Lenses Best Practices
Get the most out of Lenses with these proven strategies and tips.
1. Choose the Right Persona
Selecting the right persona ensures that responses are framed for your specific workflow. Each persona adjusts the analysis style, language, and output format.
| Persona | Best For | Example Prompts |
|---|---|---|
| Default | General analysis, quick insights, data exploration | "What was Walmart's card spend in 2024?" |
| Investor | Financial analysis, market research, sector reports | "Compare Target and Costco quarterly revenue trends for the past year." |
| Marketer | Campaign analysis, audience insights, marketing ROI | "What does DoorDash's app performance look like this quarter?" |
| Consultant | Due diligence, competitive analysis, market sizing | "Size the fast-casual dining market and identify top competitors by card spend." |
You can switch personas at any time using the dropdown at the top of the chat. If you consistently use one persona, check "Remember my persona for next time" to save your preference.
2. Ask Direct Business Questions
Use this when you know what you're looking for.
Specific prompts yield specific answers. The more precise your question, the better Lenses can select the right tools and return targeted results.
Examples
- "What was Walmart's card spend in Q1 2024?"
- "Compare Target and Costco in terms of app downloads."
- "Show me Starbucks foot traffic trends over the last 6 months."
- "What does consumer spending at McDonald's look like?"
What You'll Get
Responses include formatted tables with the resulting insights, key takeaways, and metadata on the data sources used. Lenses will also suggest follow-up questions to help you dig deeper.
3. Use Research Mode for Deeper Analysis
Toggle Research Mode on when you need comprehensive, multi-layered analysis. This mode takes more time and consumes more tokens, but delivers significantly more detailed output.
When to Use Research Mode
- Earnings preparation and financial deep-dives
- Multi-entity competitive comparisons
- Market sizing and sector analysis
- Detailed entity research reports
When Standard Mode Is Enough
- Quick single-metric lookups
- Checking a specific data point
- Exploratory "what data do you have?" questions
- Simple trend checks
4. Explore Available Data
Use this when you're not sure what data is available.
If you're exploring or don't know what data exists for a particular entity or topic, try asking:
- "What data do you have for Walmart?"
- "What insights are available for Florida?"
- "Which entities are included in the Card Spend insight?"
- "What metrics can you show me for the retail sector?"
Lenses can provide more details on the available entities and insights when prompted, helping guide your next step.
5. Tips for Better Results
Be Specific
If you want precise control over the query (e.g., entity, insight, filter, aggregation), phrase your request with specifics:
"Give me total card spend for Costco in CA from Jan to Dec 2024, by month."
Use Follow-up Questions
All responses include structured data and often metadata. Lenses uses this context to answer follow-ups, so don't hesitate to ask clarifying or extending questions:
- "Break this down by state."
- "How does this compare to the same period last year?"
- "Which competitor has the highest growth rate?"
Let Lenses Handle Tool Selection
Lenses automatically orchestrates the right MCP tools based on your question and active persona. You are also able to identify the MCP tools Lenses used in a response, by looking into the thoughts it made.
Start Broad, Then Narrow
If you're new to a topic, start with a broad question and use the suggested follow-ups to refine:
- "Tell me about Walmart's performance."
- "Focus on card spend trends for 2024."
- "Compare Q1 vs Q4 by state."