How to Read WooCommerce Trust Score Trends in TrustLens
Plugin Guide · TrustLens Analytics
How to Read WooCommerce Trust Score Trends in TrustLens
The trend chart on your TrustLens dashboard shows 30 days of your store’s average customer trust score. A rising line is not always good news. A falling line is not always a crisis. Here is how to read what it is actually telling you.
The Trust Score Trends chart sits inside the Trust Index panel on your TrustLens Command Center dashboard. It is a 30-day line chart of your store’s average customer trust score — a single daily number that compresses what happened across every customer interaction into one point on a chart.
That compression is useful. It lets you see whether your store’s overall customer health is improving, deteriorating, or holding steady. But a single average hides a lot of detail, and that detail matters when you are deciding whether to act. This guide walks through how to read each pattern the chart produces, what causes each one, and what you should actually do about it.
What the Trust Score Trends Chart Actually Shows
The Trust Score Trends chart in TrustLens displays the daily average trust score across all scored customers in your store, over the last 30 days. Each point on the chart represents one calendar day. The value shown is the mean trust score of every customer whose score was recalculated that day — on a 0-to-100 scale — along with how many distinct customers contributed to that day’s average.
To be precise about what “trust score” means here: TrustLens assigns every customer an integer score from 0 to 100. Higher means more trustworthy. Scores are calculated by eight detection modules running in the background, each contributing a positive or negative signal. The chart shows the average of those final per-customer scores across your whole customer base, day by day.
A store-level average, not individual scores
The trend chart aggregates all customers who were scored on a given day. It does not show any individual customer’s score history over time — that is visible on each customer’s own profile page. The trend chart answers the question “how is the overall health of my customer base changing?” not “is this particular customer getting better or worse?”
The chart needs at least two data points before it renders. On a fresh installation, it will be blank until at least two days of scoring history exist. If you have just run Historical Sync, the chart should populate quickly — historical orders generate scoring data that backfills the recent trend window.
The 30-day window is fixed. The chart shows the most recent 30 calendar days, ending today. There is no date range selector on the free dashboard trend chart.
How Daily Snapshots Are Built and Stored
Understanding how TrustLens stores the data behind the trend chart will help you interpret it accurately — especially on stores where scoring happens frequently.
Every time TrustLens recalculates a customer’s score (triggered by a new order, a refund, a coupon use, a dispute, or any event that might shift their signals), it writes a daily snapshot record into the database. This snapshot contains the customer’s final score for that recalculation and is tagged with the current date.
If a customer’s score is recalculated multiple times in a single day — because they placed several orders in the same afternoon, say — only the most recent calculation for that day is kept. The snapshot from earlier in the day is replaced. But snapshots from previous days are preserved as-is. Tomorrow’s recalculation does not overwrite yesterday’s.
Version 1.3.8 fixed the trend chart
Before version 1.3.8, a bug caused each score recalculation to overwrite all of a customer’s prior snapshots — not just today’s. On active stores, this meant the chart could collapse to a single point because only one snapshot survived per customer. The 1.3.8 update corrected this: previous days’ snapshots are now preserved, and the chart shows genuine day-by-day history. If you are running an older version, this is a good reason to update.
Snapshots accumulate over time. TrustLens automatically trims old snapshot records beyond roughly 400 days in a background maintenance task, so the database does not grow unboundedly. The trend chart only ever looks back 30 days, so this pruning never affects what you see on screen.
One practical consequence: score updates do not happen in real time as you watch the dashboard. Score recalculations run asynchronously via Action Scheduler (the same background system WooCommerce uses for its own tasks). There is typically a short delay between an event occurring and the score being updated. This is by design — it keeps checkout and the frontend fast while scoring happens in the background.
Reading a Rising Average
A rising trend in the Trust Score Trends chart means the daily average trust score across your scored customers has increased over the period you are looking at. That is generally a healthy signal, but it is worth knowing what can drive it.
What genuinely positive causes look like
The most meaningful rise is one that reflects your actual customer base improving in quality. If you have recently run Historical Sync on a store with several years of order history, scores will have been built from real behavioral data — loyal repeat buyers who have never filed a dispute will have high scores, which pulls the average up. A rising average in the weeks after a Historical Sync often just reflects the dataset becoming more complete.
A rising average can also follow a shift in your marketing or acquisition channels. If your last promotion attracted a different profile of buyer than usual — one that converts well and does not return — the new customers will score higher as they accumulate order history, nudging the average upward over weeks.
