Pricing

The price-per-word paradigm can give the impression that translation is the only task being undertaken when paying for services. It is often a distraction from important elements of a localization program.

Pricing
Summer in New Hampshire

Welcome to Part 3 of the Buyer Series! I am so glad you that are here. Let's dive into price discovery and how translation memory leverage impacts cost.

Cost vs Investment

In Part 1, I listed the decisions made by Diane and the management team at Company X.  They initially decided to localize their Wordpress site and Product Oh Yeah. They earmarked a budget of $10,000 for the project.

Company X does not have unlimited funds, of course. As it is considering translation for the first time, it does not even have ROI data for the effort. However, companies that start their due diligence this way do themselves a disservice. Localization is no longer about translating 1,000 words of static content and never looking back.  

It is about content that Company X's writers are evolving to meet the needs of the company's business model as well as the changing tastes of Company X's customers. While Company X may have some legacy help pages or blogs that are indeed static, the rest of it will change often or become outdated as time goes by.

Budgets therefore need to take into account various translation services: i.e., translation and editing at first, followed by updates to varying degrees. This reality is at odds with the one-and-done translation project mindset as is implied by a rough budget.

There are some challenges here that LSPs need to overcome. The main one is that the per-word pricing model gives the impression that the price covers 'translation' exclusively. But as mentioned above, there are layers of services underneath, each with its own cost. They need to be woven in and out of the workflow at specific times to achieve specific tasks. Translation → editing → QA, for instance, is a common workflow. Each step involves a different linguist and process. Budget needs to be allocated to them as they fit Company X's objectives.

I'm not saying that Jill, Diane, and Company X should be willing to pay any amount. They need to carefully evaluate the pricing disclosures, on the one hand, and the LSPs being evaluated need to make the itemization of costs very clear, on the other.

Leverage

In Part 2: Workflow, Diane asked to postponed translating Product Oh Yeah in favor of the Wordpress content. From the LSP's perspective, this is actually a good thing. When working with a customer who is new to localization for the first time, it helps to start small and spend time getting to know each other. (To get a glimpse of your readiness score, please see the Readiness Scorecard.)

To recap, here are the word counts that Beth—the Wordpress developer—shared with Jill previously:  

  • Wordpress site: 30,856 words (In-scope for the first project)
  • Product Oh Yeah: 17,888 words (Postponed)

Jill is evaluating three LSPs. She shares the source xliff files from Wordpress with each one and asks them to analyze the files and determine the billable word count. Jill should receive an email that explains the quote she is about to see along with details about the translation memory tiers and corresponding weighted price per word. Let's do that now.

Dear Jill,

Thank you again for considering LSP #1 (or #2 or #3) as a potential partner and giving us the opportunity to provide you with a quote.

The attached Excel contains our cost for translating the 30,856 words contained in your Wordpress site by translation memory tier. The price is slightly different for each language and reflects our costs of doing business in those locales.

Please keep in mind that these costs are a starting point. There are other associated costs regarding glossaries, project management and the like that we should also discuss if you would like to move forward.

We look forward to our next call to discuss.

Best regards,
LSP #1 (or #2 or #3)

Below is what Jill should expect to see: a price-per-word quote in table format that clearly defines the number of words that fall within each translation memory tier and for each language.

LSP #1 - $8,311.80 (Mid-Range)


LSP #2 - $10,005.66 (Most Expensive)


LSP #3 - $6,705.44 (Least Expensive)


Evaluation

Company X should expect little TM leverage at this point. They have no legacy translations to reuse and the majority of words will be 'new words' (words with no leverage), for which Company X will need to pay full price. That is consistent among the three quotes.

There is one tier that will apply at this stage called 'repetitions.' These are phrases and sentences that repeat within the overall body of content, irrespective of translation memory. Imagine a phrase like "Subscribe to our newsletter!" appearing in the footer of all of the pages on Company X's website. That would be considered a 'repetition.' It only needs to be translated once and then that translation can be propagated throughout the rest of the project.

It is good practice to have a linguist confirm that the repetitions fit the context. They are like free-floating strings that could appear anywhere. In other words, a sentence existing in one paragraph (context) could be reused in another paragraph (with a different context) and therefore requiring a different interpretation. Note that the term 'repetitions' is sometimes referred to as 'external matches' and charged at a much lower rate.

Per-Word Pricing Table

Jill creates this table and adds it to the Project Plan.

Language LSP #1 LSP #2 LSP #3
Korean $0.11/word $0.13/word $0.09/word
German $0.10/word $0.12/word $0.08/word
Spanish (US) $0.09/word $0.11/word $0.07/word

A Note About 100% Matches

One challenge to the tiered pricing model is that the impact of the pricing does not always become apparent right away. For instance, you may not have noticed it, but LSP #3 does not charge anything for 100% matches (the purple rows in the screenshot below).

In some use cases this might be fine. After all, the concerned segments (sentences) have been translated already, word-for-word.

However, the 100%-match segments are free floating like repetitions. They will appear sprinkled throughout the rest of the Wordpress content. The question is, therefore, will the context—the sentences before and after the relevant repetition—always be the same?

Recall that a 101% match (ICE match) means the segments above and below it have also been translated before, word-for-word. That's not true in the case of a 100% match, so having a human  editor eyeball them for context fit is a wise move. Think of it as a small sum to pay for insurance.

Which is why this is not good news. By charging zero for that tier, LSP #3 is effectively saying "We are relying solely on the TM and will not double check 100% match segments. We won't look at them." So, it is possible Company X could receive some correctly translated sentences that fit awkwardly in running text, which is a vexing thing to discover and have to fix down stream.

The Takeaway

Jill learned how critically important understanding translation memory leverage is in keeping costs down. She is also seeing details that will impact the localization program in material ways that she needs to communicate to Diane and, in so doing, help Company X make the right decision.

  • The estimates from the three LSPs are informative but require scrutiny.
  • Jill needs to update the Project Plan to give visibility of progress to Diane
  • But she also needs to start thinking about a summary statement about vendor selection, but the intricate details matter.

The picture is becoming clearer, but there is more work to be done. Part 4 of the Series covers Language Assets.

Localization Roles Used in This Series

Company X: A business with a Wordpress website and a flagship product called Product Oh Yeah. Both need to be translated.

Jill: The localization manager (LM) working for Company X. Jill's title is 'Content Manager.' She was hired to create and manage content in Wordpress, and she has no previous experience with localization.

Diane: Content Strategy VP at Company X. Appointed to oversee the localization budget. Diane wants to translate the Wordpress website and Product Oh Yeah into French (France), German, and Spanish (US) if the costs fit the initial earmark. If not, then Spanish (US) only.

Lisa: Friend of Jill working at another company, Company Y. Has some experience localizing a Wordpress website into Korean.

Beth: A Wordpress developer working at HQ for Company X.

Stories and articles that inspire.