Workflow Automation / Plan-fit comparison and test protocol / Published v0.1
Make vs Albato for Small Workflows: Current Plan Fit and Two Reproducible Tests
The useful answer depends on trigger type, step count, polling frequency, volume, and failure behavior—not on which platform displays the smaller unit count.
The short answer
- 01
Albato may fit a rare two-step Google Sheets-to-Telegram polling flow when 15-minute latency is acceptable and useful events stay inside the current Free transaction allowance.
- 02
Make is the current plan gate winner for a custom webhook, more than two steps, or the webhook-to-Sheets-and-Telegram test that must run without buying Albato Pro.
- 03
Choose neither yet when duplicate handling, partial failure, step retry, or a full rerun has not proven one event ID produces one row and one alert.
Not sure whether either platform fits the workflow?Estimate whether the workflow is worth automating
Decide on the workflow constraints, not the brand
Start with the trigger, step count, latency, volume, and failure behavior. Compare platforms only after those constraints are explicit.
- C1
Trigger cost and latency. Polling can consume units even when no useful event arrives; a webhook changes both latency and the billing topology.
- C2
Step and feature gates. A lower unit count does not help when the free plan cannot run the required webhook or number of steps.
- C3
Useful monthly events. Compare 20, 100, and 1,000 events against the actual plan and complete workflow, not one isolated module.
- C4
Duplicate and retry effects. A repeated event or full rerun can create business-side duplicates even when the platform reports a successful execution.
- C5
Account evidence. Before calling the result hands-on, preserve the plan, diagram, run IDs, counters, outputs, errors, and redacted screenshots from both platforms.
What is verified and what is still pending
The plan and billing statements below come from current public documentation. The usage numbers for the two proposed workflows are models to reproduce, not screenshots from completed CriteriaDesk account tests. That distinction is the reason this page says current plan fit rather than hands-on winner.
Current plan-fit decision
Which workflow are you actually building?
The two platform links below are affiliate links. CriteriaDesk may earn a commission at no extra cost to you. The relationship does not change the documented plan gate or allow a platform to win an untested failure check.
Plan-fit hypothesis
Albato can fit the rare two-step polling case
Use this narrow answer only when 15-minute polling is acceptable, the automation has a trigger plus one action, and useful monthly events remain within the current Free allowance. The 24-hour counter test is still required before this becomes a reproduced result.
Current plan gate
Make is the executable starting point for the webhook fan-out
Make currently exposes custom webhooks and can run the three-step test on Free. Albato Free is limited to two steps and does not include the required custom webhook/API path. Reproduce both counters before publishing a hands-on cost winner.
Stop result
Choose neither platform yet
If a duplicate event, a failed Telegram step, a step retry, or a full rerun can create duplicate rows or alerts, a low unit price is not a safe result. Run the fault pack and keep the process manual until one event ID reliably produces one row and one alert.
No commercial handoff is shown for this result.
Select a workflow shape to reveal the current documented answer.
Current public plan facts — checked August 4, 2026
| Area | Make | Albato |
|---|---|---|
| Free | 1,000 credits, two active scenarios, 15-minute minimum interval, seven days of logs | 100 transactions, five automations, no more than two steps, 15-minute interval, seven days of logs |
| Lowest paid entry observed | Core: $12 monthly, or $9 monthly when billed annually, with 10,000 credits | Pro: $22 monthly, or $15 monthly when billed annually, with 1,000 transactions |
| Standard meter | a standard non-AI module action normally consumes one credit; each polling check is an operation | the trigger is not charged; each successful post-trigger step consumes a transaction for each item |
| Custom webhook path | available; incoming webhook processing consumes credits | custom webhook/API, webhook partners, and trigger filters are not available on Free |
Verify these facts on Make pricing, Make operations, Make credits, Make webhooks, Albato pricing, Albato pricing update, and Albato transactions.
The official sources are not perfectly consistent. Make’s affiliate page mostly describes 12 months of commission but one FAQ still says 24. Albato’s plan page and dedicated message describe a seven-day trial while an FAQ still says 14. Albato materials also vary on the exact automatically purchased transaction bundle. CriteriaDesk uses the conservative public interpretation and does not publish the disputed bundle size.
Test A — Google Sheets to Telegram with 15-minute polling
Build the same two-step automation in both accounts. Observe the same synthetic sheet for 24 hours, add 10 unique rows plus two repeated event IDs, and send to separate private Telegram chats.
The preregistered monthly model is:
- Make:
2,880 polling checks + E message actions; - Albato:
E successful post-trigger transactions.
At 20 useful events, that model is roughly 2,900 Make credits versus 20 Albato
transactions. It predicts an Albato fit for this narrow low-volume case, because
equivalent 15-minute polling would exceed Make Free. It is not yet a measured
result. Hourly Make polling changes the comparison to 720 + E, but also accepts
four times the detection delay.
Albato publishes the current Google Sheets and Telegram integration path. The paired test must preserve the empty polling runs, actual detection times, and both final counters.
Test B — custom webhook to Sheets and Telegram
Send the same exportable set of 20 synthetic JSON events to both platforms:
custom webhook → create Google Sheets row → send Telegram alert.
The preregistered unit model is approximately 60 Make credits and 40 Albato transactions for 20 successful events. The smaller Albato number is not the commercial answer: Make can run this path on Free, while Albato Free does not allow the required custom webhook or three steps. Albato therefore needs an eligible trial or Pro plan for the paired test. See Make webhooks and Albato’s webhook guide.
The current plan-gate hypothesis is Make. The hands-on result remains pending until both histories, counters, outputs, and redacted screenshots exist.
The failure test can change either answer to neither
After 20 correct events:
- resend two existing
event_idvalues; - create one controlled Telegram failure;
- retry only the failed step;
- perform a full rerun;
- compare final rows, alerts, counters, and run history.
Albato documents duplicate-handling options, but also notes that when an existing record is found its data can continue into later steps. Avoiding a second row therefore does not prove that a second alert was avoided. See Albato duplicate handling.
The safety condition is simple: one event ID must produce one row and one alert. If neither implementation can provide that without disproportionate complexity, the correct result is neither platform, regardless of unit cost.
Evidence required before the hands-on update
For each platform preserve the plan and counter before the test, full diagram, trigger type and interval, mappings, duplicate/retry/error configuration, empty poll, one event, a burst, final counter, run IDs and statuses, final sheet and Telegram messages. Redact webhook URLs, tokens, addresses, account details, and chat IDs. Use synthetic data only.
The minimum row-level ledger is:
platform, workflow, plan, run_id, event_id, source_time, detected_time, completed_time, status, operations, credits, transactions, output_row_id, message_id, duplicate, retry_scope, notes.
Commercial and paid-traffic boundary
Make’s current public affiliate material describes 35% commission on subscription payments for 12 months, excludes extra credits, and requires a $100 payout threshold plus three different paying customers. Its public rules permit paid promotion but restrict the Make name and marks in keywords, search ads, copy, titles, body, or metadata without written Celonis permission. Verify Make’s affiliate page before any campaign.
Albato currently describes tiered 30%/40%/50% regular-plan commission, a 20% first-payment customer discount, and a $100 payout threshold. CriteriaDesk has not found public official paid-traffic or brand-bidding rules; the signup points to an internal FirstPromoter contract. No paid Albato promotion is permitted in this program until that contract or support provides written confirmation. See Albato affiliates and the affiliate introduction.