Back to all articlesMessaging cost planning

Budget a multi-message journey without charging every step to every recipient

Alactic Editorial9 min read

Prepared with AI assistance by Alactic Editorial. Examples identified as hypothetical are illustrative, not customer results.

A four-message journey does not necessarily send four messages to everyone who enters it. Some recipients leave after the first message, some enter alternative branches, and some messages are not delivered. Multiplying the starting audience by every planned step hides these differences.

A more useful budget follows each step’s parent population, applies a stated continuation assumption, estimates delivery, and then calculates applicable charges. The result shows both expected spend and where uncertainty matters most.

The worked example below models an already defined journey. It does not design triggers, conversation flows, or execution permissions. All audiences, percentages, rates, and costs are hypothetical planning inputs, not Alactic prices or customer results.

1. Separate eligibility, delivery, and charging

Keep four quantities distinct:

  • Eligible volume: recipients who qualify for a particular step after its conditions and exclusions.
  • Planned attempts: messages expected to be attempted for those recipients.
  • Expected deliveries: planned attempts multiplied by an estimated delivery proportion.
  • Applicable charged deliveries: delivered messages to which the relevant charge applies.

This example assumes one planned attempt per eligible recipient at each step. If an existing journey includes repeat attempts or multiple messages within a step, add separate rows. Otherwise, the budget will undercount message opportunities.

Meta messaging costs depend on delivered messages, category, and recipient market. Check exact rates and applicable exceptions against the current WhatsApp Business Platform pricing information. Do not assume every delivered message is billable.

Use a separate row whenever market, category, rate, or charging treatment differs. A blended rate belongs in a summary derived from those rows, not in place of them.

2. Establish the cohort and branch boundaries

Hypothetical journey: a business budgets one complete cohort of 10,000 initially eligible recipients: 6,000 in fictional Market A and 4,000 in fictional Market B. These labels do not represent actual countries or rate-card entries.

  • Step 1: an initial marketing message to the starting eligible population.
  • Step 2: a marketing reminder for a subset of Step 1 delivered recipients.
  • Step 3: a utility confirmation for a different subset of Step 1 delivered recipients.
  • Step 4: a marketing follow-up for a subset of Step 2 delivered recipients.

Steps 2 and 3 are mutually exclusive branches. Step 4 follows Step 2, not Step 3. Each recipient retains the same market assignment throughout this example. In a real worksheet, use the applicable recipient-market classification rather than guessing from where someone lives.

The estimate covers the whole cohort, including later steps that might fall in another billing period. That makes it a cohort budget, not necessarily a forecast for one invoice.

Permission is a prerequisite, not a budget adjustment. Honour opt-outs. Approved templates are used to initiate Platform conversations and outside the 24-hour customer-service window, which opens or resets with a user message. Approval does not guarantee delivery. Check the current WhatsApp Business Messaging Policy for precise requirements; a cost estimate does not establish permission to send.

3. Give every continuation percentage a denominator

“Half continue” is not a usable assumption until the worksheet identifies half of whom. Here, continuation percentages apply to delivered parent messages in the same market.

HYPOTHETICAL CONTINUATION AND DELIVERY INPUTS

Step 1 — parent: starting eligible audience
A: 6,000 eligible; 96% delivery
B: 4,000 eligible; 90% delivery

Step 2 — parent: Step 1 deliveries
A: 50% remain eligible; 95% delivery
B: 60% remain eligible; 90% delivery

Step 3 — parent: Step 1 deliveries
A: 30% qualify; 98% delivery
B: 25% qualify; 96% delivery
Recipients here do not also enter Step 2.

Step 4 — parent: Step 2 deliveries
A: 20% qualify; 95% delivery
B: 15% qualify; 90% delivery

In Market A, 80% of Step 1 deliveries enter one immediate branch: 50% enter Step 2 and 30% enter Step 3. The remaining 20% enter neither. In Market B, the equivalent split is 85% entering a branch and 15% entering neither.

These are net eligibility assumptions: opt-outs and other exclusions are already removed. Do not subtract them twice. If your data gives response rates before exclusions, calculate net eligibility separately.

