Why does the customer's wordcount and my wordcount differ so much?
Thread poster: Fredrik Pettersson

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
Aug 5, 2013

Why does the customer's wordcount and my wordcount differ so much?

It's a large project on 54,000 words. And the customer first sent 2 textfiles to be imported in a new TM in Workbench 2007. I then begun to translate in TagEditor, but noted that the wordcount couldn't be correct. So I checked it, and compared to the customer's original analysis. The difference was huge, many thousand more new words! Then double-checked in Studio 2011, and received the same word count there.

The customer then sent 25 new TTX-files, and I was surprised because I thought they just had missed to send me a textfile to import into the TM. But with these 25 new TTX-files, the word count is the same! And the customer insists on that they get the same wordcount as earlier.

This is the first time I've run into this situation, because the differences are huge for new words, and each of the fuzzy intervals and 100 % matches.

What settings in Trados could cause this big difference? The material to be translated is nothing special, just a normal manual for an industrial machine.


 

Łukasz Gos-Furmankiewicz  Identity Verified
Poland
Local time: 11:03
English to Polish
+ ...
Did the agency use Trados for the analysis? Aug 5, 2013

Sometimes they use other CATs, and other CATs can show fewer words than Trados.

 

Gerard de Noord  Identity Verified
France
Local time: 11:03
Member (2003)
German to Dutch
+ ...
Ask your question again in a more formal way Aug 5, 2013

Different (versions of) tools can come up with a difference of several hundreds of regular words on a 50,000 word project but that doesn't seem to be your problem.

My guess is that you're both using Trados to count and that your client is applying internal fuzzies (virtual matches that might materialize during the project) and you aren't.

So you're seeing different Trados analyze results, not different word counts.

Cheers,
Gerard


 

Eileen Cartoon  Identity Verified
Local time: 11:03
Italian to English
Paul (from SDL) explained it clearly once Aug 5, 2013

you can read his explanation at

http://kb.sdl.com/?portalid=23&articleid=4548#tab:homeTab:crumb:7:artId:4548

It explains very clearly the differences.


 

Gerard de Noord  Identity Verified
France
Local time: 11:03
Member (2003)
German to Dutch
+ ...
Paul from SDL needs a formal question too Aug 5, 2013

Eileen Cartoon wrote:

you can read his explanation at

http://kb.sdl.com/?portalid=23&articleid=4548#tab:homeTab:crumb:7:artId:4548

It explains very clearly the differences.


Hi Eileen,

Its the virtual fuzzies hater again. I read the article you link to but in that article Paul from SDL only explains small differences, while Frederik is talking about a difference of thousands of words.

So we have to wait for Frederik to rephrase his problem in a way that allows us to help him.

Cheers,
Gerard


 

Annamaria Amik  Identity Verified
Local time: 12:03
Romanian to English
+ ...
Could be a format issue? Aug 5, 2013

Not sure if my answer helps, but I had a similar problem a while ago. As far as I recall, I got the correct wordcount / analysis (the one communicated by the client) when I changed the format (from docx to doc or viceversa, don't remember) of the original file, reopened it in TagEditor and ran the analysis on that file.

[Edited at 2013-08-05 21:36 GMT]


 

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
TOPIC STARTER
Reply from the customer Aug 6, 2013

I received this reply from my customer:

Adjunto una captura de las propiedades de segmentación de las que habla Trados, por si fuera problema de penalizaciones en tu ordenador. He comparado ambos análisis y no es normal tantísima diferencia. He vuelto a hacerlo en dos versiones diferentes de Trados y sale el mismo recuento, con lo cual creo que puede que tu ordenador no cuente las coincidencias “fuzzy” o las cuente como “no match” directamente. En nuestro análisis tenemos 35.000 repeticiones del 100% y en el tuyo sólo 10.000, por ejemplo.

Vuelvo a adjuntar ttx, memorias y análisis para comprobar una vez más, siguiendo esa base de penalizaciones, si sigue dando error.

Voy a intentar hacer el análisis con memoQ para comprobar dónde está el error. En caso de que sea nuestro, hablaríamos con el cliente para que se pudiera pagar todo el trabajo, por supuesto.

I think he means my settings for penalties in Workbench (penalties for formatting etc.) is different from his:

Customer's penalty settings
http://screencast.com/t/VDgyjAKke8B

My penalty settings
http://screencast.com/t/nhV2MrkE

But as you can see, we have exactly the same penalty settings!


 

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
TOPIC STARTER
The customer approved my wordcount! Aug 6, 2013

The customer finally approved my wordcount, said that his TM was corrupt.

 


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

Why does the customer's wordcount and my wordcount differ so much?

Advanced search







SDL Trados Studio 2019 Freelance
The leading translation software used by over 250,000 translators.

SDL Trados Studio 2019 has evolved to bring translators a brand new experience. Designed with user experience at its core, Studio 2019 transforms how new users get up and running, helps experienced users make the most of the powerful features, ensures new

More info »
Déjà Vu X3
Try it, Love it

Find out why Déjà Vu is today the most flexible, customizable and user-friendly tool on the market. See the brand new features in action: *Completely redesigned user interface *Live Preview *Inline spell checking *Inline

More info »



Forums
  • All of ProZ.com
  • Term search
  • Jobs
  • Forums
  • Multiple search