Rising averages that need caution
A rising average that is not accompanied by a rising customer count can mean something different: you may simply have fewer customers being scored on some days. If a quiet period means only your best customers are placing orders, the average can rise even if your overall customer quality has not changed. The chart shows both the average score and the number of distinct customers who contributed to each point — look at both together.
A rising average can also follow a wave of blocking. If TrustLens (or you manually) blocked several Critical or Risk customers in the past few weeks, those customers are no longer generating new orders, so they contribute less to recent score calculations. The bottom drops out of your distribution, pulling the average up. That is not bad news — but it is different from the average rising because new customers are genuinely trustworthy.
Reading a Falling Average
A falling trend is the one that tends to get store owners’ attention. Before you take action, it is worth diagnosing what is actually behind the decline, because the correct response varies significantly depending on the cause.
New customers diluting the average
The most common cause of a falling average on a growing store is also the most benign: new customers. When someone places their first order, TrustLens does not have much behavioral history to work with. New customers tend to score in the Normal range (around 50) until they accumulate enough orders to establish a pattern. If you have been running a promotion or paid campaign and a surge of new buyers came through, the average can fall simply because a high proportion of your recent scoring activity consists of people TrustLens does not know yet.
In this case, the right response is usually to wait and watch. As those customers place additional orders, their scores will differentiate. Loyal buyers drift toward Trusted and VIP. Anyone who starts filing disputes or returning excessively will move toward Caution or Risk. The distribution will sort itself out.
Genuine risk accumulation
A falling average driven by genuine risk accumulation looks different. Instead of the customer count rising alongside the falling average (the dilution pattern), the average falls while the customer count stays roughly flat. This suggests existing customers are being rescored downward — they are doing things that TrustLens is penalizing: filing chargebacks, returning at increasing rates, using coupons repeatedly, or triggering linked-account flags.
When a falling average warrants immediate attention
A sustained fall of more than 5–10 points over two weeks, not explained by a known influx of new customers, is worth investigating. Open the Customer list, sort by segment, and look at how many customers have moved into Risk or Critical recently. If the number is growing, look at their profiles for the signal that is driving the change — chargeback history, return rates, and linked accounts are the most common culprits.
This is where TrustLens earns its keep as an early-warning system. A chargeback that arrives today was typically placed 30–90 days ago. The behavioral signals that preceded it — an unusually high return rate, a dispute filed at another address in the same linked-account group, an order pattern that does not match a genuine repeat buyer — will often have pushed the customer’s score down before the dispute was filed. A falling store average can sometimes be the first indicator that a chargeback wave is building. If you want to go deeper on the relationship between trust scoring and chargeback prevention, the chargeback prevention playbook covers how to use those early signals to act before disputes land.
Reading Sudden Spikes and Sharp Dips
One-day deviations — a point significantly above or below the surrounding trend — usually have a simple explanation. Understanding them prevents unnecessary alarm and unnecessary inaction in equal measure.
Single-day spikes (unusually high average)
A day where the average spikes noticeably above the surrounding trend typically means a small number of high-scoring customers placed orders that day. If your VIP and Trusted customers tend to shop on weekends, you may see a regular spike pattern. Alternatively, if you sent an email to your loyalty list, the orders that followed will skew high because those customers have long, clean histories.
Spikes can also appear right after you manually allowlist a group of trusted customers. Allowlisted customers have their score locked at 100, which will pull the day’s average upward if several of them place orders on the same day.
Single-day dips (unusually low average)
A single-day dip is more commonly explained by a batch of risky events landing together. A wave of card-testing activity at checkout will generate score recalculations for every affected device (and, where email addresses can be resolved, for those customers). A marketing campaign that attracted a cluster of first-time buyers will show a low average on the day they converted, simply because new customers have not yet established history.
Historical Sync can also produce a one-day dip in the trend chart. When the sync first runs, it scores many customers from existing order data, including customers with poor histories. If those customers are all scored on the same day, that day’s average may look anomalously low compared to the surrounding trend. It will normalize quickly.
A dip worth reading carefully
A dip that spans multiple consecutive days — not just a single point — and that is not explained by a promotion or sync event is worth examining. Check the high-risk customer list for any sudden additions. If several customers moved into Risk or Critical within that window, read their individual profiles for the signals that drove the change. That is more actionable than trying to interpret the average in isolation.
What the Segment Distribution Adds to the Picture
Immediately below the Trust Score Trends chart on the same dashboard panel is the Customer Segments stacked bar. This shows the current proportion of your scored customer base in each of the six segments: VIP, Trusted, Normal, Caution, Risk, and Critical.
The segment distribution and the trend chart answer different questions. The trend tells you whether the average is moving and in what direction. The segment bar tells you what the distribution looks like right now — how much of your base is in the healthy segments versus the concerning ones.
