AI

AI Column Types and When to Use Each

AI Column Types and When to Use Each This article describes the different AI Column modes available in TexAu and when to use each one. What an AI Column does An AI Column sends data from your table to an AI model and writes the model's response into a new column. Use it to classify, summarize, extract, score, or transform text at scale. Add an AI Column 1. Click + Add column at the right edge of your table. 2. Select AI Column. 3. Choose the AI Column type you want. Available AI Column types Text generation Use this when you want the model to write something new based on your data. Examples: - Write a personalized cold email opening line based on a person's job title and company - Summarize a company's website content into one sentence - Generate a follow-up reason based on the prospect's LinkedIn headline How to configure: 1. Write a prompt in the Prompt field. Use {{column_name}} to include column values. 2. Select the AI model (e.g., GPT-4, Claude, or Gemini). 3. Set the maximum response length. 4. Click Save. Classification Use this when you want to assign each row to one of a fixed set of categories. Examples: - Classify a job title as executive, manager, or individual contributor - Label industries as tech, healthcare, finance, or other - Identify whether an email address is personal or work How to configure: 1. Enter the categories you want to classify into. Add one per line. 2. Write the classification prompt. Include the categories in the prompt. 3. Select the AI model. 4. Click Save. The output is always one of the categories you defined. Extraction Use this when you want to pull a specific piece of information from a longer text. Examples: - Extract the company name from a person's LinkedIn headline - Pull the years of experience from a biography or job description - Extract the technology stack from a website's scraped content How to configure: 1. Describe exactly what you want to extract. 2. Specify the expected format (e.g., "Return only the company name as plain text"). 3. Select the AI model. 4. Click Save. Scoring Use this when you want to assign a numeric score to each row based on criteria you define. Examples: - Score each prospect from 1-10 based on their job title and company size - Rate how relevant a company's description is to your product - Score email subject line effectiveness based on personalization and length How to configure: 1. Define the scoring criteria in the prompt. 2. Specify the score range (e.g., 1–10). 3. Instruct the model to return only the numeric score. 4. Select the AI model. 5. Click Save. Choose the right AI model Different models trade off speed, cost, and quality: - Use GPT-4 or Claude for nuanced tasks like scoring or evaluation - Use GPT-3.5 or Gemini Flash for fast, high-volume classification or extraction tasks where quality requirements are lower The model selection appears in the column configuration. Switch models at any time. Credit cost AI Column credit cost depends on the model selected and the length of the prompt and response. Check the action details panel when configuring the column to see the estimated cost per row. Troubleshooting AI Column returns inconsistent results across rows. The prompt is too open-ended. Add specific constraints such as "Return only one word" or list the exact values you accept. Test on 10–20 rows before running on the full table. The output column contains raw JSON instead of the value I expected. Your prompt is asking the model to return structured JSON. Either parse the JSON using a formula column downstream, or rewrite the prompt to return plain text. The column returns empty responses for some rows. Check that the input columns referenced in the prompt are not empty for those rows. An empty input often causes the model to skip or return blank output. I want to use a model not listed in the dropdown. The available models depend on your AI Column configuration. Contact support to inquire about adding a specific model.

