TrustLens Segments Explained: Low, Medium, High, and When Each Matters

You open the TrustLens customer list and filter for Caution. Forty names. You click the first one: score of 34. The segment badge says Caution. But what does Caution actually mean for how you process that customer’s next order?

The segment name gives you a direction. The score gives you a position within that segment. Neither one, on its own, tells you what to do. That third layer — the operational decision — is what this post is about.

The thresholds and scoring mechanics below are verified against TrustLens version 1.2.3 source code. The default thresholds are configurable in TrustLens Settings, so if your store has adjusted them, your segment boundaries will differ from what is shown here.

Why Segments Matter More Than the Raw Score

A score of 34 and a score of 47 are both in the Caution segment. Operationally, they call for the same initial approach: review the signal breakdown before doing anything else. You would not treat a 34 differently from a 47 on first glance, any more than a doctor treats a patient whose temperature is 38.2°C differently from one at 38.8°C before checking symptoms.

Segments do something useful: they collapse the 0–100 range into six categories that correspond to recognizable operational modes. Each mode has a default posture — a starting position for how you engage with that customer. The signal breakdown behind the score tells you whether to deviate from that default.

The segments also matter because TrustLens uses them as triggers. Automation rules in TrustLens Pro fire when a customer’s segment changes, not when their raw score shifts by five points. The segment is the decision unit; the score is the evidence underneath it.

How TrustLens Assigns a Segment

TrustLens calculates a trust score for every WooCommerce customer using eight detection modules — returns, orders, coupons, categories, linked accounts, shipping anomalies, chargebacks, and card-testing defense. The engine adds the adjustments from all active modules to a baseline of 50, applies an account-age loyalty bonus of up to +15 points, and clamps the result to the 0–100 range. The wstl_get_segment_from_score() function in the plugin then maps that number to a segment name using configurable thresholds.

Two things are worth knowing before you look at a segment badge:

• The minimum-order gate. Customers with fewer than three completed orders (the default minimum) stay in the Normal segment regardless of what their signals say. TrustLens will not classify a new customer as Caution or Risk based on one or two orders because the data is not yet reliable. Signals accumulate during this period — they just do not move the segment until the threshold is met.
• Allowlisted customers always score 100. A customer you have added to the allowlist has their score locked at 100 and their segment set to VIP. No detection module can affect them. This is deliberate; allowlisting is how you protect customers you know well enough to vouch for.

If you want to understand how each detection module contributes to a specific customer’s score, the signal breakdown on their profile page shows every adjustment with a reason text. The full walk-through of all eight detection modules covers each one in detail.

The Six Segments at a Glance

TrustLens maps every score to one of six named segments. The default thresholds — verified from the plugin’s wstl_get_segment_from_score() function — are:

A score of exactly 90 is VIP. A score of 89 is Trusted. The boundary is the threshold value itself — customers at or above the threshold land in the higher segment. A score of exactly 10 is Risk; a score of 9 is Critical.

VIP (Score 90–100) — Protect and Reward

A VIP customer in TrustLens has accumulated enough positive signals to score at or above 90. In practice this means: a long clean order history, a low return rate, no coupon abuse flags, and typically the full +15 account-age loyalty bonus that kicks in at 365 days of order history. These are your most reliable buyers. The score reflects accumulated trust, not just longevity.

What the signals look like

The Order Pattern module can contribute up to +15 points for customers with ten or more clean orders (orders minus refunds). The Return Abuse module adds +10 points when a customer has five or more orders and a return rate at or below 5%. The account-age bonus adds +15 for customers who have been ordering for a year or more. A customer with all three of these conditions starts at 50 + 15 + 10 + 15 = 90 — right at the VIP boundary before any other module runs.

What to do

The most important action for VIP customers is protecting them from false positives. Add your most reliable VIPs to the allowlist. An allowlisted customer’s score is locked at 100; no signal from any detection module can affect it. This means a VIP who has a single atypical event — a product return that is genuinely defective, a billing address change mid-move — will not slip into Caution because of one anomaly.

