In this article:
General Social Reports FAQ
Is Social Reports a standalone product or part of Dig One?
Social Reports (AKA OneCliq) is integrated into the Dig One platform. The platform houses two complementary capabilities: Studies (validation and quantification) and Social Reports (exploration and cultural context).
How can I use OneCliq with jobs-to-be-done or demand moment frameworks?
The key is to focus on the language people actually use to describe their behaviors. For example, if you notice a trend where people say they get a “special drink” as a reward during the day, you might run queries like:
“When do people feel the need to get a special drink?”
“Why do people reward themselves with coffee or tea?”
“What trends exist around people treating themselves to a drink?”
This helps capture authentic conversations that reflect those demand moments.
It’s also useful to break down broad strategic objectives into behavioral questions. For instance, instead of asking “Should [brand] get into functional beverages?” in Social Reports, reframe it as:
“Why do people get fast food for breakfast?”
“What motivates people to choose functional or energy-boosting drinks?”
This helps capture authentic demand moments without relying on rigid frameworks
Are there groups of people harder to find with Social Reports?
Yes. For example, the super-wealthy often don’t post openly about their experiences online.
Does OneCliq filter bots, trolls, or fake content?
Yes, OneCliq uses a relevancy algorithm to remove unhelpful or harmful content (spam, bot-driven, etc.). That said, it’s also valuable to see general conversations that might influence perception even if not entirely genuine.
What happens if a report fails?
Reports fail for two main reasons: (1) too few posts/comments (threshold is ~100, sometimes 50 for niche cases), or (2) a temporary system error. In both cases, you’ll see an error message with a retry option. If retrying multiple times doesn’t work, let the team know.
Can source comments be exported to Excel for further analysis?
A list of sources with links can be found in platform so you can further explore the posts found in your query.
Writing a Query FAQ
Can I narrow my query to be region-specific?
Social media platforms do not indicate location or geographical data, so OneCliq results may not be representative of regional insights.
Although OneCliq doesn’t collect demographic or geographic metadata, you can steer the system toward market-specific conversations by:
adding geographic context directly into your prompt (i.e. “camping experiences in Montreal”)
writing queries in the target language when relevant.
These approaches help the system filter out noise and increases the likelihood of surfacing region-specific conversations.
How many comments/posts are needed to draw a report?
Onecliq requires a minimum of 100 comments to draw conclusions and build a report. We encourage combining your OneCliq results with an Dig One Idea Screen to get the most representative data.
How many questions should I include in a prompt?
OneCliq works best with one main question and up to two closely related follow-up questions. Avoid overloading a single prompt with unrelated questions — it will dilute results. It’s better to run multiple reports in parallel and review them together.
Is there a character limit for prompts?
650 characters including spaces. (Roughly 15-20 words).
If I ask multiple questions in one prompt, will that confuse my results?
It depends on how related the questions are. OneCliq works best when you keep your query focused. If the questions are closely connected (e.g., opinions on the NBA playoffs and what people are excited about), you can combine them or run them as separate but related reports.
If the questions are very different (e.g., NBA playoffs opinions, Hello Kitty activation, and All-Star weekend complaints), they should be separate reports to avoid a high-level or diluted summary. A good strategy is to run multiple focused reports in parallel.
For example:
“What are people’s opinions on the NBA playoffs?”
“What are they excited about this year’s playoffs?”
“What do fans care about with the LA Clippers?”
“What did they care about in last year’s playoffs?”
By reviewing them together, you get a more complete, nuanced picture—without losing focus in any single report.
Should I ask “why” questions?
Yes. Adding “why” often surfaces richer insights upfront, especially around motivations or frustrations. For example: “Why do people feel disappointed with the Jaguar rebrand?”
Why would I choose only certain sources instead of all of them?
Each platform offers different strengths. TikTok = Gen Z. Reddit = authentic long-form storytelling. Facebook/YouTube = older demographics. Instagram = more visual, less detailed. Many clients select all sources for balance, but narrowing can help if you want specific audiences.
Analyzing Social Reports FAQ
Does OneCliq source from content in other languages?
OneCliq matches data in the same language as the query, similar to how Google Search works (e.g., an English question pulls English content).
Why does my report reference the same post more than once?
OneCliq is designed to prioritize relevance over raw volume. The platform uses a relevancy algorithm that scores posts based on how closely they match your query, surfacing only the content that is genuinely aligned rather than pulling in a broad set of low-value noise.
It’s important to view each OneCliq report as a snapshot of the conversation, not the full universe of posts on a topic. We often recommend running multiple reports from different angles to build a more complete picture.
Can source comments be exported to Excel for further analysis?
A list of sources with links can be found in platform so you can further explore the posts found in your query.
Does OneCliq analyze video content as well as comments?
For TikTok, Instagram, and Facebook video posts, the video is analyzed for context alongside comments. For YouTube, typically only the video description is analyzed, unless no description exists.
Can results differ between platforms?
Yes. Different platforms can surface conflicting opinions — this isn’t an error, it reflects real-world differences in how communities talk online.
Can I tell if a report is based on too few comments to trust?
It depends on context. For niche topics, even 100 comments can be valuable. For broad topics, you’d want hundreds or thousands. Be cautious with quantitative charts if the sample is very small. Qualitative insights may still be useful.
How should I interpret percentages in the charts?
Percentages always refer to the dataset of that specific report—not the whole internet. For example, “21% talked about game format” means 21% of the 30,000 comments in your dataset, not all online conversations.
Methodology FAQ
Are searches and analyses private or used to train the model?
All searches and analyses are entirely private. No data from your query is used to train the model or shared externally.
Does OneCliq process videos as a source?
OneCliq analyzes video descriptions and titles from YouTube. Our algorithm also uses videos from TikTok, Facebook and Instagram to provide context and better understand the source comments and posts used in reporting.
If I run 2 of the exact same Social Reports, will the results be the same?
Re-running the same OneCliq query won’t always produce identical results, even when the inputs are unchanged. Because OneCliq uses qualitative methods to interpret and synthesize the conversation, there’s always a small degree of variability in how themes are grouped or phrased, much like any qualitative research process.
How does OneCliq differ from ChatGPT or other AI search engines?
While tools like ChatGPT are powerful for summarizing information, they are not built to systematically analyze massive volumes of raw social data. To replicate what our Social Report tool does, you would first need to manually source tens of thousands of relevant comments for every topic or report and repeat that process each time.
Even then, general AI tools are not designed to apply consistent qualitative rigor across entire datasets. They often summarize based on limited sampling rather than comprehensively reviewing and classifying every comment. That can make it difficult to assess the reliability or representativeness of the insights.
Our Social Report Tool is purpose-built to analyze millions of comments with the rigor of a trained qualitative researcher. It systematically reviews and classifies the full dataset, not just a subset, delivering structured and defensible insights in as little as 10 to 20 minutes per report.
How are emotions and topics classified?
Unlike machine learning (which needs predefined labels), OneCliq uses generative AI to identify patterns organically. This allows it to surface unique, context-specific topics and emotions in each report.
How do we handle biases like a “vocal minority” dominating results?
OneCliq reflects public perception, not a representative sample. Heavy posters may skew results. Use OneCliq to surface themes and emotions, then triangulate with surveys, interviews, or other data to validate.