Manage Saved Prompts for the AI Column

Manage Saved Prompts for the AI Column This article explains how to create, save, edit, and reuse prompts in TexAu. What saved prompts are Saved prompts are reusable instructions for the AI Column. When you write a prompt that works well, save it so you can apply it to any table without rewriting it. Saved prompts can reference column values using the {{column_name}} syntax. Before you begin You need at least one AI Column in a table. If you have not set one up yet, read the article on AI Column types first. Create a saved prompt You can save a prompt from the AI Column configuration or from the Prompts page in Settings. From an AI Column 1. Open a table that has an AI Column. 2. Click the column header and select Edit column. 3. Enter your prompt in the Prompt field. 4. Click Save as prompt. 5. Enter a name for the prompt. 6. Click Save. From Settings 1. Go to Settings > Prompts. 2. Click New prompt. 3. Enter a name for the prompt. 4. Type the prompt text. Use {{column_name}} to reference column values. For example: Summarize the LinkedIn headline for {{First Name}} {{Last Name}}. Keep it to one sentence. 5. Click Save. Use a saved prompt in an AI Column 1. Open the table with the AI Column. 2. Click the column header and select Edit column. 3. Click Load prompt. 4. Select the saved prompt from the list. 5. Click Apply. The prompt populates the prompt field. You can edit it before saving. Edit a saved prompt 1. Go to Settings > Prompts. 2. Find the prompt and click Edit. 3. Make your changes. 4. Click Save. Editing a saved prompt does not update columns that already use it. Columns save a copy of the prompt text at configuration time. Delete a saved prompt 1. Go to Settings > Prompts. 2. Find the prompt and click Delete. 3. Confirm the deletion. Deleting a prompt does not remove it from columns already using it. Those columns retain the prompt text in their configuration. Prompt writing tips - Be specific about format. "Return only the job title, no other text" gives more predictable output than "tell me the job title". - Test on a small batch of rows before running on a full table. - Include expected output format when the output column is used downstream. For example: "Return a JSON object with keys: seniority_level, department." - Keep prompts concise. Shorter prompts are faster and cheaper to run. Troubleshooting The {{column_name}} placeholder is not populating. Check that the column name in the placeholder exactly matches the header in the table, including capitalization and spaces. Column names are case-sensitive in prompt templates. Saved prompts are not appearing in the Load prompt list. Refresh the page. If the prompt was just created, it may take a moment to appear. I cannot find the Prompts option. Prompts is located under Settings > Prompts (click the gear icon in the sidebar, then select the Prompts tab). AI Column features are available on specific plans. Check Settings > Billing to confirm your plan includes AI Column access.

Using AI Enrichment in TexAu

https://www.loom.com/share/4ba8d173d75043b1830276da0c7ca4b2 Using AI Enrichment in TexAu AI enrichment in TexAu lets you enrich your data without endless manual work. Whether you're generating personalized outreach messages, summarizing company information, classifying leads, or scoring data quality, TexAu's AI integration makes it simple. This guide covers everything you need to know about using AI to enrich your data. What Can You Do With AI Enrichment? AI enrichment opens up a world of possibilities. Here are some common use cases: Personalized Messaging: Generate custom email openers or LinkedIn connection requests tailored to each prospect. Lead Scoring: Ask AI to score leads on quality, fit, or readiness to purchase based on their profile data. Company Research Summaries: Turn raw company information into concise, actionable summaries for your team. Data Classification: Automatically categorize leads by industry, company size, or other attributes. Insight Extraction: Pull key information from messy text fields or company descriptions. Sentiment Analysis: Understand sentiment from customer reviews, feedback, or social posts. The possibilities are truly endless once you get comfortable with the workflow. How AI Enrichment Works in TexAu Here's the basic flow: 1. You create an AI column in your table 2. You write a prompt that tells AI what to do 3. You can reference other column values in your prompt (like {{firstName}} or {{companyName}}) 4. TexAu sends your data + prompt to the AI model you've chosen 5. The AI processes it and returns results 6. Those results