected as an anomaly automatically. A smooth growth curve stays invisible to filters.

That is why the plan is built so that volume increases by roughly 20–25 messages each day. Exact values are not critical — the principle matters. If one day you send 5 messages more or fewer, that is fine. What is dangerous is doubling volume overnight.

Pauses between sends shrink in proportion to volume growth. In the first days — about 30 minutes between messages. By day 14 — about 4–5 minutes. AndroidSender randomizes the delay within each range automatically, so the rhythm never repeats — and identical pauses are what expose a bot.

Tip: do not try to speed up the plan by skipping days or doubling volume. Sharp jumps are the cause of most blocks during warm-up.

Three metrics to watch

These numbers are not theory. Accounts that stay within the ranges below work steadily. Those that fall below get restrictions.

Metric Normal Below normal — signal
Delivered (two grey ticks) ≥ 90% Dirty list — dead or non-existent numbers
Read (two blue ticks) ≥ 80% Irrelevant audience or weak first screen
Replied (any reply) ≥ 60% Change the message or the list

If numbers are below normal — do not keep sending in the hope you will “pick up speed.” Stop, find the cause, fix it.

Important: account blocking is not the only risk. A low Response Ratio gradually reduces deliverability: WhatsApp starts quietly “holding back” messages even before an explicit block.

Days 1–2: manual only

Do not touch the extension. The goal of the first two days is not volume but reply rate. Every message sent should get a reply. So write little, but get a reaction every time.

Text Voice Images Voice calls
Day 1 2–3 1 (5–10 sec) 1 1–2 (20–60 sec)
Day 2 3–4 1–2 (5–10 sec) 1–2 2–3 (30–120 sec)

Total: day 1 — about 5 interactions, day 2 — about 9.

Rules for these two days:

  • Send from at most 1–2 warmed accounts to the new one — no more
  • Always reply from the new account to every incoming message
  • Quote messages in replies, add reactions — this increases the “weight” of the conversation
  • Contacts saved in the device phone book

Tip: variety of content types — text, voice, image, call — mimics live user behavior. An account that only texts looks suspicious.

Fast start on days 3–4: 30 incoming messages

On days 3–4 the account is at its most vulnerable point: minimal history, the algorithm is watching under a microscope. This is where incoming traffic with guaranteed replies matters most.

How the service works

  1. The course author sends 30 messages from 30 different real warmed accounts to your new number
  2. You reply from the account being warmed to every incoming message
  3. Going forward, the extension continues the conversation with these same 30 accounts

In one day the account gets what usually builds up over a week. Response Ratio jumps sharply, Unanswered Counter stays at zero.

Cost

  • 500 rub — standalone one-time service
  • Free — included when you pay for full account warm-up (1,000 rub/month)

Days 3–14: AndroidSender enabled

From day 3 the extension is connected. Warm-up runs automatically.

Days 3–4: one-to-one mode

Each message goes to one contact and gets a reply before the next send. Volume is small — quality of each exchange matters, not quantity.

Days 5–14: one-to-many mode

Sending to several contacts with pauses. Volume grows smoothly each day.

Day Messages / day Pause between sends Mode
3 20 ~30 min (1767–1833 sec) one-to-one
4 43 ~28 min (1633–1700 sec) one-to-one
5 65 ~26 min (1500–1567 sec) one-to-many
6 88 ~23 min (1367–1433 sec) one-to-many
7 111 ~21 min (1233–1300 sec) one-to-many
8 134 ~18 min (1100–1167 sec) one-to-many
9 156 ~16 min (967–1033 sec) one-to-many
10 179 ~14 min (833–900 sec) one-to-many
11 202 ~9 min (533–600 sec) one-to-many
12 225 ~8 min (450–567 sec) one-to-many
13 247 ~6 min (317–450 sec) one-to-many
14 270 ~4 min (183–317 sec) one-to-many

Important: AndroidSender randomizes the delay within each range automatically. Identical pauses signal a bot. Random ones signal a human. Never set a fixed delay value manually.

Three cases from practice

Real scenarios — what went right, what went wrong, and how they recovered.

Case 1: Successful warm-up ✅

The account completed 14 days strictly by the plan. Metrics on day 14: delivered 96%, read 84%, replied 67%. Bulk messaging started in week 5 — the account ran 3 months without blocks.

What worked: clean list, not a single skipped day, Spintax with 4+ nesting levels.

Case 2: Poor metrics mid-way ❌

On day 9 deliverability dropped to 71% — the list had many inactive numbers. Sending was stopped immediately, the list was cleaned with a validator. After that metrics returned to normal in 4 days.

Lesson: validate the list before warm-up starts, not during. Dead numbers cannot be “waited out” — they must be removed.

Case 3: Recovery after a block ❌

The account got a temporary block on day 12 because of a sharp volume spike: the owner manually sent ~100 messages on top of automatic warm-up. After unblock — 5 days of clean warm-up without campaigns, only incoming and mutual messages. On day 6 the account returned to working mode.

Lesson: during warm-up do not add manual sends on top. The algorithm sees total volume, not the source.

AndroidSender settings before you start

Check every item before launch.

  • Spintax template — at least 4 nesting levels, no message should repeat
  • Typing simulation — enable
  • Randomize line breaks — enable
  • Working hours — daytime only in your region (night sends are an anomaly for a live user)
  • Delay range — strictly per the table above, not a fixed value

More on installing and configuring the extension:

Transition to bulk messaging

Campaigns start after the account has passed warm-up steadily for at least 10 days without warning signs.

Warm-up does not stop when campaigns start — it runs in parallel. Incoming from the trust pool keeps Response Ratio high. That is real protection during sending: warm-up replenishes the reserve of trust while the campaign spends it.

Tip: if a campaign loads the account harder than usual — for example, you expand the list or change the offer — temporarily increase warm-up intensity, do not reduce it.

If metrics drop

Stop campaigns immediately. Return to clean warm-up mode: only incoming and mutual messages, minimal volume. At least 3–5 days of recovery before the next attempt.

Typical causes of falling metrics and what to do:

Symptom Likely cause Action
Deliverability < 90% Dead numbers in the list List validator, remove inactive
Read rate < 80% Weak first screen of the message Rewrite the first 2 lines of the template
Replies < 60% Irrelevant audience or weak offer Change segment or rework the message
Sharp drop in all metrics Temporary restriction from WhatsApp Full pause for 5 days, warm-up only