The model also assumes undelivered parent messages create no downstream eligibility. If the defined journey includes another path for those recipients, model that path explicitly rather than inflating an unrelated continuation percentage.

4. Establish consistent rates and explicit charging populations

The following hypothetical rates apply throughout one modeled pricing period. Journey position does not determine the rate: all marketing rows with the same market and pricing treatment use the same input.

HYPOTHETICAL RATE INPUTS — currency units (CU)

Market A | marketing | charged: CU 0.040 per delivery
Market B | marketing | charged: CU 0.060 per delivery
Market A | utility   | charged: CU 0.015 per delivery
Market B | utility   | charged: CU 0.020 per delivery

Keep rates as row-level inputs, but record their shared basis. A change to the Market B marketing rate for this pricing period must reach every affected row, not just the initial message.

For utility messages, the example conditionally assumes that checking the applicable pricing rules establishes two populations: charged deliveries and deliveries covered by an applicable exception. It does not identify or claim entitlement to any actual exception.

HYPOTHETICAL UTILITY DELIVERY SPLIT
Use only if the applicable exception is confirmed.

Market A total utility deliveries: 1,693.44
  Charged population:              1,270.08
  Exception-covered population:      423.36
  Charged share: 1,270.08 / 1,693.44 = 75%

Market B total utility deliveries:   864.00
  Charged population:                691.20
  Exception-covered population:      172.80
  Charged share: 691.20 / 864.00 = 80%

The 75% and 80% shares summarize these subrows; they are not adjustable budget discounts. In practice, derive the populations from message-level classification under the current rules. If no applicable exception is confirmed, do not use these illustrative excluded populations. Budget the relevant deliveries under the verified charging treatment instead.

5. Fill the reusable journey worksheet

The artifact below keeps the parent calculation visible and separates utility charging treatments. Fractional recipients are statistical expectations across the cohort, not executable recipient counts.

HYPOTHETICAL FILLED WORKSHEET

Row formulas:
Eligible = parent volume × net continuation share
Delivered = eligible × delivery proportion
Cost = charged deliveries × applicable unit rate

STEP 1 — MARKETING
A: Eligible 6,000; delivery 96%
   Delivered and charged: 5,760
   Rate CU 0.040; cost CU 230.4000
B: Eligible 4,000; delivery 90%
   Delivered and charged: 3,600
   Rate CU 0.060; cost CU 216.0000
STEP 1 TOTAL: CU 446.4000

STEP 2 — MARKETING REMINDER
A: Eligible 5,760 × 50% = 2,880; delivery 95%
   Delivered and charged: 2,736
   Rate CU 0.040; cost CU 109.4400
B: Eligible 3,600 × 60% = 2,160; delivery 90%
   Delivered and charged: 1,944
   Rate CU 0.060; cost CU 116.6400
STEP 2 TOTAL: CU 226.0800

STEP 3 — UTILITY CONFIRMATION
A: Eligible 5,760 × 30% = 1,728; delivery 98%
   Total delivered: 1,693.44
   Charged subrow: 1,270.08 × CU 0.015 = CU 19.0512
   Conditional exception subrow: 423.36; cost CU 0
B: Eligible 3,600 × 25% = 900; delivery 96%
   Total delivered: 864
   Charged subrow: 691.20 × CU 0.020 = CU 13.8240
   Conditional exception subrow: 172.80; cost CU 0
STEP 3 TOTAL: CU 32.8752

STEP 4 — MARKETING FOLLOW-UP
A: Eligible 2,736 × 20% = 547.2; delivery 95%
   Delivered and charged: 519.84
   Rate CU 0.040; cost CU 20.7936
B: Eligible 1,944 × 15% = 291.6; delivery 90%
   Delivered and charged: 262.44
   Rate CU 0.060; cost CU 15.7464
STEP 4 TOTAL: CU 36.5400

To reuse this structure, retain fields for cohort, step, parent, market, category, eligibility, delivery, charging treatment, rate, and cost. Add the rate-card period and the evidence date for each assumption. Split charging treatments beneath the delivery total so that the same recipients are not counted twice as additional deliveries.