appear in your table Once you set it up, it runs automatically on new rows or whenever you trigger it. Setting Up an AI Column Let's walk through creating your first AI enrichment column. Step 1: Create a New Column In your table, click the + Add Column button. Choose AI Enrichment as the column type. Step 2: Choose Your AI Model TexAu supports three AI providers, each with different models: OpenAI: GPT-4, GPT-4 Turbo, and GPT-3.5 Turbo. GPT-4 is the most capable but also the most expensive. Great for complex reasoning, detailed writing, and nuanced analysis. Anthropic (Claude): Claude 3 Opus, Claude 3 Sonnet, and Claude 3 Haiku. Claude models excel at nuanced writing, long-form content, and detailed analysis. Opus is the most capable; Haiku is the fastest and most affordable. Google Gemini: Gemini Pro is Google's multimodal model. Good all-around performance with competitive pricing. Tip: If you're not sure which model to start with, Claude Sonnet or GPT-4 Turbo are excellent middle-ground choices. They balance quality, speed, and cost well. Your available models depend on your TexAu plan. Premium plans unlock access to the most advanced models. Standard plans get access to solid workhorse models like GPT-3.5 Turbo and Claude Haiku. Step 3: Write Your Prompt Your prompt tells the AI exactly what to do with each row of data. Give your AI column a clear name (e.g., "Personalized Email Opener" or "Lead Quality Score"). Then write your prompt in the prompt editor. Here's an example: Write a personalized 2-sentence LinkedIn connection request for {{firstName}} who works at {{companyName}} as a {{jobTitle}}. Make it friendly and mention something specific about their company. Notice how we're using {{columnName}} syntax to reference other columns? That's how you make your prompts dynamic and personal to each row. Step 4: Test Your Prompt Before running on all your data, test your prompt with a few sample rows. This helps you see if the results are what you're looking for. If not, refine your prompt and try again. Step 5: Run Your Enrichment Once you're happy with the results, you can run the enrichment on your entire table or just new rows going forward. TexAu will process each row through your AI model and populate the column with results. Choosing the Right AI Model With three different providers and multiple models, you might wonder which one to choose. Here's a breakdown: When to Use OpenAI (GPT-4 / GPT-3.5 Turbo) Best for: Structured data extraction, complex reasoning, technical writing. Trade-off: Can be pricier than alternatives. GPT-4 is slower but more capable. GPT-3.5 Turbo is faster and more affordable. Use GPT-4 when you need the absolute best results on complex tasks. Use GPT-3.5 Turbo when you want solid quality at a good price point. When to Use Anthropic Claude (Claude 3 Opus / Sonnet / Haiku) Best for: Long-form content, nuanced writing, personalized messaging, creative tasks. Trade-off: Slower than some alternatives. Claude Haiku is very affordable. Claude shines when you want thoughtful, well-written outputs. If you're generating outreach messages or detailed summaries, Claude often produces more natural-sounding results. Haiku is surprisingly capable for its price, making it great for bulk enrichment on a budget. When to Use Google Gemini Best for: General-purpose tasks, good balance of speed and quality, competitive pricing. Trade-off: Slightly less specialized than OpenAI or Claude for certain tasks. Gemini is an excellent all-rounder. If you're not sure which model to pick and want a solid, cost-effective option, Gemini is a great choice. Pro tip: Try different models on small batches of data. You'll quickly develop a sense for which one produces results you like best. Your needs and preferences matter most. Writing Effective Prompts Good prompts are the key to great AI enrichment. See the Writing Effective Prompts guide for a full walkthrough, but here are the essentials: Be Specific: Instead of "write a message," try "write a friendly, 3-sentence LinkedIn connection request that mentions their recent company news." Provide Context: The more your prompt tells AI about the situation, the better the results. Use Column Variables: Reference {{firstName}}, {{companyName}}, {{jobTitle}}, etc., to make your enrichment personal and relevant to each row. Set Output Format Expectations: If you want a number, say "respond with only a number from 1-10." If you want JSON, ask for it. Clear formatting = cleaner results. Keep It Conversational: Write prompts in plain English. You don't need fancy phrasing. "List three talking points" works great. Understanding