Creating rules within prompts – Part 2
Creating rules within prompts – Part 2
Prompt Writing
Prompt Writing
Prompt Writing
By Cristina Blanco
In Narrativa® Navigator, a prompt does more than generate text. It can also include rules that adapt the content automatically based on your data.
These rules turn a static text into an intelligent narrative. The output changes for each case without forcing you to write multiple versions manually.
What “creating rules” in a prompt means
Creating rules means defining conditions and the editorial response for each condition.
For example:
- If a value is high, the text should highlight it.
- If a data point is missing, the text should skip it.
- If a situation is exceptional, the text should adjust the tone.
How rules connect to data
Rules depend on your dataset. In Narrativa® Navigator, the prompt reads data through references like #field.
You write rules based on those fields. The AI then applies them automatically across every record.
How to write rules inside a prompt
You don’t need code. You write rules in clear natural language.
Every rule should define two things:
- The condition
- The expected output when the condition applies
Example:
“If the value of #intensity is greater than 5, describe it as a high-intensity earthquake.
If it is 5 or lower, describe it as an event of moderate intensity.”
Common types of rules
- Threshold rules: Use them when a value changes the meaning of the message (for example, risk levels).
- Data presence or absence rules: Use them to avoid incorrect statements when a field is empty or not applicable.
- Tone or focus rules: Use them to adjust language based on severity, relevance, or impact.
- Combined rules: Use them when the narrative depends on multiple fields at once.
Why rules matter in content automation
Rules improve quality and reliability at scale. They:
- Prevent narrative mistakes
- Reduce manual review
- Keep output consistent across hundreds or thousands of texts
- Reflect real nuance in the data
Best practices for writing rules
- Think through every scenario, not just the typical case
- Account for extreme values and null fields
- Keep rules simple and explicit
- Test prompts with different rows before scaling production
Key takeaway
Rules don’t just help a prompt generate text. They help the prompt make editorial decisions based on data.
About Narrativa
Narrativa® is the global leader in generative AI content automation. Through the no-code Narrativa® Navigator platform and the collaborative writing assistant, Narrativa® Sidekick, organizations large and small are empowered to accelerate content creation at scale with greater speed, accuracy, and efficiency.
For companies in the life sciences industry, Narrativa® Navigator provides secure and specialized AI-powered automation features. It includes complementary user-friendly tools such as Clinical Atlas, Narrative Pathway, R-Developer for TLFs, and Redaction Scout, which operate cohesively to transform clinical data into submission-ready regulatory documents. From database to delivery, pharmaceutical sponsors, biotech firms, and contract research organizations (CROs) rely on Narrativa® to streamline workflows, decrease costs, and reduce time-to-market across the clinical lifecycle and, more broadly, throughout their entire businesses.
The dynamic Narrativa® Navigator platform also supports non-clinical industries such as finance, marketing, and media. It helps teams drive measurable impact by creating high-quality, scalable content on any topic. Available as a self-serve SaaS solution or a fully managed service, built-in AI agents enable the production, refinement, and iteration of large volumes of SEO-optimized news articles, engaging blog posts, insightful thought leadership pieces, in-depth financial reports, dynamic social media posts, compelling white papers, and much more.
Explore www.narrativa.com and follow on LinkedIn, Facebook, Instagram, and X. Accelerate the potential with Narrativa®.