Beyond protection: VIP customers are the ones most worth rewarding. If you run loyalty programs, early-access sales, or personalized pricing, this is the segment to target. TrustLens identifies them; what you do with that identification is your call.

Trusted (Score 70–89) — Standard Processing, No Action

Trusted customers have reliable behavioral records. They have not yet accumulated the score needed for VIP status — typically because they are newer customers who have not yet earned the full account-age bonus, or because their order volume is smaller than your long-standing VIPs. But their signals are clean.

What to do

For most Trusted customers, the right action is no action at all. They are behaving well. The segment confirms that fact and lets you get on with running the store.

The thing worth watching: Trusted customers who start accumulating return signals or coupon-then-refund patterns may drift toward Normal or Caution. This is the system working correctly. A customer who moves from Trusted to Caution has had something change in their behavior, and that change is worth noting when their next order comes in.

If a Trusted customer contacts you about a dispute or an unusual order, their profile gives you quick context. Look at the event timeline to see whether the current situation is consistent with their history. A customer with 40 clean orders and one dispute is in a very different position from someone whose Trusted status came from eight orders and nothing unusual — yet.

Normal (Score 50–69) — Watch the Trend, Not the Score

Normal is the segment that holds two distinct customer types, and understanding which type you are looking at is more useful than the segment name itself.

New customers in Normal

Customers below the minimum-order threshold (default: three orders) stay in Normal regardless of what their signals say. A customer with one order and a Normal badge means “not enough data yet” — not “clean history.” This is intentional. TrustLens does not want to flag new customers as risky based on insufficient evidence.

Established customers in Normal

A customer with 20 orders and a score of 58 is in Normal for a different reason: they have mixed signals that net out to the middle range. Some moderate penalty signals being offset by account age and clean order volume, for example, produces a mid-range score. Open the signal breakdown to see what is going on.

What to do

For established Normal customers, the main thing to watch is direction of movement. A customer who has been in Normal for two years with a stable score has genuinely average behavior. A customer whose score has dropped from 65 to 52 over the past six months is trending toward Caution. The trust score trend chart on the TrustLens dashboard shows store-wide movement; the event timeline on a specific profile shows it for that individual customer.

No enforcement action is warranted for the Normal segment. Monitoring is.

Caution (Score 30–49) — Read the Signals Before You Act

Caution is the segment where early intervention is most valuable — and where the most mistakes happen. It is easy to over-react (blocking customers who have elevated signals for entirely legitimate reasons) or under-react (ignoring signals that are about to get worse).

A score in the 30–49 range means the detection modules have registered meaningful risk signals, but not at a level that indicates consistent, deliberate abuse. Common combinations that produce a Caution score:

• A return rate between 25% and 40% — elevated but not in the critical range. The Return Abuse module applies -10 points for return rates at or above 25% (verified from class-module-returns.php).
• One or two coupon-then-refund events. The Coupon Abuse module applies -5 points for the first occurrence, -15 for two, -25 for three or more.
• A link to one other customer account at a slightly lower score.
• Early-stage cancellation patterns that reduce the order quality score.
• An account-age bonus that is partial but not yet full, producing a score that cannot compensate for moderate penalties.

What to do

Open the profile and read the signal breakdown. Then look at the event timeline. Ask yourself two questions before doing anything else:

1. Is this a pattern or a period? A run of returns in a single quarter could reflect a supplier quality issue or a policy change — not abuse. A consistent return rate across 18 months is a different situation entirely.
2. Is this trending up or down? A Caution customer whose score has been stable at 42 for a year is a very different concern from one whose score has dropped from 64 to 42 in the past 90 days.

For most Caution customers, the right response is monitoring rather than enforcement. Note the customer, check their profile monthly, and watch whether the trajectory continues. Blocking a Caution customer outright is usually premature — false positives in this range are common, and the cost of blocking a legitimate customer is real.

When some level of intervention feels warranted, a graduated response fits better than a block: hold orders from this customer for manual review before fulfillment, or — with TrustLens Pro — configure an automation rule to alert you when their next order comes in so you can decide in real time.

