Cost calculation and statistics

Project statistics

Project statistics show the word and character count, and are calculated by looking at your original documents, and by the number of translation memory matches and repetitions found in them.

This data can be found under the Statistics tab, in the project you have selected. Statistics may take some time to calculate, depending on the number and size of documents in the selected project, and the contents of the translation memory databases (TMs) in use.

Statistics

You are free to add more documents and/or TMs to the project, or update existing TMs at any time. Also, you can download statistics in either of the following formats:

  • Trados XML (for use in SDL Trados)
  • Smartcat XLSX (for use in MS Excel or any other spreadsheet software)

Statistics by contributor

This type of statistics contains information about the amount of work assigned to a particular contributor, and the amount of work completed by that contributor. A separate file is generated for each person assigned, language pair and workflow stage.

TM matches and repetitions are calculated for completed work in real time. An inserted TM match is detected instantly, once the segment is confirmed by a translator.

Different versions of statistics

Should you choose to refresh the total statistics, they will be calculated based on the current content of the translation memory. Imagine you have finished translating a document, thus updating the TM with new translations. By clicking the “Refresh Statistics” button, you can have the translated segments recalculated as 102% matches (see details below) since the translation memory has been previously updated with new segments. Note that previous statistics data will not be saved after refreshing.

The statistics for assigned tasks are calculated for each person assigned at the moment of their assignment, and can be refreshed only if you add another assignment or change the active one for that user.

Data for the completed work will be up-to-date at the exact moment you download the report.

TM matches and repetitions

Translation memory match percentages represent the degree of match between the source document segment and a translation memory segment. The minimum threshold for fuzzy matches is 50%.

Here’s an example:

TM segment Source segment Match percentage
Main Settings Main Settings 100%
Home and office Home & office 70%
Research and development Institute for research and development 62%
User interface design and development Institute for research and development 37% (no match)

There are different types of TM matches:

  • 100% (Exact match): The source segment is exactly the same as the match from the TM.
  • 101% (Context match): The source segment and one of its adjacent segments are exactly the same as the matches from the project translation memory.
  • 102% (Double context match): The source segment and both of its adjacent segments are exactly the same as the matches from the translation memory.
  • Repetition: The source segment, which appears several times in the document, has been translated during the current translation session, and stored to the TM.
  • Cross-file repetition: Generally the same as normal repetition, but the source segment appears in multiple documents of the project.

Units of measurement

The following units of measurement are supported for statistics:

  • words
  • pages (one page equals 250 words)
  • characters with spaces
  • characters without spaces

Please let us know if you would like one page to be considered as having, for example, 1,800 characters without spaces, in your projects. We will customize these settings at your request.

Payable words and client net rates

Payable words are calculated based on the net rate selected for a particular project. You can customize net rates, and link them to particular clients. By selecting a client in your project, you can automatically select this client’s net rate.

Net rates

Keep in mind that the standard net rate is applied when working with freelancers from the Marketplace.

Project cost calculation

The cost of work is calculated automatically by multiplying the number of payable words in original documents by the translator’s per-word or, for hieroglyphic languages, per-symbol rate according to the net rate.

MT matches and repetitions are calculated at discounted rates. However, discounts do not apply for the work of editors, proofreaders, and post-editors.

Standard net rate for translators

TM matches and repetitions Percentage of the full per-word rate
New words 100% (full rate)
50–74% 100% (full rate)
75–84% 70%
85–94% 40%
95–99% 10%
100–102% Unpaid
Repetitions, including cross-file repetitions Unpaid

As an example, let’s take a document containing 31 words. Of these, 14 words are new, 12 words are fuzzy matches, and 7 words are repetitions.

Now let’s do the math: (14 × 1) + (12 × 0,4) + (7 × 0) = 18.8.

You can study statistics on each translator, and set other discounted rates.

Note that only the first translator to confirm a segment will be paid for it in full. You can add additional workflow stages, should the translation need editing.

Only completed work is paid. The system calculates the number of segments translated (i.e. confirmed) by each translator. Net rates are applied in real time; the system determines whether or not the segment was retrieved from translation memory upon confirmation. This is important when multiple translators are involved in a project.

For example, Josh has translated a segment, and the segment has been stored to the TM. Later, Andy inserted this translation into a matching target segment, and will be paid at a discounted rate for this segment. If those two segments are not identical, and no translation memory is connected to the project, they will be calculated at a full rate at the moment of the project’s start.

The customer can see the cost of each project member’s work under the Team tab, while the project members can see how much they have earned on the project page.

Cost of work

The freelancer’s rate is established at the moment they accept an invitation. However, you can request that we change the rate, upon a mutual agreement with the freelancer.

The expected cost and the billing cost are different. Why?

The final project cost may not be the same as the cost estimate you see before the start of the project. Possible reasons are:

  • Discounts for TM matches and repetitions were applied which resulted in a lower final cost of the project.
  • Different translators have different rates. The first translator to complete a repeated segment receives the full payment for it; other translators assigned with that segment will be paid at a discounted rate.
  • Tasks were changed. Any changes in the project concerning assignments and roles affects the calculation.
  • New documents or TMs were added. By uploading a new TM base, you might add more TM matches and repetitions to the project, while uploading new documents will lead to an increased word count.
Was this article helpful?