Understanding and Maximizing Your Trellis Credits
Table of Contents
Trellis relies on a credit system to power its agentic AI capabilities, helping ensure predictable pricing and feature availability for all users. Here is a guide to understanding how credits are used and how you can get the most value from every prompt.
How Trellis Credits Work
All users receive a free monthly usage allowance of approximately 100 credits.
- Users without Trellis Plus: Your 100-credit limit resets on your monthly billing date.
- Trellis Plus for Extended Access: If your team requires more analysis, you can upgrade to Trellis Plus, which grants extended access up to a higher fair use limit. Trellis Plus costs $35 per user per month. For Trellis Plus users, your credit limit resets each month based on your subscription date.
- Note on Usage: Credits are only consumed when you use Trellis Chat or Trellis Studio. Other Sprout AI Assist experiences do not count toward your usage limits.
What Factors Impact Credit Usage?
Credit usage is primarily driven by the number of tool calls Trellis must make and the amount of data it has to process, rather than the simple length of your prompt.
| High Credit Usage | Low Credit Usage |
| Broad or Vague Questions Queries asking for general "themes" or "interesting things" without narrowing the scope will consume a high number of credits. | Narrow, Specific Questions scoped to a single network, a recent time window, or a specific metric use fewer credits. |
| Large Data Sets Queries spanning long periods (months or years) or asking Trellis to digest a large number of messages. | Specific Follow-Ups If Trellis already fetched the data for a detailed first response, asking for more detail on that data (like displaying the messages themselves) will likely be inexpensive. |
| Multi-Source Questions Questions that require pulling data from multiple sources simultaneously (e.g., an inbox and listening data, or Instagram and LinkedIn). | Clarifying Interactions When Trellis asks you a clarifying question instead of providing an answer, this costs very few credits. |
| Compounding Follow-ups Asking for a new kind of analysis within a long, existing chat thread, which causes token usage to balloon because Trellis has to review the whole thread context. | Writing a Long, Specific Prompt A long, specific prompt uses credits more efficiently than a short, vague one, as it clearly defines what Trellis should process. |
3 Essential Tips for Conserving Credits
Follow these best practices to ensure you get the maximum value from your credit allocation:
1. Be Specific in Your Prompts
Include details such as the specific topics, profiles, and date ranges you want Trellis to analyze. This dramatically reduces the need for back-and-forth clarifying questions, which conserves credits.
2. Start a Fresh Chat for New Topics
To prevent excessive credit use, start a new chat whenever you switch to a new conversation topic or goal. The longer your chat thread, the larger the context window Trellis must review before each response, which quickly burns through credits.
3. Leverage Conversation Starters and Trellis Studio
While Trellis Skills generally use more credits than a simple prompt, they are professionally tested and designed to return high-value, actionable insights, providing a better return on investment.
Examples
Not all queries are created equal. To help you get the most out of your token allocation, we've grouped queries into three classes based on complexity and resource usage.
Class 1 — Quick Queries
These are your everyday, low-cost interactions. Perfect for daily check-ins and ad-hoc questions. They return fast and use minimal tokens.
- Checking trending themes in your Listening data
- Comparing engagement across tagged campaigns
- Generating a simple sentiment trend chart
- Identifying themes driving negative sentiment
- Pulling your daily inbox briefing
- Comparing top vs. lowest performing posts
Class 2 — Standard Analysis
These queries handle meaningful analysis over moderate datasets. Best suited for weekly reviews, content planning, and campaign optimization.
- Generating post ideas from listening trends
- Identifying trending themes across your listening data
- Summarizing what messages are saying about a topic
- Monthly content performance summaries compared to prior periods
- Pulling and summarizing sentiment across a set of messages (e.g., top 100 posts on a topic)
Class 3 — Deep Analysis
These are the most resource-intensive queries: large-scale, cross-source, or multi-year analyses. Reserve them for strategic deep-dives like quarterly reports and competitive intelligence.
- Identifying common themes across a topic and analyzing trends over time
- Multi-year competitive analysis with per-year theme breakdowns
- Cross-source comparisons (e.g., inbox vs. listening data with sentiment analysis)
- Full content performance analysis spanning multiple months
Was this article helpful?
