A guideline to choosing transcription service providers

Over the past decade or so, technology has simplified the Audio-to-text conversion process, for many industries. One such simplification is the shift from Manual/ human-dependent transcription to Automated/ tech-enabled transcription; in Consumer Research, Pharma/Healthcare, Legal services etc.

Application of technology to Audio-to-text conversion means that transcription can now be easily outsourced to equipped service providers. As these tech applications improved in terms of accuracy and became affordable, their spread has steadily grown. Industry forums now list dozens of professional transcription service providers. This makes it deceptively easy to find a service provider meeting your needs of transcription volume and quality. However, it is only after the first few projects are delivered, that you can ascertain whether you chose right.

This ‘hit-and-miss’ effort can be easily minimized, if the evaluation criteria for choosing a service provider are clear. Broadly speaking, any outsourcing decision is weighed basis 3 major factors:

1. Speed to Insights: How much faster can the service provider deliver transcription?

Typically, human transcription productivity ranges from 40-100 words per minute, depending on the transcriber’s expertise. Which means a beginner is likely to take approximately 4 hours to completely transcribe a one-hour audio file, whereas a professional could do the same task within 3 hours. Transcription software can cut this down to as low as 10 minutes. Even if we have to add time to conduct quality checks, the total time taken would be significantly lesser. 

Faster turnaround on transcripts allows the researcher to spend more time on analysis and culling out insights.

2. Meaningful output: How much more accurately can the service provider transcribe?

Qualitative researchers manage both, Domestic and International clients. This means data is captured in multiple languages. Hence, it becomes critical to evaluate a service provider on Accuracy, particularly for multi-market studies. Accuracy is determined at two levels: 

2.1 does the service provider use a combination of advanced speech recognition technology and native human speakers with experience across dialects and industries? 

2.2 does the service provider’s process include thorough quality checks for accuracy?

Accurate transcription reassures researchers about having an authentic grasp of the consumers’ opinion; which in turn enables them to confidently massage the dataset for true-to-life insights.

3. Commercial value: What is the price at which the service provider delivers transcripts? 

Pricing becomes an important criterion particularly for large-volume requirements. Pay-per-use prices might work for one-off requirements, but not for repeated, large-volume transcription needs.

Affordable pricing without compromising Quality helps researchers manage profitability; across large and small projects.

Apart from these, certain other factors that can further enable choice are:

Customer support

Check whether the service provider offers multiple channels of communication (email, phone, live chat); supported by a responsive support team.

Flexibility

Check whether the service provider is open to providing a free trial; sharing sample transcripts; providing Automated transcriptions while they manually check the transcript for accuracy etc.

Security 

Particularly when dealing with sensitive data, seek providers that have encrypted file transfers, secure storage and are willing to work using your cloud servers.

All in all, comparing at least 2-3 service providers can help arrive at your ideal partner. Eventually, the objective should be to make Transcription a seamless and efficient task in your QDA process.