Account health on Shopee, Lazada and TikTok Shop, side by side

Three years into running stores across Shopee, Lazada and TikTok Shop, the thing that still catches sellers off guard isn’t the fees or the commission tiers, it’s account health. You can have healthy margins and a five-star rating and still wake up to a store flagged for review because a courier missed three pickups in a row, or because one listing photo tripped an IP infringement filter you never knew existed.

I’ve watched more than one seller’s Shopee badge drop from Preferred Seller to plain Standard inside two weeks, not from anything dramatic, just a run of late scans during a slow week at the warehouse. On TikTok Shop, a single upheld counterfeit strike can freeze your ability to go live or add new SKUs while you’re mid-appeal. Lazada will quietly throttle your search visibility before it ever sends you a warning that reads like a warning.

The three platforms measure different things, use different point systems, and forgive mistakes on different timelines. None of them explain this well in one place. This is my attempt to put Shopee’s Penalty Points, Lazada’s seller performance metrics and TikTok Shop’s Account Health score next to each other, with what actually moves the needle and what to do once you’re flagged.

background and prior art

Marketplace account health scoring isn’t new. Amazon has run a version of this since around 2016 with its Order Defect Rate and Account Health Rating, and the Southeast Asian platforms built their own systems on similar logic once seller counts got large enough that manual review stopped scaling. Shopee’s Penalty Point system rolled out market by market from around 2019 onward. Lazada’s performance metrics dashboard followed a similar path inside Seller Center. TikTok Shop is the newest of the three here, its Southeast Asia marketplace only properly launched in 2022, and its Account Health tab was built from day one as an explicit, named feature rather than bolted onto an older dashboard the way some of Shopee’s and Lazada’s tooling feels.

seasellerdesk isn’t affiliated with Shopee, Lazada or TikTok Shop. Everything below comes from reading each platform’s public seller centre documentation and from running stores through all three ourselves. Where a number matters, I’ve said where to go check it live, because these thresholds get revised by the platforms without much notice and often differ market to market.

the core mechanism

All three systems run on the same basic logic even though the branding differs: track a handful of fulfilment and compliance metrics over a rolling window, convert breaches into points or a score, and escalate consequences as the number gets worse. Where they diverge is which metrics count, how forgiving the rolling window is, and what actually happens once you cross a line.

shopee: penalty points

Shopee’s system lives in Seller Centre under Penalty Points. It tracks late shipment rate (orders shipped after the promised ship-by date, as a percentage of total orders in the period), non-fulfilment rate (cancelled or undelivered orders), and listing or product violations such as prohibited items, misleading claims, or IP infringement flags. Violations convert into points on a scale that scales with severity, a late shipment run costs a handful of points, a confirmed counterfeit listing costs a lot more, and the points sit on a rolling basis, usually somewhere between 30 and 90 days depending on the violation type.

Cross a points threshold and Shopee starts limiting things: fewer daily listing slots, suppressed search ranking, and eventually a full account suspension if the total keeps climbing. On the other end, Shopee also runs a positive tier, Preferred Seller and Star Seller badges, that reward a low cancellation rate, fast chat response and consistent shipping. Losing that badge is often the first visible sign something’s wrong, well before penalty points show up on the dashboard. It matters beyond visibility too, badge status can affect which commission tier you sit in, which is where this overlaps with the money side of things I covered in Shopee, Lazada and TikTok Shop seller fees compared.

I’m deliberately not quoting exact point values or cutoffs here. Shopee has run different tables in the Philippines and Indonesia than in Singapore in past updates, and they get revised, sometimes seasonally around 11.11 and 12.12 when fulfilment load spikes across the board. The only number worth planning around is the one on your own Seller Centre > Penalty Points page, checked on the day you’re reading this, as of September 2026.

Filing a dispute is its own small ritual, and it’s worth knowing before you need it. Attach your proof, submit through Penalty Points, then wait. Ours have come back anywhere from three days to just under two weeks, depending on how backed up the review queue is that month. One dispute got reversed in four days flat; a near-identical one sat untouched for eleven. I never found a pattern that explains the gap.

lazada: seller performance metrics

Lazada’s version sits inside Seller Center under a performance dashboard, tracking a similar set: late shipment rate, failed delivery or rejection rate, cancellation rate, and chat response rate. Instead of a single points ledger, Lazada tends to bucket stores into a small number of standing tiers, something like preferred, standard and at-risk, based on whether your rolling metrics sit inside or outside its target bands. LazMall, Lazada’s official-brand storefront format, runs a stricter version of the same bands than the open marketplace does.

