To get the most out of your monthly usage, try starting a new chat session whenever you're moving on to a different topic or task! Because our AI looks back at your entire conversation history to provide the best answers, very long chat threads use up your token allowance much faster, even for short replies. Think of a new chat like a clean slate: it keeps the AI focused, helps you stay within your fair use limits, and ensures you have plenty of tokens left for the rest of the month.

1. Be Clear and Specific About Your Goal

State what you want Trellis to achieve, not just the data you want it to look at.

Instead of: "Show me my post performance"
Try this: "Which post types drove the highest engagement rate on Instagram last month, and what should I do more of?"

2. Provide the Right Context

Always include a date range and profile (or campaign) when relevant. The more specific you are, the more useful the output.

Try this: "Summarize the sentiment around Nike in the 'Brand Health' topic for the month of July."

3. Ask for Meaning, Not Just Metrics

Frame your prompts to get analysis and recommendations, not just raw numbers. Use words like compare and analyze, or question words like what, why, how, or should to push Trellis toward deeper analysis.

Try this: "Compare engagement rate for my profile this month vs. last month."
Try this: "Why did engagement drop on LinkedIn last week?"
Instead of: "Show impressions by network"
Try this: "Which network drove the most reach last month, and what does that tell us about where to invest?"

4. Break Down Complex Tasks

For multi-part needs, guide Trellis step by step or ask it to structure results clearly.

Try this: "Rank my posts from last month by engagement rate, identify what the top performers have in common, then suggest three content ideas based on those patterns."

5. Combine Domains for Richer Insights

Trellis can connect listening trends, inbox sentiment, and post performance in a single thread. Use this to find gaps between what's resonating externally and what you're publishing.

Try this: "What topics are trending in my listening data that I haven't been posting about?"
Try this: "What opportunities should I surface to my product team, based on my listening and inbox data?"

6. Think of Trellis as a Teammate

If you're having trouble getting the answer you want, ask Trellis to help you create a useful prompt.

Try this: "How should I ask you this question if I want to understand which campaign drove the most conversions last quarter?"
Try this: "What tools do you have access to?"

Trellis can also help you shape and position insights for different audiences:

"Help me position this insight for my director."
"Reframe this finding for my content team."
"How would I present this to my influencer marketing team?"
"How should I frame this recommendation for my creative team?"

7. Use Trellis's Capabilities

Remember Trellis can do things like sentiment analysis, create charts, and summarize messages. You can ask it to use these specific functions.

Try this: "Plot the volume and sentiment over time for Adidas in the last three months and summarize any emerging topics you find."

8. Keep It Open for Discovery

If you're looking for new patterns, ask questions that encourage Trellis to explore the data.

Try this: "Are there any new slang terms or hashtags emerging in conversations about Starbucks this month?"

Sample Prompts You Can Use Now

These prompts are designed to give you helpful, actionable answers. Pick the category that matches your goal.

Daily/Weekly Performance Monitoring

Goal	Example Prompt
Performance Snapshot	"Give me a performance snapshot for [profile] over the past 7 days — impressions, engagements, and link clicks. How does that compare to the prior 7 days?"
Period Comparison	"Compare engagement rate for [profile] this month vs. last month."
Top & Bottom Posts	"What were my top 5 and bottom 5 posts this week by engagement rate? What do the top performers have in common?"
Weekly Trend	"How did my engagement rate change this week vs. last week across all profiles? What content drove any significant shifts?"
Profile Health Check	"Which of my profiles had the biggest performance change this week? Should I be concerned about anything?"
Performance Lifts	"Calculate performance changes and percentage lifts across all my [campaigns/channels] over [selected time frame], with clear comparisons and trend summaries."
Channel Analysis	"Analyze which social networks are driving the highest engagement rate, and identify the top-performing channels by content type and audience behavior."

Content Strategy & Optimization

Goal	Example Prompt
Best Formats	"What post types (video, image, link, text, etc.) drove the highest engagement rate for [profile] in the past 30 days?"
Top Content Themes	"Rank my posts from the past month by engagement rate. Show the actual top messages. What themes or formats show up most in the top performers?"
Week-over-Week	"Compare my top-performing content from this week to last week. What's changed, and what does that suggest for my strategy?"
Underperformers	"Show me my lowest-performing posts from the past 30 days by [metric name]. What patterns do you see, and what should I stop doing?"
Scheduled Content Review	"Look at my upcoming scheduled content. Based on what's performed well recently, what would you optimize or change?"

