Data is a word that refers to all the information that you have. It may live in tables, spreadsheets and a variety of different systems. Metrics, more specifically, is a collection of the same type of data that comprises a method for measuring something.
A key performance indicator (KPI) provides proof of whether a certain condition exists or shows us whether our goals have been met. KPIs help you bring visibility to important indicators and prioritize them. They also guide investment decisions.
A word of caution about KPIs: they are only as good as you make them. If you simply set arbitrary goals, the organization may very well orient itself towards meeting them, but that may turn out to be a waste of resources.
These are the metrics you can start with. They focus on some of the most basic, yet critical factors in the success of a translation project.
Total deliveries that are on-time or early ÷ Total deliveries
When you’re measuring delivery time, you need to know what you are comparing yourself to. Is it the requested date? Is it the renegotiated date based on a change order or delay? Or is it the actual delivery date? It’s important to think very carefully about what you mean by “on time.” Take some time to focus on what you mean, exactly, in all of these different scenarios.
Weighted word count ÷ (Working days - Holidays)
You can compute this by dividing the number of words, pages, articles or any other unit by the number of working days spent producing them. This can be an important productivity statistic regardless of whether you are an LSP trying to manage your partners, a client buying services from multiple LSPs or even if you’re working internally with a team of translators. It is especially helpful in making more predictable deliveries.
Total escalations ÷ Total products or deliveries per region
This is a less-structured data point. It refers to feedback received from end-users about a mistake or an issue, but these end-users can be your CEO, customers, channel partners or even in-country sales folk. You could compare the total number of escalations against total products or deliveries per region, or use some other comparison that makes more sense to you.
Total errors ÷ Total word count
It’s important to measure errors so you can get a handle on your translation quality. Errors have various root causes: there may be inconsistencies from using multiple language vendors on the same project, you may have an underqualified translator or your QA processes may have fallen short. Or, you may be achieving very high quality, but just don’t know it. To measure language quality, you will need to define what an error is and determine how severe it should be rated.
Total passed tests ÷ Total tests
Pass/fail scores are the second-most important metric behind linguistic quality assessment (LQA) scores. Either a translation passes or it doesn’t. Again, you need to define what constitutes passing and failing very carefully. Does it pass based on quality, on-time delivery or other factors?
As a general rule, you should try to understand and establish the essential metrics first before you go on to these advanced ones.
This may not be of much use to you if your translation team is entirely in-house. To calculate it, you will need to look at the total outsourced cost including project management, engineering, translation and review, then divide it by the total number of words. This gives you a sense of your total program costs as a number. If you are leveraging your TMS or translation memory efficiently, this number should be below your new word rate. Words that you can leverage through a TMS are those that you don’t have to touch at all during the localization process. Keep in mind, though, that larger projects may have a higher number of new words, which will bring the average up. So, do consider project types when calculating the fully-loaded price per word.
Divide the total number of employees engaged in globalization by the total number of your company’s full-time and part-time employees. The globalization headcount should include all staff dedicated to product development, global marketing, globalization, localization and internationalization.
Figure out this percentage by dividing the in-house localization cost by the outsourced localization cost. The in-house cost should include full-time and part-time employees that count towards your company’s localization budget as well as localization tools management, QA, engineering, travel and fully-loaded expenses. The outsourced cost should include work done by third parties (including contractors) and technology purchased from third parties for bundling.
This percentage is your total localization cost divided by your total revenue from localized products, content and services. The localization cost should include both in-house as well as outsourced costs. The total localized product revenue would include measurable gross revenue for all sales of local-language product versions or measurable gross revenue by each “ship to” country.
Determine this by dividing the total number of employees engaged in localization in roles like localization engineer, project manager and translator by the total number of your company’s full-time and part-time employees.
This is calculated by dividing the total outsourced cost by the total number of words.
Obtain this metric by dividing the total localized product revenue by your total globalization costs.
We list below some key online engagement metrics that, if measured before and after launching a localized product, can give a fair idea of the efficiency of your localization program.
