They measure the same kind of feedback over different windows. The rating used in level criteria covers the past two years, while the profile rating reflects all-time ratings. A seller can therefore show one number on a level overview and another on the public profile without either being wrong. Compare them as two labeled metrics over two labeled periods, not as evidence of hidden reviews or inflated claims.
Find public social profile clues with our TraceSeek app.
Name the two numbers before you compare them
Start by writing down each figure with its own label and period, side by side, not one above the other. The level figure belongs to a periodic evaluation of criteria that include rating, response rate, orders, unique clients and earnings. The profile figure belongs to the seller's all-time rating display. When two numbers carry different windows, a gap between them is expected arithmetic, not a discrepancy to explain away. [Fiverr: level metrics and rating windows]
- Record the level metric with its two-year label.
- Record the profile metric with its all-time label.
- Do not merge them into a single average of your own.
Why the windows split apart
A recent buyer's rating enters both windows, but old ratings stay only in the all-time profile figure once they fall outside the two-year level period. A seller whose early work drew very different ratings than recent work will show the widest separation. Steady feedback across years tends to keep the two figures close. This is a denominator-window comparison: the same rating type, counted over different spans of time. [Fiverr: level metrics and rating windows]
Keep observation, limit and next step apart
An observation is only the visible pair of figures with their labels. The limit is that neither number tells you review order, which buyers were most recent, or how a seller earns. The next step is practical: read a sample of individual reviews and seller responses as separate objects rather than treating a summary figure as a verdict. The table below shows how to hold those distinctions together. [Fiverr: level metrics and rating windows]
| observation | limit | next step |
|---|---|---|
| Level figure labeled with a two-year rating window | Does not show all-time history | Read it only as the level-period metric |
| Profile figure labeled all-time | Does not show recency or trend | Read recent reviews separately for a sense of direction |
| The two numbers differ | Does not imply a fabricated rating or hidden feedback | Compare labels and periods rather than accusing anyone |
A hypothetical worksheet
Suppose a seller's level overview shows a rating near 4.6 over the two-year window and the public profile shows 4.8 all-time. The labeled reading is: level-window metric 4.6, all-time profile metric 4.8, both from buyer ratings. The gap is consistent with older, higher ratings still counted in the all-time figure. This example is hypothetical and illustrates labeling, not any real seller. Levels can change through periodic evaluation, and a badge is not a personal character or income certification. [Fiverr: level metrics and rating windows]
- Write each number with its window label.
- State the arithmetic reason for any gap.
- Leave earnings, intent and character out of the note.
When one source is missing, private or ambiguous
If the level overview was not shared, the profile rating is absent, or the two clearly come from different sellers, record that exact limit and stop. Do not fill the gap with estimates of personal earnings or guesses about hidden reviews. If the artifact is voluntarily shared, use only what it shows and note the missing side plainly. Some sellers share overviews inconsistently, so an incomplete comparison is a normal outcome, not a finding. [Fiverr: level metrics and rating windows]
What you can and cannot take from the pair
You can take two labeled figures over two stated periods and a plain explanation of why they may diverge. You cannot conclude that a seller is dishonest, that reviews were ordered or highlighted in a particular way, or that a rating window implies anything about income or reliability beyond the stated metrics. The App or page version you are reading may show these labels differently, so check the options shown where you view the profile before drawing a conclusion. [Fiverr: level metrics and rating windows]
A couple more questions
Does a lower level rating than profile rating mean something is wrong?
No. The level rating covers the past two years, while the profile rating is all-time. Older ratings outside the level window can keep the profile figure higher, so a gap is expected and does not imply a fabricated rating.
Can I tell which reviews count toward the level rating?
Not from the two summary figures. The level figure is a windowed metric, and individual reviews are separate objects you read on their own. Use the labeled numbers only for what they state.
What should I do if the overview is not shared or the figures seem to come from different sellers?
Record the precise limit and stop there. Do not estimate earnings or infer hidden feedback; an incomplete comparison is simply incomplete.
Sources & editorial notes
Official references checked on October 24, 2026. Checklists, comparison tables and example wording are DateClues editorial tools, not official tests or validated risk scores.
General information. No guide or tool can certify a person’s identity, intentions or safety. Read our editorial standards.