Risk (Score 10–29) — A Policy Decision Is Overdue

Risk means the detection modules have found multiple signals across different dimensions. A single elevated return rate can produce a Caution score. A Risk score typically reflects a combination: elevated return rate plus coupon-then-refund patterns, or shipping anomalies plus linked accounts flagged at higher risk, or chargebacks alongside a high cancellation rate.

The signals are not ambiguous at this level. Something is wrong with this customer relationship — either they are abusing your policies deliberately, or your policies are misconfigured for their purchasing pattern in ways that look like abuse. The profile will tell you which.

What to do

Open the profile and look at the signal breakdown honestly. If the signals reflect a genuine purchasing pattern problem (returns in a category where your return policy is too generous, for example), the right response might be a policy adjustment — not a customer block. If the signals reflect deliberate abuse (coupon-then-refund on multiple accounts linked by address, combined with chargebacks), a block is warranted.

In the free version of TrustLens, blocking is a manual action: you click Block Customer on the profile page. The block takes effect immediately on both Classic and Block checkout. If you want this to happen automatically when a customer enters Risk, TrustLens Pro automation rules let you configure that — with conditions so you do not accidentally block customers whose Risk score is driven entirely by one large legitimate dispute.

Even if you decide not to block immediately, a Risk customer warrants a clear decision: document it in the admin notes field on their profile. If someone on your team processes an order from this customer later, they need to see that a risk decision was already made — and why.

Critical (Score 0–9) — Every Order Costs You Something

Critical represents the bottom of the score range: customers where the accumulated signals across multiple detection modules are severe and consistent. A score of 0–9 requires multiple penalty signals stacking on top of each other — it is not produced by a single elevated metric.

What produces a Critical score in practice: a very high return rate (at or above the configurable critical threshold, default 60%, applying -40 points from the Return Abuse module) combined with full-refund wardrobing penalties, chargeback history, coupon-then-refund patterns, and links to other high-risk accounts. The account-age bonus cannot compensate for this level of accumulated penalty.

What to do

At Critical, the decision is usually binary: block the customer, or document your cost exposure and continue serving them with full awareness. There are legitimate reasons to choose the second option — a long-standing business account whose returns reflect genuine product issues, for example — but that decision should be intentional and documented, not a passive default.

If you have chargebacks associated with this customer and are building a dispute response for your payment processor, TrustLens Pro’s Dispute Evidence Report generates a print-ready behavioral risk summary — trust score, signal history, return analysis against your store average, linked accounts, and the full event timeline — in a format suitable for processor submissions. The detailed guide to TrustLens Dispute Evidence Reports covers what the report includes and how to use it.

For Critical customers in the free version: block them from the profile page. Blocking prevents them from adding items to cart or completing checkout — it applies to both Classic and Block checkout. The block is logged and can be reversed at any time.

What Actually Moves a Score Between Segments

Understanding which signals produce which score adjustments helps you read a profile more quickly. Here are the verified penalty and bonus amounts from the TrustLens source code, for the most impactful signals:

All adjustments are additive. The total adjustment is added to the base score of 50 and clamped to the 0–100 range. Category-aware scoring, linked accounts, shipping anomalies, chargebacks, and card-testing signals also contribute — their specific adjustments depend on configuration and the severity of the detected pattern.

A Graduated Response Framework Across All Six Tiers

Rather than treating every segment as a binary “act or don’t act” decision, a graduated response matches the level of intervention to the level of evidence. Here is a practical framework that fits most WooCommerce stores:

Tier 1: VIP and Trusted — frictionless processing

Orders from VIP and Trusted customers go through standard fulfillment. No holds, no manual checks, no friction. If you use TrustLens Pro automation rules, configure them to skip this tier entirely. The only workflow step is occasionally reviewing whether to move a VIP to the allowlist.

Tier 2: Normal — standard processing with trend awareness

Normal customers also process normally. The additional step: pay attention to segment-change notifications if you have them configured. A Normal customer whose score drops to Caution is a signal — not an emergency, but worth a look at their profile.

Tier 3: Caution — review-before-fulfillment for high-value orders

For Caution customers, consider a threshold-based review policy: orders above a certain value (say, $150 or $300 depending on your AOV) get a brief profile check before fulfillment. Below that threshold, process normally and watch the trend. The goal is not to create friction for every Caution order — it is to catch the few that are genuinely problematic without annoying customers who are simply in a risk range due to coincidence.