Long-standing customers with clean histories. Low refund rates, no disputes, established order patterns. A growing VIP segment means your loyal base is holding and expanding.
Good customers with enough history to be confident in. Some minor signals may exist but nothing concerning. Most of your healthy repeat buyers will sit here.
Customers with limited history or a mixed signal picture. This is where new customers start. A high Normal proportion in a growing store is expected and usually not concerning.
Customers whose behavior has raised some flags. Worth watching but not necessarily acting on. An order from a Caution customer is fine to fulfill — just check the profile before shipping high-value items.
Customers with multiple significant signals. These are the ones worth reviewing manually before fulfilling orders. The Live Telemetry panel tracks how many new high-risk customers have appeared in the last 7 days.
Customers with the strongest negative signals. In many cases this means confirmed disputes, very high return rates, or verified multi-account fraud. These profiles deserve a direct look before any further orders go through.
Read the two visualizations together. A falling trend with a shrinking Risk and Critical segment means your interventions are working — you are removing problematic customers from the scoring pool. A falling trend with a growing Risk segment means the problem is getting bigger, not smaller.
The default thresholds for each segment (VIP at 90, Trusted at 70, Normal at 50, Caution at 30, Risk at 10) are configurable in TrustLens Settings if your store’s profile warrants different cutoffs. Adjusting these thresholds will shift how customers distribute across segments and, consequently, how the trend average reads — worth keeping in mind if you make changes and then see an unexpected jump.
From Chart to Action: Drilling Into Individual Customers
The trend chart is a navigation tool, not a conclusion. Its job is to tell you when something is worth investigating — not what to do about it. When the chart moves in a way that warrants attention, the next step is to move from the store-level average to individual customer profiles.
The high-risk customer list on the right side of the Command Center shows customers currently in the Risk and Critical segments. Clicking any customer from that list opens their full profile, which shows the specific signals that moved their score: how many refunds they have filed, whether their return rate has climbed above your store average, whether linked accounts share their shipping address, and whether any disputes are recorded. For a fuller picture of how TrustLens builds that profile, see the guide to what TrustLens actually does.
There is no automatic action that follows from a trend movement in the free version of TrustLens. What you see on the trend chart tells you whether to look; what you find in the customer profiles tells you what to do. That might be nothing — or it might be manually blocking a customer, opening a closer watch on their next order, or checking whether a Caution customer who just placed a large order is worth shipping to without verification.
A useful weekly check-in routine
Open the dashboard once a week. Glance at the trend direction and whether it changed from the previous week. Look at the segment bar to see if the Risk or Critical count has grown. If either number moved significantly, open the customer list filtered by segment and read the profiles of anyone who changed segments that week. That takes about five minutes and covers most of what you need from the trend data.
When the Chart Is Blank or Shows a Flat Line
The trend chart only renders when there are at least two days of scoring data. On a new installation, this means you will see a blank chart for the first day.
The fastest way to populate the chart is to run Historical Sync from the dashboard. This processes your existing WooCommerce orders in the background and builds trust profiles from your store’s historical data. Once the sync is complete, the trend chart should show meaningful data if your store has sufficient order history.
A flat line — where the chart renders but the value barely moves — usually means one of two things. Either your store has very low scoring activity (few orders in the window, few customers being rescored), or the distribution of your customer base is stable with no new customers joining and no existing customers changing segment. Both are benign. If you are expecting movement but not seeing it, check that the TrustLens detection modules are enabled in Settings and that Action Scheduler is running normally (visible at Tools → Scheduled Actions in your WordPress admin).
Free vs Pro in the Analytics Picture
The Trust Score Trends chart on the Command Center dashboard is a free feature. All eight detection modules run in the free version, score every customer, and contribute to the trend data. There are no restrictions on the trend window or scoring depth based on your plan.
Where Pro expands the analytics picture is in the Chargeback Monitor. The free Chargeback Monitor shows a blended chargeback ratio speedometer — your overall dispute rate compared to the major card-network thresholds (Visa, Mastercard, Amex, Discover). Pro adds a 12-month trend chart of that ratio, per-brand breakdown (tracking each network separately), a trailing-30-day window, a dispute-deadline worklist, and daily email alerts when your ratio approaches a threshold. That chargeback ratio trend is a different chart from the Trust Score Trends chart and answers a different question: not “how healthy is my customer base?” but “how is my dispute rate tracking against the card-network fine lines?”