6. Reconcile the total and identify the largest contributor

Sum unrounded row costs, then round the reporting total. Prematurely rounding expected volumes to whole recipients creates avoidable reconciliation differences.

HYPOTHETICAL COHORT RECONCILIATION

Eligible message opportunities:
  Step 1                         10,000.00
  Step 2                          5,040.00
  Step 3                          2,628.00
  Step 4                            838.80
  Total                          18,506.80

Expected delivered messages:      17,379.72
Applicable charged deliveries:   16,783.56
Conditional exception deliveries:   596.16

Expected Meta messaging cost:
  Step 1                         CU 446.4000
  Step 2                         CU 226.0800
  Step 3                         CU  32.8752
  Step 4                         CU  36.5400
  Total before rounding          CU 741.8952
  Reported total                 CU 741.90

Separate hypothetical planning lines:
  Provider/platform allocation   CU 120.00
  AI usage allowance             CU  18.00
Combined planning total          CU 879.90

The provider/platform allocation is a fictional share of a fixed-period expense, not necessarily incremental cash caused by this cohort. The AI allowance is separate and does not imply that each message uses AI. Replace both with the applicable billing basis and usage forecast.

Step 1 contributes approximately 60% of expected Meta messaging cost. Step 2 contributes about 30%; together they account for approximately 91%. These are the first rows to investigate when prioritizing uncertain inputs.

The model contains 18,506.8 expected message opportunities rather than the 40,000 implied by four messages for every starting recipient. That difference reflects conditional structure, not a claimed saving or delivery result.

7. Test uncertainty without changing the pricing logic

A sensitivity case should change a named input and all its dependants. For a hypothetical rate sensitivity, increase Market B’s marketing rate from CU 0.060 to CU 0.070 for the entire modeled pricing period.

The affected charged deliveries are 3,600 in Step 1, 1,944 in Step 2, and 262.44 in Step 4: 5,806.44 in total. Multiplying by the CU 0.010 increase adds CU 58.0644. Expected Meta cost becomes CU 799.9596, and the combined planning total rounds to CU 937.96. Changing only one step would require a documented reason, such as a different applicable rate-card period.

Test the conditional utility treatment separately. If none of the illustrative exception-covered deliveries qualifies, the added cost is 423.36 × CU 0.015 plus 172.80 × CU 0.020: CU 9.8064. With all other base inputs unchanged, expected Meta cost becomes CU 751.7016 and the combined total rounds to CU 889.70.

Continuation changes propagate too. Increasing Step 2 eligibility changes Step 4’s parent volume. Recalculate both, and keep mutually exclusive Step 2 and Step 3 shares within the available Step 1 deliveries. A higher-response case and a higher-reminder case cannot independently claim the same recipients.

8. Compare actuals at the layer where the forecast failed

When costs differ from the estimate, avoid immediately replacing the unit rate with a blended actual. First locate the mismatch:

  1. Eligibility: compare branch entrants with the specified parent population. Check for double-subtracted exclusions or overlapping branches.
  2. Delivery: distinguish campaign preparation, template review, attempted messages, and delivery. These are separate states.
  3. Charging: verify market, category, charging treatment, and rate-card period. Reconcile exception-covered populations separately.
  4. Scope: check for repeat messages, extra cohorts, or later steps appearing in another reporting period.
  5. Other costs: reconcile provider/platform and AI expenses without folding them into a Meta message rate.

Give provisional assumptions an owner and an evidence date. Lower delivery may reduce expected messaging charges while revealing an operational problem; it is not automatically a budget success.

The useful handoff is a reconciled cohort worksheet with explicit branch denominators, verified pricing inputs, and a named largest-cost step. Comparing those same fields with actuals makes the next estimate more explainable—not merely more precise-looking.

Messaging cost planningWhatsApp BusinessJourney budgeting

Put the next conversation to work.

Explore Alactic Communications for WhatsApp conversations, campaigns and AI-assisted workflows. Choose the product and plan that fit your team.

Explore Communications