Standard Page: Some Historical Background

I live in a country where the traditional unit is the “standard page”, so, for the younger among you, let’s look at its origin first.

Before computers, we used to type on a machine called “typewriter” – maybe you have seen it in a museum-:). A standard page (SP) was then defined as 30 lines of 60 characters each (spaces included). We used to just count lines, tedious extra work to be done after completing each translation. Of course the lines may have been shorter or longer than 60 characters (mostly longer, I believe), but there was no way to count characters manually. For translating books and generally materials intended for print, paper sheets with a frame were used – the type was supposed to fit in the frame, which delimited the SP. By definition, the target lines/pages were always counted.

This results, today, in the definition of a SP as 1800 characters with spaces (30x60). Another definition used is 1500 characters without spaces – this is considered roughly equivalent, but is it really? According to another standard, 1 SP holds 250 words (which is not true in all languages – see below). Therefore, this would include around 250 spaces. 1500 + 250 = 1750, not 1800! So if you count in SP 1500, you gain 50 characters per page (more or less depending on the language). This is negligible when you translate 5 pages, but in a book of 500 pages, the difference will be substantial!

Words per SP

When I first started working in other countries and needed to provide quotes in words, I took a number of texts in each language involved and calculated average numbers of words per SP (then defined as 1500 characters without spaces). The results were as follows:
English.....282
French......290
Czech.......257
Bulgarian...244
Romanian..252

My samples were not large enough, so these figures are to be regarded with caution – but it is clear that the standard of 250 words/SP is just a rule of thumb to quickly calculate how long a job will be in units familiar to you, and are not to be used for quoting purposes, unless the agency defines a SP as 250 words – so far, I encountered it just once. In such a case, it is easier just to count in words.

Source vs. Target

The target is the work that you must “actually do”, physically type, so it seems logical to base your rates on that. I compared source texts and their translations (my own) into other languages – in languages that I am able to translate in both directions, my sample included equal numbers of samples in both directions (to eliminate any possible poor translation that results in lengthening the target). I found that English and Czech were about the same length (the same number of characters), while French, Bulgarian and Romanian tended to be much longer – 20% on average, but in a range from nearly zero to 40% for French (e.g., very polite letters, some legalese… If you work in only 2-3 language pairs, you can take samples and calculate the source/target ratio per specialty field to be more accurate.) Therefore, it can make a huge difference whether your rate is set for the target or for the source! Again, these figures are approximate, as the samples weren’t large enough. For illustration: I just completed a translation of the same Czech text (in the field of accounting) into both English and French, and the French target is 12% longer than the English one.

If we apply the numbers of words per SP to the average differences in number of characters between the source and the target, we will get the following source/target ratios in terms of words:

Czech to French: 252 words/SP in source, 290 in target, target longer by 20% on average in terms of characters. There will be 20% more SPs (characters) in the target, and each target SP will be composed of 290 words – the multiplication will give us the number of target words, which will then be divided by the source words to arrive at the ratio:

290*1.2/252 = 1.38

Therefore, in number of words, a French translation will be 38% longer on average than its Czech source. It is then the translator’s interest to have the rate based on the target!

French to English: 290 words/SP in source, 282 in target, source longer by 20% on average.

282/(290*1.2) = 0.81

In number of words, an average English translation will represent only 81% of its French source, so you should ask to be paid by source word!
Etc.

And Then You Convert Currencies…

…while taking into account their possible evolution up to the time you are likely to get the payment, if significantly fluctuating currencies are involved – based on the assumption that you convert the money on receipt, which is not necessarily true, but you will be free to do that and it’s up to you to decide. Also fees for receiving the payment, if significant, should be considered – on that subject, see my first “Let’s count” article on payment methods. As an example, let’s take a PayPal charge amounting to 4%. Of course, the rate you calculate using this method may not be acceptable for the client – but you will have an idea on how the rate offered by them compares to your usual rates.

Example
For a French to English translation, let us say your rate is €0.10 per source word (this is an example for easy calculation – not for discussion about high or low rates!). You should quote per target standard page in £. The entry data are the following:
- Words per SP: source 290, target 282
- Ratio source/target in characters/SP: 1.2/1
- Exchange rate GBP/EUR (as I am writing this): 1.188, but the lowest rate over the past 3 months was 1.139 – let‘s take that as a “worst case scenario” rate; however, if you are paid by PayPal and intend to convert the money there, you should use the rate you are likely to get there, which is much lower! In the case of a wire transfer, the rate you will get in a bank is usually about 2% lower than the official rate. This is just an example.
- PayPal fee 4% (also the worst case scenario in most cases – see article about payment methods)

First, let’s convert your rate to source SP: 0.10 * 290 = €29

1.2 source SP will correspond to one target SP; therefore, if your usual rate is based on the source, you will need more money per target SP: 29*1.2 = €34.8

Then convert it to GBP using the current and the “worst case scenario” exchange rates (be sure to apply them in the right direction! The amount in £ should be lower than the amount in €!):
34.8/1.188 = £29.29
34.8/1.139 = £30.55

And finally, let’s add the PayPal charge:
29.29 * 1.04 = £30.46
30.55 * 1.04 = £31.77

Your quote should therefore be in between these two figures – you will probably propose £32 and if the client says it’s too much, you will be able to go down (depending also on how likely the “worst case” scenario seems to be) – your minimum will be £30.46 as calculated above if you want to keep your usual per-source-word rate.

You can prepare a conversion table in Excel per language pair and/or specialty field, where the parameters will be in cells to which your formula will refer (this will make it possible to update exchange rates for example). For example, in the “£/target SP PayPal” cell of the “French to English” table, the above calculation would be resumed as follows:

=[cell containing your rate]* [cell containing the number of words/SP in the source]* [cell containing the source/target ratio]/ [cell containing the exchange rate (or multiplied by it if you use the inverse rate)]* [1 + cell containing the percentage of charges expressed as a number, here: 0.04]

which may look like that (example!): =A10*B2*B4/B5*(1+F5)

0.1*290*1.2/1.188*(1+0.04) = 30.46

The source/target ratio must also be applied in the right direction – which will be opposite if your usual rates are based on the target and you want to calculate the per-source rate. To check that: a target language longer that the source will increase the source rate compared to the target rate, and decrease the target rate compared to the source rate. The opposite is true for a target shorter than the source.

When you have a few hours, do your homework (the words/SP and source/target statistics) for your language pairs – you may be very surprised by the results!