Pro also adds Automation Rules, which can turn a trend pattern into automatic action — for example, holding orders for review when a customer enters the Risk segment, or sending a webhook when a score drops below a threshold. The third post in this series, how TrustLens Automation Rules work and which responses to automate vs keep manual, covers the full trigger-condition-action model. For now, the free version gives you complete visibility; acting on that visibility is a manual step unless you have Pro.
Common Questions
Does the trend chart update in real time?
No. Score updates run asynchronously via Action Scheduler, so there is a short lag between an event occurring and its effect appearing on the chart. The dashboard refreshes when you load it, but it does not auto-update in the background while you are watching. Reload the page to get the latest data.
Why does my average look low even though I know most customers are good?
The trend chart shows the average of customers scored on each day, not the average of all customers in your database. If your store sees bursts of first-time buyers — after a flash sale or a promotional email to a cold audience — those customers will score in the Normal range (around 50) because TrustLens does not have enough history to differentiate them yet. The average can look lower than your “real” customer quality during and after acquisition campaigns. The score improves as customers return and accumulate history.
Can I see how an individual customer’s score has changed over time?
Not as a time-series chart. The trend chart on the dashboard is a store-level aggregate. Each customer’s detail page shows their current signals and the reason for their current score, but does not display a personal historical score chart. To understand a customer’s trajectory, look at their total order count, refund history, chargeback count, and return rate on the profile page.
What score should I consider “healthy” as a store average?
There is no universal benchmark because it depends heavily on your store’s customer profile, product category, and return policy. A store selling high-ticket electronics will have a naturally lower average than one selling digital downloads, because physical goods generate more returns and disputes. A more useful question is whether your average is stable or moving, and in which direction. A consistent average around 60–70 is typical for established stores with a reasonable repeat buyer base. An average below 50 that is declining is worth investigating. An average above 70 probably reflects either a mature, loyal customer base or a relatively quiet store where only established buyers are active.
How does the trend relate to my chargeback risk?
A falling trend is a leading indicator of potential chargeback risk, not a lagging one. Chargebacks are filed weeks or months after the order — by the time the dispute arrives, the behavioral signals that predicted it are already recorded in TrustLens. A customer who goes on to file a chargeback will often show declining signals (missed deliveries, multiple return requests, linked accounts with dispute histories) before the formal dispute is submitted. That is the value of a trend-based view: you see the early warning before the formal damage arrives. The chargeback prevention playbook explains how to act on those early signals.
What happens to the trend data if I rotate my WordPress secret keys?
This is important to know. TrustLens uses your WordPress authentication key as the hashing material for customer identifiers. If you rotate those keys (through a security plugin’s “regenerate keys” feature or by editing wp-config.php), TrustLens can no longer match returning customers to their existing records. The trend chart data remains in the database, but the scoring system will treat returning customers as new because their hashed identifiers have changed. Running Historical Sync after a key rotation rebuilds the profiles from existing order data using the new keys. The trend chart will reflect the gap if there is a period between the key rotation and the sync.
What to take away from this
- The Trust Score Trends chart shows a 30-day daily average across all scored customers. It is a store-level health indicator, not an individual customer tracker. Use it to know when something is worth investigating — then open the customer list for the specifics.
- A falling average has two distinct causes: dilution from new customers, and genuine risk accumulation. Look at whether the customer count is rising alongside the falling average. If it is, you are probably seeing a surge of new, unscorable buyers. If the count is flat and the average is falling, existing customers are deteriorating — that warrants a closer look.
- Single-day spikes and dips are usually explainable. A spike after emailing your loyalty list, a dip after a sale that attracted first-time buyers, a bump after running Historical Sync — these are normal. A sustained multi-day trend in either direction is more meaningful than any single point.
- Read the segment distribution bar alongside the trend chart. The stacked bar shows what proportion of your customer base is in each segment right now. A falling average with a growing Risk or Critical count is more concerning than a falling average with stable segment proportions.
- The trend chart is a free feature. All eight detection modules and the full 30-day trend are available on the free plan. Pro adds a separate 12-month chargeback ratio trend and the automation rules that let TrustLens act on what the trend chart surfaces.
- Version 1.3.8 fixed the trend chart’s core behavior. If you are on an older version and the chart collapsed to a single point or showed implausible flatness, the 1.3.8 update corrected the snapshot-preservation bug that caused this.
TrustLens is available on the Webstepper plugin page, and the free version is on the WordPress plugin repository. The Trust Score Trends chart, the segment distribution, and all eight detection modules are part of the free plan with no restrictions. If you want to understand how TrustLens translates those signals into a complete risk picture for each customer, the guide to what TrustLens actually does walks through each detection module in detail.