The practical difference from Shopee is speed of feedback. Lazada’s dashboard updates can lag a day or two behind what actually happened, so a bad week doesn’t always show up as a status change until you’re already partway into the next one. That lag has bitten me before. We fixed a shipping bottleneck on a Tuesday and the dashboard didn’t reflect it as recovering until the following Monday, by which point we’d already spent a few anxious days assuming the fix hadn’t worked.

These days we track our own courier pickup logs instead of waiting on the dashboard, since those update same-day. It’s an extra spreadsheet nobody asked us to keep, but it’s caught more than one false alarm before it ate an afternoon.

tiktok shop: account health

TikTok Shop calls its version exactly that, Account Health, and it’s the most visually explicit of the three: a single score or risk banding built from two buckets, quality metrics (fulfilment rate, cancellation rate, return and refund rate, late shipment rate) and compliance violations (listing rule breaks, IP infringement, restricted goods). Each violation carries a point deduction and a validity period before it ages off. Cross into a worse banding and TikTok Shop starts removing functionality: no live selling, no new product uploads, reduced traffic to your storefront, and at the bottom, suspension.

TikTok Shop’s Southeast Asia footprint is also the most market-dependent of the three right now. Indonesia’s operation runs through PT Tokopedia, after TikTok invested roughly $1.5 billion into GoTo’s e-commerce arm in December 2023, following Indonesia’s government restricting direct social-commerce selling under a 2023 trade regulation (not legal advice, just recent history worth knowing if you sell into Indonesia). What that means practically is that account health mechanics for an Indonesian seller may not map one to one onto what a Singapore or Malaysia seller sees in their own dashboard. If Indonesia is one of your markets, check your own Seller Centre rather than assume the Singapore rules carry over.

Of the three, I trust TikTok Shop’s score the least, not because it’s badly built, but because it moves the fastest and forgives the slowest. A single bad batch of stock can tank two weeks of otherwise clean metrics before the rolling window lets it go.

The metric every seller asks about first is late shipment rate, and it’s calculated the same basic way across all three platforms even where the label differs:

late shipment rate = (orders shipped after the promised ship-by date) / (total orders in the rolling period) Ã, 100

What changes between platforms is the rolling period length, the promised ship-by cutoff (itself a function of your listed processing time), and what counts as “shipped.” A printed label is not a shipped order on any of these three. It’s the courier’s pickup scan that stops the clock.

worked examples

These are illustrative, built from the mechanics above rather than a real seller’s live numbers, since current thresholds are only worth trusting from your own dashboard.

A homeware store on Shopee Philippines running 300 orders a month: if 18 of those ship late inside a 30-day window, that’s a 6% late shipment rate. Whether that trips a warning or a points deduction depends on where Shopee’s current cutoff sits for that market, but the mechanics matter more than the exact number. That 6% is calculated against total orders in the window, not just the ones a customer complained about, so a handful of quiet late scans during a busy week move the needle just as much as one angry review does.

A beauty brand on TikTok Shop Singapore sitting at a 2.5% return and refund rate, comfortably inside a healthy band, ships one bad batch of a poor-fitting product line and returns jump to 6% for two weeks. Because Account Health blends multiple quality metrics into a single score, a spike in just one of them, even briefly, can knock the overall status down a tier. Recovery isn’t instant either, it tracks the rolling window, not the day you fix the underlying problem.

