‘80% of success is showing up’- Woody Allen
He could have been referring to multi-lingual text analysis. Is that good news?
It might be. If you are extracting multi-lingual data from social posts you’re far ahead of the competition. The bar isn’t high, and your competitors probably aren’t close to extracting data in its native language. If they are, they aren’t doing it very well.
‘80% of success is showing up’- Woody Allen
He could have been referring to multi-lingual text analysis. Is that good news?
It might be. If you are extracting multi-lingual data from social posts you’re far ahead of the competition. The bar isn’t high, and your competitors probably aren’t close to extracting data in its native language. If they are, they aren’t doing it very well.
Native language text analysis is complex but it’s not impossible. To get far ahead of the competition, make a plan that celebrates small wins. You will position your business for solid gains that tie directly to action.
This was a remarkable point from last week’s Social Media Analytics Summit (#SMAS12) from Text Analytics News. The United States is a multi-lingual society. Yet everywhere is more multi-lingual than here. Foreign languages are no longer ‘foreign’, but existing sentiment analysis tools (social, text, video) don’t reflect this reality or meet the need. That’s a disadvantage, and we have to look no further than recent demographic data to see why.
Consider the data from the 2010 US Census:
- 20% of the population (over the age of 5) speaks a language other than English at home
- 35 million – US residents speak Spanish in the home
- 50% of US population growth is coming from Hispanics
- Asian Americans are the fastest growing population segment- 46% increase 2000 vs 2010
- 2.6 million- US residents speak Chinese in the home
- Non-native speakers prefer to conduct business in their native language
This is a business and a cultural issue. Businesses must adapt their analysis to meet the needs of their customers.
How?
Start with a cross-lingual solution that works in the native language of the text. Some of the existing solutions rely on translations. Translations are labor intensive and inaccurate. And if the translation doesn’t work, the analytics don’t work. Ad technology solutions that incorporate key features of volume, velocity and variety (text, video, languages et al) are beginning to address this gap. These cross lingual solutions work to improve efficiency and efficacy, but that’s not the entire picture. Efficiency doesn’t directly convert to insight, and insight does not directly correlate to ROI. To get to ROI, you have to tie to action; it’s your efficacy.
Work backwards. Start with a conclusion and work back. Make a plan that rewards small wins, allows for incremental improvement and has a clear path to ROI. As mentioned you get to ROI and efficacy by isolating action, and action by working in the native language. Advertising technology can help, and it will get better more precise.
Make a plan, and stick with the plan. While taking the time to consider and evaluate:
- Advertising technology- it should work with any language
- ROI- focus on conversion, but evaluate
- Plan- take the detailed view, iterate, celebrate small wins
When data is left unexamined, you’re potentially leaving money on the table.
And the numbers that the C-suite ultimately cares about are preceded by a currency sign.
Is your business ready to meet the challenge? Talk to an expert, they should offer the insight and consultative approach you need. And in this way you stand to get to action from multi lingual text analysis.
For further information/insight, I’ll direct you to by Meta Brown of LinguaSys, whose SMAS12 session- Capitalize on Multi-Lingual Social Analytics– formed the basis for this post.
