Live Interviews

Follow these steps to get up and running on a simple live interview project using CoLoop.

​

Recording Interviews

When recording live interviews you must ensure you collect high quality audio. We recommend doing the following:

1. Request High-Quality Audio from Moderators/Field Agencies: Communicate the importance of high-quality audio to moderators and field agencies, emphasizing its impact on the project’s success.
2. Initial Quality Check: Listen to the first recording from each market to verify the audio quality. This initial effort can significantly improve the overall results.
3. Importance for Live Interviews: High-quality audio is especially crucial when conducting live interviews in a facility. Clear audio enhances the accuracy and reliability of the data collected.
4. Use of Recording Devices: For in-person interviews, it is advisable to bring a dictaphone or place a laptop or phone in the middle of the table to capture clear and consistent audio.
5. Get participants to introduce themselves: Particularly for focus groups we’d recommend getting participants to introduce themselves. This will save time when labelling speakers.

​

Create your project

1. Upload your Discussion Guide
2. Enter your project objectives
3. Generate a project description

Discussion guides must be in DOCX format. If yours isn’t just copy and paste the contents into a Word doc and upload this. The AI will use this to generate a project description and set of objectives.

​

Why is the project description important?

• Project description is used to provide context to the AI to help it understand what it’s looking at
• The description will be used to evaluate evidence and decide which points are relevant to your question
• Here is an example of a project description, you can read more about this here

Section	Example
Project Description	“This project is exploring different concepts encouraging people in the UK to quit smoking”
Methodology	“The project involved discussions around how simple, unexpected, concrete, credible, emotional, motivating and how much storytelling there was within 3 different concepts”
Description of any stimulus	Concept 1: a pair of lungs made from a wilted flower on the left side, and a vibrant, colorful bouquet of flowers on the right side. The text reads “QUIT SMOKING AND BREATHE”
	Concept 2: features a handwritten note saying, “I’m not scared of spiders. I’m scared of my Mum dying from smoking,” next to a real spider. It also includes a statistic about the number of people dying each week in the UK from smoking-related diseases.
	Concept 3: shows a diseased, decomposing part of the body inside a cigarette paper, with the text: “Every cigarette rots you from the inside out.”
Overall Objectives	“The project aims to understand which of these concepts is the most motivating for quitting smoking, and how much the sticky ideas framework (simple, unexpected, concrete, credible, emotional, storytelling) impact motivation.”

​

Add Your Research Material

For live interview projects we recommend the following file types

• Audio and video files: We strongly recommend using these whenever possible for the following reasons:
  1. No potential formatting required (unlike docx / txt)
  2. We have very high accuracy transcription built in
  3. Will enable you to create clips downstream
• Text and docx transcripts: We recommend using these if you are working in a language we don’t support yet or have gold standard human written or curated transcripts available
• Spreadsheet (xlsx): These can be used to automatically apply segments or upload corresponding interview notes alongside your raw data (read more about this here)

​

Uploading Audio & Video

After uploading audio or video CoLoop will prompt you to enter…

• Estimated numbers of speakers (both moderator and respondents)
• Language or accent details (If in doubt, select US English…)
• Keywords for specific brands or products (make sure you hit the enter key after each)
• Providing these details (especially keywords!) will help to improve the transcription accuracy

​

Uploading DOCX / TXT Transcripts

• If you’re working with a language that we don’t support for transcription just yet, we recommend getting transcripts from another source (3rd party moderators, transcription services etc.).
• You must ensure these transcripts contain labelled speakers.
• Unsupported transcript formats will return the error shown below
• If you see this please double check our formatting guides or reach out to support@coloop.ai

​

Labelling your speakers

Once CoLoop finishes transcribing, label your speakers…

• You must give every speaker a unique name within the project
• You must give every speaker a role of either researcher or participant
• Giving 2 speakers the same name will cause them to be merged together, you should only do this if you are certain they are the same person!

​

Correcting transcripts

In some cases CoLoop’s transcription may mistranscribe certain words or phrases

• ‘Low confidence’ words will be highlighted in red
• You can click on these to correct them
• Alternatively you can select any text range and edit it directly
• In both cases you can click “Show all instances” to correct the word or phrase across all of your transcripts simultaneously.

​

Adding Segments

Click on the segments tab up at the top to add segments such as age, gender, location or others to your research material

• Segments can be used to filter and compare different characteristics of your research participants
• Segments can also be imported from xslx sheets or 3rd party data sources such as Recollective or Field Notes.
• If importing from xlsx files make sure your xlsx file matches the format shown below (more details here)
• You must ensure names in your xlsx sheet match up with how they appear in the project to be correctly applied.

​

Generating Analysis Grids

• Start by opening up the analysis grid and creating a new grid.
• You can create multiple grids within a project
• These are arranged like pages in Excel along the bottom row
• Analysis grids are great for identifying key themes and motifs within your data

​

How to use the analysis grids

• We recommend using the analysis grids in a hypothesis led way
• Use the suggested questions and your own intuition to test out ideas, objectives and questions
• These can be questions directly from your discussion guide or related ones
• CoLoop will summarise the main themes across all of your interviews in the overview up at the top
• Generated summaries and bullets can be checked against the supporting evidence by clicking on them

​

Arranging grids by segment

• You can also setup custom analysis grids to compare across segments
• These grids will summarise the consensus from each of the selected segments
• These are most useful for projects with multiple focus groups or strong segmentation components e.g. across markets

​

Tips when using analysis grids

Analysis grids are a great way to explore, test and validate different hypotheses about your research material but remember…

• Analysis grids provide primary analysis they will not try to interpret the information, they will just report on what is there
• The themes are suggestive, it’s your job as a researcher to explore and validate them
• Be sure to click through on the bullets and check the evidence!

​

Evidence Panels

Click on any sentence in the overview or one of the responses to see the related quotes. These are useful for finding verbatim or quickly validating a given theme or point.

1. Suggested quotes are based on the point you clicked on
2. Scroll through the list, the ‘best’ one might be lower down
3. Click “Open” to see the full context in the transcript

​

Finding quotes and further details

The AI chat is most useful in the later stage of the report once you’ve developed an outline and are starting to fill it with key details and additional evidence. It’s useful for:

• Finding specific quotes or evidence to build out a slide
• Helping you remember key details from your interviews or research material
• Summarising further details around a particular theme or point
• AI Chats will show full working, files and participants considered as well as citations for every generated point

​

Filtering by file or segment

• AI chats can be filtered by file or by segment
• This helps to improve accuracy by limiting the amount of data to be considered

Filtering by file

• Click “Specific Files” when creating a new chat to choose specific files to include
• Once you’ve selected a subset of files click create to setup a new AI chat

Filtering by segment

• When prompting the AI chat press the @ key or click on the “@ Segment” button to reference a segment
• These work like mentions in a group chat or on LinkedIn
• You can tag multiple segments in a prompt and we recommend working them into your query e.g. “What did @Group 1 participants say about x?”
• Correctly tagged segments are decorated with a green highlight.

​

Resetting the AI Chat

• The AI Chat considers the preceding message(s) when generating its response
• For this reason we recommend pressing the ‘reset’ button if you’re changing topics or simply creating a new AI chat

​

Creating Video & Audio Clips

• Click on any of the points in the AI chats to open the evidence panel
• Click the “Open” button on any of the quotes in the evidence panel
• This will open the section in the transcript the quote came from
• Select a text segment in a transcript a press create clip to jump to that section
• Click the download button at the top to save an audio or video clip of that quote

​

Recommended Guides