A fashion accessories seller running the same catalogue on both Shopee Malaysia and Lazada Malaysia will often see the two platforms disagree about how healthy the store is in the same week, because Lazada’s dashboard lag and Shopee’s closer-to-real-time points ledger respond to the same underlying courier delay on different timelines. That’s not a bug in either system. It’s two different measurement windows looking at the same warehouse problem.

edge cases and failure modes

A handful of pitfalls show up over and over running stores across all three platforms.

opening a second store to dodge a bad rating: all three platforms link stores to the same business registration, bank account and device fingerprints, and treat an unauthorised duplicate store as a violation in its own right, on top of whatever got the first store into trouble in the first place. Shopee, Lazada and TikTok Shop each publish rules on how many stores you can legitimately run and under what conditions, I went through those market by market in running multiple Shopee and Lazada stores, the actual rules. Short version: don’t try to outrun a penalty by spinning up a new shopfront, it tends to end with both stores gone.
seasonal load spikes: 11.11, 12.12 and similar campaign days push order volume up faster than most warehouses can scale pickers, and a late shipment rate that looked fine in October can spike hard during a single campaign week. Build slack into your processing time setting before the campaign, not during it, since editing it mid-campaign doesn’t retroactively fix orders already in the window.
one bad SKU dragging the whole store down: because TikTok Shop and Shopee both compute rates against total store orders, a single problem product with high returns or frequent stockouts can pull your whole account’s metrics down even if every other listing is clean. Pull the SKU or fix its listed processing time the moment you notice, don’t wait for the rolling window to age it out on its own.
appealing a wrongful IP or counterfeit flag: all three have a dispute path (Shopee: Penalty Points > dispute, Lazada: a case through Seller Center support, TikTok Shop: an appeal button under the specific Account Health violation), and all three want documentary proof, invoices, brand authorisation letters, trademark registration numbers. Keep that paperwork on hand before you list a brand’s products, not after a takedown notice lands.
ignoring seller centre banners and in-app notices: Shopee, Lazada and TikTok Shop all push policy changes through a Seller Centre banner or in-app notification before the change takes effect, not through email, and it’s easy to swipe past those the same way you’d swipe past a cookie prompt. A change to IP enforcement or to what counts as a late shipment can sit there for weeks before it shows up in your numbers, and by the time it hits your metrics the grace period is already gone.
managing seller centre logins while travelling or running things remotely: a login from an IP address the platform doesn’t associate with your usual account location is exactly the kind of signal that triggers a security or device review. That’s a separate system from account health scoring, but a frozen login during a review still means you can’t act on orders, and that does hit your fulfilment metrics. If your operation is genuinely based in Singapore but you or a staffer sometimes log in from elsewhere, Singapore Mobile Proxy sells real Singapore mobile IPs with sticky sessions built for keeping that login consistent. It’s Singapore-only though, so it’s no help if your warehouse is actually in Jakarta or Manila and you’re just looking for a shortcut.

what we learned in production

The single biggest lever across all three platforms is processing time, the number you set for how long you need before a courier picks up. Padding it by even a day buys real room against late shipment rate without touching anything else in the business, and it’s the fix most sellers reach for last, usually after they’ve already eaten a few penalty points.

The second thing is more of a confession than a lesson. We once assumed a printed shipping label counted as “shipped” for TikTok Shop’s clock. It doesn’t, the courier’s pickup scan does, and it cost us two avoidable violations before someone on the team actually read the fine print in Seller Centre. That one stung. If chat response and shipping compliance are split across multiple staff logging in from different phones, that setup is worth doing properly rather than improvising, I wrote up how we split that across devices in running seller apps across multiple phones. And if you’re pricing anywhere near a category band that might trigger a manual quality review, we track that separately too, covered in tracking competitor prices on Shopee and Lazada. None of this replaces reading your own dashboard, which changes more often than any article can keep up with, including this one.

references and further reading

Shopee Seller Centre (Singapore) - the live source for Singapore penalty points, shipping settings and Preferred/Star Seller criteria.
Lazada Seller Center (Singapore) - performance metrics dashboard and dispute submission.
TikTok Shop Seller Center (Singapore) - Account Health tab, violation history and appeals.
Amazon Seller Central - Amazon’s own account health system, the older model the Southeast Asian platforms broadly followed once seller volume outgrew manual review, useful for direct comparison.

For the full list of the site’s deep-dives on running stores across these three platforms, start at /blog/.

Written by Xavier Fok

disclosure: this article may contain affiliate links. if you buy through them we may earn a commission at no extra cost to you. verdicts are independent of payouts. last reviewed by Xavier Fok on 2026-09-12.