Tag Performance Analysis

Goal	Example Prompt
Tag vs. Average	"How did posts tagged with [tag name] perform compared to my overall average this month? What drove any differences?"
Tag Deep Dive	"Show me my top 10 posts from the [tag name] tag ranked by impressions. What themes resonated most?"
Tag Retrospective	"What were the top and bottom performing posts with [tag name] tag from [start date] to [end date]? What should we do differently next time?"

Inbox Intelligence

Goal	Example Prompt
Daily Briefing	"Show me my daily inbox briefing. What are the most common themes in messages I've received today?"
Topic Deep Dive	"What questions or concerns are showing up most in my inbox about [product/topic] this month?"
Escalation Check	"Are there any patterns in negative inbox messages this week that I should escalate? If so, which ones, and how should I respond?"

Network & Profile Comparison

Goal	Example Prompt
Best Network	"Which network drove the most [key metric] last month? Show the metric for each network, and analyze the actual post copy — tell me the main themes, topics, and angles that performed best."
Head-to-Head	"Compare my performance on [network] vs. [network] over the past 30 days. Where should I be investing more effort?"
Profile Benchmarking	"How does [key metric] vary across my different profiles from the past 7 days? Which profile is over-performing or underperforming?"
Long-Range Trend	"Show me impressions by network for the past 90 days. How has the breakdown shifted over time?"

Content Ideation & Planning

Goal	Example Prompt
Idea Generation	"What are useful patterns across my top performing posts this month? Use that insight to brainstorm 5 new post ideas for me to add to my content calendar."
Listening-Led Ideas	"What themes from my listening data are getting traction right now that I haven't posted about yet? Give me 3 content ideas."
Content Series	"Analyze my best-performing content from this quarter and recommend one new content series I can launch next month."
Calendar Gaps	"Review my scheduled posts for the next two weeks and identify any gaps or opportunities based on my top-performing content."

Cross-Domain Insights

Goal	Example Prompt
Listening to Content Pipeline	"Analyze my listening data from the last 7 days and identify the top trending topics. For each trend, explain why it matters, suggest 3 post angles I could use, and recommend which network each idea fits best."
Inbox vs. Listening	"How does sentiment about [topic] compare between my inbox and listening data? What themes overlap, and where do the conversations diverge?"
Cross-Source Complaints	"Analyze my inbox messages and listening data from this month. Identify recurring complaints that appear in both sources, then summarize the common thread across them."

Tip: Always include a date range when combining inbox and listening data to keep results focused and efficient.

Brand Health & Sentiment

Goal	Example Prompt
Recurring Concerns	"What are the recurring concerns appearing in my inbox over the past month? How have they trended over time?"
Weekly Sentiment	"Summarize the sentiment in my inbox for [profile] this week. What's positive, what's negative, and what needs attention?"

Troubleshooting Results

If you're not getting the answer you want:

Ask Trellis to help rephrase: "How should I ask this question given your capabilities if I want to know [insight of interest]?"
Narrow the scope: Try limiting to a single profile, network, or shorter date range before broadening.
Ask for structure: "Please organize your results by [network / tag / time period]."
Break it up: Ask one question at a time and use Trellis's answer to inform your next prompt.

Good to Know

Hourly data is not supported. Day-level reporting is the smallest time window. For posting time recommendations, use Sprout's Optimal Send Times (OST) feature.
Listening-based prompts require Listening to be enabled for your group, with at least one listening topic configured.
Tag-based prompts require existing tags. Tags must already exist in Sprout and be actively applied to content. If no matching tags are found, Trellis will let you know.
Cross-domain queries can be resource-intensive. Always include a clear date range when querying across both inbox and listening data to stay within your usage limits.

Was this article helpful?

0 out of 0 found this helpful

Pro Tips for Writing Great Prompts for Trellis

Table of Contents

A Note on Token Usage