SEO should drive traffic to the online content you provide in the given locale. This metric shows the number of unique visitors from the locale to a particular piece of content—be it an eBook, blog post or landing page. Analyzing this number will provide further insights listed below.
This quantifies those visitors landing on, or actually preferring, content in their language. A high number of visits to your localized content will show the quality of your inbound marketing as well as the level of user experience your localized content offers.
Are they higher or lower than what you see for other languages? This is an area where optimization of your local content pays off...or of where more work needs to be done.
Closely related to the bounce rate, this metric shows how much time visitors spend on your localized site. How does the metric differ across individual markets? More time spent on your pages usually points to a higher engagement rate.
People who return to your content over and over again are experiencing that content positively. This metric measures the “stickiness” of your local content, as well as the overall quality of your marketing directed to the given market. Looking at the source of your traffic can also tell you what type of content performs best.
The number of conversions—visitors that proceed to perform an action, such as completing a form or signing up to a demo/trial that may eventually lead to a purchase—is an important metric.
The percentage of consumers in each locale that take this further step will tell you if your proposition is attractive for your targeted local audiences. True customization of the content can really help increase your conversion rates.
This metric will show you if your email campaigns are properly customized for your specific international markets.
In this episode of Globally Speaking, Chris Englund, formerly Senior Manager of Localization at HubSpot, provides details about HubSpot’s successful localization program and talks about how your localization approach can drive value and how you can prove program ROI.
Review your strategic objectives.
Identify a list of potential indicators.
Use baseline performance to determine a target.
Eliminate indicators that don't pass a reality check.
Plan data collection.
Align everyone's goal with the KPIs.
Create a dashboard with the results.
Once you establish what you want to measure, clearly and consistently communicate your goals and expectations to the team.
Localization dashboards help build a quick view of the data that matters most to you. Remember to include only the highest-priority indicators and keep them updated and validated regularly. You may also need to assign someone to be in charge of data entry and analysis to manage the data.
Some of the things that you may want to include in your dashboard are:
Schedule - For the percentage of on-time deliveries, the KPI should be understood as “as agreed” rather than “as requested.” The former data impacts the actual production line, while the latter reflects a perfect-world scenario.
Quality - Include a pass/fail rate, which, without getting too involved with the data, will let you know generally if the quality is okay or not.
Cost - This could contain periodic costs to date, average cost per deliverable (page, word, etc.) or costs per duration like week or month.
If you didn’t have a meeting and if you didn’t document the agreement between both parties, then it doesn’t exist.
For quality, how many errors per thousand words are acceptable? What percentage of late deliveries is acceptable?
As moms have been known to say: If it’s worth doing, then it’s worth doing right. So start measuring against KPIs from the very start of your program so you can establish your baseline.
The KPIs you build at the beginning of your program may turn out to be unreasonably difficult to hit or your program may outgrow them. Recalibrate your agreed KPIs as the program evolves.
While some organizations may view localization as a cost, more forward-looking and mature companies consider it an investment that enables international revenue growth. There is a range of leading and trailing indicators that look at the contribution of localization to generating sales in international markets.
International market growth rate and relative market share in a given local market are two that provide the most relevant information. However, the details are critical when it comes to understanding how product localization actually influences revenue. If an original, nonlocalized product was available in the given market, it makes sense to establish a baseline revenue and calculate the increase in sales once the product/content has been localized, then adjust that by the “organic” market growth rate, meaning the rate of growth that would incur regardless.
Most investments in product localization are hardly discrete events. Companies invest in ongoing local product releases, expanding the amount of content made available in a given language over time. As a result, many localization ROI calculations may fail to account for the ongoing nature of such an investment in localized assets.
So, an alternative option is to calculate the return on localization assets. This metric uses the income derived from localized products or services divided by the total or average value of the existing localization assets in the same period.
Nothing spurs action faster than data. So how do you mine the mountains of localization program data for meaningful business insights? We selected a series of real-life case studies to walk you through how you can use your localization data to approach actual problems utilizing several different tools and strategies.