How to test an SMM panel: 12-step protocol with pass/fail thresholds

How to test an SMM panel: the short answer

To test an SMM panel, place 3 minimum orders on a throwaway account, log delivery start time, count delivered units at 24 hours, then recount at day 7 and day 30 to calculate your drop rate. Everything below makes that sentence replicable and defensible.

Most guides tell you what to look for. This one tells you the exact steps to run, in order, with defined pass/fail thresholds at each gate. If a panel fails any gate, you stop spending. If it clears all twelve, you have a tested baseline — not a guarantee, but a baseline.

One structural fact before you start: Instagram's Platform Policy explicitly states that accounts may be disabled for purchasing followers or engagement (Instagram Help Center, Platform Policy). That risk is present on every order, at every panel, regardless of how well the panel performs on the tests below. Run every test on accounts you can afford to lose.

Should you use a fake account to test an SMM panel?

Yes, always. Create a throwaway account on the target platform — a fresh profile with a few posts, a real-looking username, and no connection to any client or personal account. This is non-negotiable. If the delivery pattern triggers a platform review, the throwaway absorbs it. If the panel asks for your account password at any point, stop immediately and do not proceed. No legitimate SMM panel requires login credentials. Panels operate by targeting a public profile URL or username, never by logging in as you.

Set up one throwaway per platform you intend to test. A TikTok throwaway for TikTok views, a YouTube throwaway for YouTube watch hours, a Telegram throwaway channel for Telegram channel members. Never cross-contaminate a test.

Step 1 — Vet the panel before depositing anything

Before you fund a balance, run these four checks:

1. Search the panel's domain on BlackHatWorld and Reddit. Look for threads where buyers report actual delivery outcomes — not reviews on the panel's own site. Independent forum complaints about order status freezing, partial orders never resolving, or refund requests ignored are the most useful signals you will find without spending money.
2. Check Trustpilot. Low review counts are a yellow flag, not a red one — newer panels have thin records. Patterns of identical 5-star reviews posted in short windows, or reviewer profiles with no other history, are red flags.
3. Open a support ticket with a real question before depositing. Ask something specific: "Do you offer a refill guarantee on Instagram follower service ID [X]?" Log the response time and the quality of the answer. A panel that takes four days to answer a pre-sales question will take longer when your order stalls.
4. Confirm the panel does not ask for passwords. Check the order form for any field requesting account login credentials.

Pass threshold: Support responds within 24 hours with a substantive answer. No password field exists on any order form.

Step 2 — Identify the exact service IDs you will test

This is the step every incumbent skips, and it is the one that costs resellers the most money.

A panel is not a monolith. Service ID #203 and service ID #204 on the same panel can draw from different upstream suppliers with different retention curves. When you read a review that says "this panel has 15% drops," you do not know which service ID was tested. That number is useless to you unless you know the service ID.

For each platform and metric you plan to resell, identify the two or three service IDs the panel offers for that category — for example, "HQ followers," "Real followers," and "Premium followers" for Instagram. You will test each one separately.

What is the difference between HQ and Real followers on an SMM panel?

Panels use these labels inconsistently, and there is no industry-wide standard definition. In practice, "HQ" (high-quality) typically refers to accounts that have profile pictures and some post history — they look less obviously automated than blank bot accounts. "Real followers" is a marketing claim, not a verified category; no panel can guarantee the source accounts are operated by real humans. "Premium" usually signals a higher price tier with the same upstream ambiguity.

The only way to know whether the price premium on a higher tier is justified is to test each service ID separately and compare 30-day retention. Do not assume the premium tier outperforms the standard tier until your own data shows it.

Steps 3–7 — Place and monitor the test orders

Step 3: Fund a minimal balance. Deposit the smallest amount the panel allows. You are not committing client budget yet.

Step 4: Place 3 identical orders, not 1. Order the same service ID, the same quantity (use the panel's minimum), on the same throwaway account, spaced 24–48 hours apart. One order tells you what happened once. Three orders tell you whether the panel is consistent. Variance across three identical orders is a direct signal about supplier-chain stability.

Step 5: Log delivery start time. Note the exact time you placed the order and the exact time the follower or view count on the throwaway account began moving. This is your start-to-delivery latency. Cross-reference it against the panel's claimed delivery window in the service description.