AI Credit Costs Every time you run an AI enrichment, TexAu consumes AI credits from your plan's monthly allocation. The cost depends on a few factors: Model Choice: Advanced models (like GPT-4 or Claude Opus) cost more than budget models (like GPT-3.5 Turbo or Claude Haiku). Prompt + Data Size: Longer prompts and more input data = higher cost. Typically, enriching a single row costs anywhere from a fraction of a cent to a few cents, depending on your model. Output Length: If you ask AI to write a long essay, it costs more than asking for a one-word classification. When you create an AI column, TexAu estimates the cost per row and per 1,000 rows. You can see these estimates before you start, so there are no surprises. Budget tip: If you're enriching thousands of rows, start with a budget model like Claude Haiku or GPT-3.5 Turbo. You'll stretch your credits much further. Tips for Getting Better AI Results Even with the same prompt, tweaks and iterations can significantly improve results. Here are some best practices: 1. Be Explicit About What You Want Instead of: "Summarize this company." Try: "Write a 2-3 sentence summary of what {{company}} does, focusing on their primary product and target customer." Specificity wins every time. 2. Use Examples in Your Prompt Showing AI what good output looks like helps it match your style: Generate a lead quality score from 1-10 for {{firstName}}. Use this scale: - 1-3: Not a fit (wrong industry, too small, etc.) - 4-6: Possible fit (could be relevant) - 7-10: Strong fit (high priority to contact) Answer with only the number. 3. Test Before Running at Scale Always test your prompt on 5-10 sample rows first. Spot-check the results. It takes 30 seconds and saves you credits and time. 4. Iterate Based on Results If results aren't quite right, adjust your prompt. Maybe be more specific, add examples, or ask for a different output format. Small tweaks often lead to big improvements. 5. Reference Relevant Columns If you're enriching with data from multiple columns (like {{firstName}}, {{company}}, {{lastInteraction}}), include all relevant context. More context = better, more personalized results. 6. Keep It Simple You don't need a novel-length prompt. Clear, concise instructions usually work best. If your prompt is taking forever to write, simplify it. What's Next? Ready to write your first AI enrichment prompt? Head over to the Writing Effective Prompts guide for detailed use cases and prompt templates. Or, jump straight to setting up your first AI column in TexAu.

Writing Effective Prompts for AI Enrichment

Writing Effective Prompts for AI Enrichment A great prompt is the difference between mediocre AI results and results that genuinely move the needle for your business. The good news? Learning to write prompts that work is surprisingly straightforward once you understand a few key principles. Think of your prompt as a conversation with a very smart colleague who's willing to help but needs clear instructions. The more specific and thoughtful you are, the better your colleague (the AI) can help you. Prompt Writing Best Practices Before we get into specific examples, here are the fundamentals of effective prompt writing: 1. Be Specific Vague prompts produce vague results. Always tell AI exactly what you want. Weak prompt: "Write about the company." Strong prompt: "Write a 2-sentence summary of {{company}}'s main products and target market. Focus on what makes them unique. Keep it under 50 words." The strong version tells AI the length, focus, and format. No guessing. 2. Provide Context The more background information you give, the better results you'll get. If you're writing a message, tell AI who it's for and what you want to achieve. Weak prompt: "Write an email." Strong prompt: "Write a friendly, professional email to {{firstName}}, who works in {{department}} at {{company}}. We're reaching out to offer {{productName}}. The email should address a pain point specific to their industry and include a clear call to action. Keep it to 3-4 paragraphs." Context helps AI understand the situation and adjust its tone and content accordingly. 3. Use Column Variables Liberally Column variables like {{firstName}}, {{company}}, {{industry}} make your enrichment personal and dynamic. They're the secret sauce that transforms generic outputs into relevant, targeted results. Without variables: "Write a message." With variables: "Write a personalized message to {{firstName}} at {{company}}. They work in {{industry}} and their company size is {{employeeCount}}. Mention something relevant to {{industry}} businesses." Variables turn a one-size-fits-all prompt into 1,000 personalized enrichments. 