Tier 4: Risk — deliberate decision required before each order

For Risk customers, the default posture is a hold-and-review before fulfillment. This does not mean blocking — it means that before you ship, someone looks at the order and the profile. TrustLens Pro can automate this by sending an alert when a Risk customer places an order; the free version requires manual monitoring of the customer list.

Tier 5: Critical — block or explicit cost acceptance

For Critical customers, process no order without a deliberate decision. Either block the customer now, or explicitly decide to accept the exposure and document why. There is no passive option at Critical.

Configuring Thresholds for Your Store

The default segment thresholds — 90 for VIP, 70 for Trusted, 50 for Normal, 30 for Caution, 10 for Risk — suit most stores on first install. But they are configurable in TrustLens Settings, and some stores have good reasons to adjust them.

A store with generous return policies and a high-volume customer base might want to raise the Caution threshold to 40 — treating the 40–49 range as a watch zone rather than a review zone — because a return rate of 30% is normal for their product category. A store that has been burned by repeat fraud might lower the Risk threshold to 15, expanding the Risk tier to catch more edge-case patterns.

Two things to keep in mind when adjusting thresholds:

1. Lowering a threshold does not change how scores are calculated. It changes which bucket a score falls into. A customer at 28 is still at 28 regardless of whether your Risk threshold is 10 or 20. The threshold only changes the label and the automation trigger — not the underlying evidence.
2. Threshold changes affect all customers immediately. If you raise the Trusted threshold from 70 to 75, customers who were previously Trusted at 72 will drop to Normal at the next score recalculation. Review your customer list after any threshold change.

If you have adjusted your thresholds from the defaults, the score ranges in this post will not match your store’s segment boundaries. Always check TrustLens Settings → General for your current thresholds before interpreting segment data.

For a deeper look at the mechanics behind how TrustLens builds a customer profile — including how each of the eight detection modules actually calculates its adjustment — the guide to reading a TrustLens customer profile walks through every section in detail.

Frequently Asked Questions

Can a customer move from Critical back to Normal?

Yes. Trust scores are recalculated dynamically whenever a relevant event occurs. A customer who stopped returning merchandise, has not filed any new disputes, and has accumulated additional clean orders over the following months will see their score improve. Whether they leave Critical depends on how many positive signals accumulate against the remaining penalty signals. The account-age bonus continues to grow, which helps. But a customer with a -40 penalty from a very high return rate will not recover to Normal through account age alone — the underlying behavior needs to change.

Does blocking a customer erase their history?

No. Blocking a customer prevents them from completing checkout. Their trust score, segment, signal history, and event timeline remain intact. If you unblock them later — because a dispute was resolved, or because new evidence changed your assessment — their history is still there. You can also add a note to their profile explaining the decision and the conditions under which you would reconsider.

What happens to guest customers who later create an account?

TrustLens identifies customers by a keyed HMAC-SHA256 hash of their email address. A guest who checks out with the same email address as a registered customer maps to the same hash, and their order history is combined. If a guest later creates an account using the same email, their existing trust profile carries over automatically.

Is the Caution segment the right place to apply a first-order discount restriction?

Not automatically. First-order discount abuse is tracked separately by the Coupon Abuse module, which looks at whether a customer has used first-order coupons and then refunded. A Caution customer whose score is driven entirely by a moderate return rate — with no coupon signals — is not a coupon abuser. If you want to restrict first-order discounts based on trust level, use TrustLens Pro automation rules to apply that condition specifically to customers with nonzero first-order coupon refund counts, rather than blocking discounts for the entire Caution segment.

How quickly does a score update after a new order or refund?

TrustLens queues a score recalculation via Action Scheduler whenever a relevant event occurs — an order completes, a refund is processed, a dispute is recorded. Action Scheduler processes these jobs asynchronously in the background, typically within a minute or two on healthy sites. The score you see on a profile is current as of the last recalculation event. You can trigger a manual recalculation from the profile page at any time.