Also check the panel's API order status update cadence. A panel can claim "1–6 hour delivery" but only push order status updates once every 24 hours. If you are running an API-connected reseller panel, that cadence gap means your downstream customers see stale status — a support cost that compounds at scale.

Step 6: Count delivered units at 24 hours. Compare what was delivered against what was ordered. A partial order — an order that stops before reaching the ordered quantity and enters "partial" status — should trigger a refund or a completion, depending on the panel's stated policy. Test whether that policy executes automatically or requires a manual ticket.

Pass threshold: At least 80% of ordered units delivered within the panel's stated delivery window across all 3 orders.

Step 7: Enable drip feed if available, on one of the three orders. Drip feed spreads delivery over hours or days rather than delivering in a single spike. Test it because it changes the delivery pattern your downstream clients see, and because platform-side anomaly detection is more sensitive to sudden spikes than to gradual growth. Compare the 24-hour count on the drip-feed order against the standard-delivery orders.

Steps 8–10 — Measure quality at the follower level

Step 8: Inspect delivered accounts. On the throwaway, open the follower list. Spot-check 20–30 of the new followers. Look for: profile pictures present, some post history, non-zero following counts, bios that are not blank. Bot accounts typically cluster on one or more of: no profile photo, zero posts, following-to-follower ratios that are extreme in either direction, usernames that are random alphanumeric strings.

This is a qualitative check, not a pass/fail count. You are looking for the distribution — if 90% of the spot-checked accounts are blank shells, the service tier is not what the label implied.

Step 9: Recount at day 7. Record the follower or view count on day 7. Calculate drop rate: (units delivered at 24h − units present at day 7) ÷ units delivered at 24h. This is your 7-day drop rate.

Pass threshold: 7-day drop rate below 15%. This is the threshold cited by practitioners in the reseller community; we note that no peer-reviewed source benchmarks this figure, and your acceptable threshold may differ by service type and client expectation.

Step 10: Recount at day 30. Calculate your 30-day drop rate using the same formula. This is the number that determines your real cost per 1,000 retained — a metric we analyzed in depth in our piece on real cost per 1k retained, where sticker price and 30-day net cost diverged by multiples across tested panels.

Pass threshold: 30-day drop rate below 20% on the service ID you plan to resell.

Steps 11–12 — Test the safety net

Step 11: Trigger the refill guarantee. If your order drops beyond the threshold the panel advertises in its refill policy, submit a refill request. Time how long the refill takes to execute. A refill guarantee that takes 10 business days to process is not a functional safety net for a reseller with client SLAs.

Pass threshold: Refill initiated within the panel's stated window, with a confirmable order status update.

Step 12: Test the API if you plan to integrate. Panels including JAP (JustAnotherPanel), Peakerr, SMMRaja, and GetMyLikes publish API documentation. Pull the service list endpoint, place a test order via API, and poll the order status endpoint. Confirm that the status codes match what their documentation describes. An undocumented status code on a real order is a support ticket waiting to happen at scale.

How many test orders should I place before trusting a panel with client accounts?

Three minimum orders per service ID you plan to sell, measured through 30 days, before any client account is touched. That is the floor. If you plan to resell multiple service IDs — say, Instagram followers and TikTok views — that is three orders per service ID, not three orders total. Shortcutting the count because early results look good is the single most common reason resellers absorb drop-rate losses they did not budget for.

If the data from your three orders shows high variance — one order delivered fast, one took twice as long, one partially stalled — you do not have a passing panel. You have an inconsistent one. Inconsistency in a supplier chain compounds when you scale.

A note on what this protocol cannot tell you

This playbook measures a panel's performance at a point in time, on a specific service ID, under current platform conditions. Upstream supplier chains change. Platform algorithm updates affect delivery patterns for all panels simultaneously — so a result you logged in one month may not hold the next. Re-run the minimum test on any service ID you have not ordered in 60 days before committing fresh client budget to it. Data on this is thin across the industry; no published study tracks panel performance degradation over time with controlled methodology. That gap is real, and any guide that does not acknowledge it is overstating its own certainty.