4. Set Clear Output Format Expectations Tell AI exactly what format you want. Numbers, JSON, bullet points, paragraphs. Be explicit. Unclear: "Tell me if this lead is good." Clear: "Rate this lead from 1-10. Answer with only the number. Use this scale: 1-3 (not qualified), 4-7 (might work), 8-10 (very qualified)." Clear format expectations = cleaner, more usable results. 5. Use System Prompts vs. User Prompts In TexAu, you're writing a "user prompt," the specific request for each row. Think of system prompts as standing instructions that would apply to all requests (e.g., "You are a sales expert. Be concise. Use simple language."). In TexAu, we handle system instructions for you. Your user prompt should focus on the specific task at hand. 6. Keep It Conversational You don't need to use fancy language or technical jargon. Write like you're talking to a smart colleague. AI understands natural, conversational English perfectly well, and often prefers it. Stiff: "Enumerate three salient characteristics of the aforementioned entity's operational paradigm." Natural: "List three key things about how {{company}} operates." Simple, conversational prompts often produce better results. Good vs. Bad Prompt Examples Let's look at real side-by-side comparisons to see how these principles play out: Example 1: Lead Scoring Bad Prompt: Score this lead. Why It's Bad: No context, no scoring criteria, no format, no guidance on what makes a good lead. Good Prompt: Score this lead on a scale of 1-10 based on fit for our {{productName}} solution. Company: {{company}} Industry: {{industry}} Employee Count: {{employeeCount}} Use this scale: - 1-3: Poor fit (wrong industry, too small, not a buyer) - 4-6: Possible fit (could use our solution) - 7-10: Excellent fit (high-priority prospect) Answer with only the number. Why It's Better: Specifies criteria, provides context, sets format, gives examples of scoring tiers. Example 2: Personalized Messaging Bad Prompt: Write an outreach message. Why It's Bad: Who is it for? What's the goal? What's the tone? Way too open-ended. Good Prompt: Write a friendly 2-sentence LinkedIn connection request for {{firstName}}, who works as a {{jobTitle}} at {{company}}. Research shows {{company}} is in the {{industry}} space. Mention something specific about their company or industry to show you've done your homework. Keep it warm and genuine. No sales pitch. Why It's Better: Names the recipient, specifies length, asks for a specific detail, sets the tone clearly. Example 3: Company Research Bad Prompt: Tell me about {{company}}. Why It's Bad: Vague output, no length guidance, no focus on what matters to your business. Good Prompt: Write a 3-4 sentence summary of {{company}} for a sales rep preparing for an outreach call. Focus on: 1. What {{company}} does and their main products 2. Who their likely target customers are 3. Any recent news or trends in their industry Keep it concise and actionable. Avoid buzzwords. Why It's Better: Specific length, focus areas, intended audience, and tone all clarified. Common Prompt Patterns You Can Use Here are battle-tested prompt patterns from our most successful customers. Copy, customize, and use them as a starting point: Pattern 1: Lead Scoring Score {{firstName}} from {{company}} on fit for our solution (scale 1-10). Context: - Their industry: {{industry}} - Company size: {{employeeCount}} employees - Their role: {{jobTitle}} - Budget indicator: {{budgetLevel}} Scoring scale: - 1-3: Not a fit - 4-6: Could be a fit - 7-10: Strong fit for our solution Answer with only the number (1-10). Pattern 2: Personalized Email Open Write a 2-3 sentence email opening for {{firstName}} at {{company}}. Their role: {{jobTitle}} Their industry: {{industry}} Key detail: {{companyFocus}} Make it personalized and specific to their situation. Avoid generic language. Show that you've done your research. Start with the opening line only. Don't include greeting or sign-off. Pattern 3: Company Summary for Sales Summarize {{company}} for a sales rep in 3 sentences. What they do: - Primary product/service: {{productType}} - Target market: {{targetMarket}} - Company size: {{employeeCount}} Focus on: What they do, who they sell to, and what makes them stand out in their market. Be concise and practical. Pattern 4: Lead Classification Classify {{firstName}} from {{company}} into ONE of these categories: - Hot Lead: Actively buying, budget available, urgent - Warm Lead: Interested, in research phase - Cold Lead: Potential fit, early awareness stage Context: - Their role: {{jobTitle}} - Industry: {{industry}} - Recent activity: {{lastInteraction}} Answer with only the category name. Pattern 5: Sentiment Analysis Analyze the sentiment of this customer feedback: "{{feedback}}" Rate sentiment as one of: Positive, Neutral, Negative Then provide a 1-sentence summary of the main point. Format your response as: Sentiment: [One word] Summary: [One sentence] Pattern 6: Data Extraction & Cleaning Extract the key information from this text: "{{rawText}}" Return as comma-separated values: [Company Name], [Industry], [Headcount Range], [Contact Person Name] If any information is not clearly stated or cannot be determined, write "Unknown". Pattern 7: Insight Generation Based on {{company}}'s profile, identify ONE key opportunity for our solution: - Company: {{company}} - Industry: {{industry}} - Size: {{employeeCount}} employees - Current tools: {{currentTools}} What's their biggest pain point that our product solves? Be specific and business-focused. One sentence only. Testing & Iterating on Your Prompts Great prompts aren't born perfect. They're refined through testing. Here's our process: Step 1: Write Your First Draft Don't overthink it. Write your initial prompt with the best practices in mind. Step 2: Test on Sample Rows Create your AI column and test it on 5-10 representative rows. Run the enrichment and review the results. Ask yourself: - Are the outputs what I expected? - Is the tone right? - Is the format clean and usable? - Did the AI understand what I wanted? Step 3: Spot Check & Analyze Look at 3-5 outputs closely. Are they good? Are there patterns in what works and what doesn't? Step 4: Refine Your Prompt Based on what you see, adjust your prompt. Maybe: - Be more specific about tone or format - Add an example of what good output looks like - Simplify the instruction (you asked for too much) - Add context the AI might have missed - Clarify scoring criteria or definitions Step 5: Test Again Run it on your sample rows again. Does it improve? Step 6: Scale It Up Once you're happy, run it on your full dataset or set it up to run on new rows automatically. Testing tip: Save your test results and notes. Over time, you'll develop intuition for what works. You might even build a library of prompts that work amazingly well for your specific business. Prompt Templates You Can Customize We've built a collection of ready-to-use prompt templates in TexAu. When creating a new AI column, check out our Template Library for prompts designed for common tasks. You can copy them and customize with your own column names and business details. Look for templates in these categories: - Sales & Outreach - Lead Scoring & Qualification - Company Research - Data Classification - Sentiment & Feedback Analysis - Content Generation - Insight & Opportunity Identification Common Mistakes to Avoid Mistake 1: Asking for Too Much in One Prompt Avoid: "Write an email, score the lead, and summarize the company." Instead: Create separate AI columns for each task. It's cleaner and often produces better results. Mistake 2: Forgetting to Mention Column Names in Examples If you reference {{jobTitle}} in your prompt, make sure you actually have that column in your table. Typos or non-existent columns break things. Mistake 3: Being Too Vague About Tone AI can sound robotic, overly casual, or off-brand if you don't set clear tone expectations. Always specify: "professional but friendly," "formal and technical," "casual and conversational," etc. Mistake 4: Ignoring Output Format If you want a number or JSON or bullet points, say so explicitly. Don't assume AI will guess. Mistake 5: Not Testing Before Running at Scale Test on 5 rows before running 5,000 rows. A small prompt tweak saves thousands of credits. Pro Tips from Our Best Users Here's what the customers who get the best results from AI enrichment do differently: Tip 1: They write prompts for their use cases, not generic ones. Specificity to your business > generic prompts. Tip 2: They iterate constantly. They're not afraid to tweak and refine. Each iteration gets better. Tip 3: They use variables liberally. The more personalization, the better the engagement. Tip 4: They start simple. They write clear, concise prompts before trying complex ones. Tip 5: They document what works. They keep notes on which prompts produce which results so they can reuse and build on them. What's Next? Ready to write your first prompt? Head to your TexAu workspace and create an AI column. Use the patterns and examples above as your starting point. Need help with a specific use case? Check out our Using AI Enrichment guide for detailed walkthroughs of popular AI enrichment scenarios. Happy prompting!