> Parel documentation. Index of every page: https://parel.si/docs/llms.txt. This page in HTML: https://parel.si/docs/playbooks

# Playbooks

> Whole jobs a founder or marketer runs every week, written as prompts to paste into an agent with Parel connected. Each names the tools it uses and what a typical run costs at today's prices.

Whole jobs a founder or marketer runs every week, written as prompts to paste into your agent once Parel is connected. Each one, step by step with an example of what comes back, is on [parel.si/playbooks](https://parel.si/playbooks) ([Claude](https://parel.si/docs/quickstart-claude.md), [Claude Code](https://parel.si/docs/quickstart-claude-code.md), [Cursor](https://parel.si/docs/quickstart-cursor.md), [n8n](https://parel.si/docs/quickstart-n8n.md)). Replace the parts in square brackets with your own values. The cost of a typical run is computed from each tool's price today (₹93.93 per US dollar on 17 November 2026) times the calls it makes. Failed calls are free, so a run normally costs the figure shown or less. See [if a source bills more than expected](https://parel.si/docs/how-it-works.md#if-a-source-bills-more-than-expected). A step marked _switching on soon_ uses a tool that is decided but not yet switched on; the agent skips it until it is, and the prompt still works for the rest.

## Weekly rank report

Where you rank on Google in India for the twenty keywords that matter, and which way interest is moving.

**About ₹12.60 a run** · 24 searches · A few minutes · [Open it step by step](https://parel.si/playbooks/weekly-rank-report)

1. **Where do you rank today?** Checks the live Google ranking in India for each keyword: your position, your page, and the three sites above you. Uses Google ranking (`rank_check`), 20 × ₹0.29.
2. **Is interest growing?** Reads the search trend for each keyword, five at a time, so a slipping keyword isn’t mistaken for a lost ranking. Uses Keyword trends (`keyword_data`), 4 × ₹1.70.

**What you get:** A table of twenty keywords with your position, who is above you and the trend, plus three suggestions. Rank checks take ten to twenty-five seconds each; a run is a few minutes.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Run my weekly rank report. Site: [yourbrand.in]. Keywords: [twenty keywords, comma separated]. For each keyword, check the live Google ranking in India (location India, language en) and note my position, the URL that ranks, and the three sites above me. Then check Google Trends interest in India for the keywords, five at a time. Give me a table sorted by my position, call out anything that moved since last week if I pasted last week's table, and end with the three keywords where a small push would gain the most. Tell me the total cost at the end.
```

## AI visibility check

Does ChatGPT or Perplexity recommend your brand when a buyer asks? Ten buyer questions, both engines.

**About ₹10.40 a run** · 20 searches · Several minutes · [Open it step by step](https://parel.si/playbooks/ai-visibility-check)

1. **Does ChatGPT name you?** Asks ChatGPT ten questions the way an Indian buyer would and notes every brand it names and every site it cites. Uses AI visibility (`check_ai_visibility`), 10 × ₹0.37.
2. **Does Perplexity agree?** Asks Perplexity the same ten questions, so you see whether one engine is the problem or both. Uses Perplexity visibility (`check_perplexity`), 10 × ₹0.67.

**What you get:** A scorecard: your share of mentions against competitors, and the sites the AI answers lean on. ChatGPT answers take 20 to 40 seconds each; a run is several minutes.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Check my brand's AI visibility. Brand: [your brand]. Competitors: [two or three names]. Ask these ten questions as an Indian buyer would, to both ChatGPT and Perplexity, country IN: [ten questions, e.g. "best cold-pressed hair oil in India", "which protein bar should I buy in India"]. For each answer, record whether my brand is named, which brands are named instead, and which sources the answer cites. Summarise: how often I appear, who wins, and which cited sources I should try to be on. Tell me the total cost.
```

## Creator shortlist

Twenty Instagram creators vetted before you pay any of them: reach, real engagement, posting rhythm, red flags.

**About ₹9.60 a run** · 40 searches · About five minutes · [Open it step by step](https://parel.si/playbooks/creator-shortlist)

1. **Who are they, really?** Reads each profile: followers, following, posts, bio and whether it’s verified. Uses Instagram profile (`get_instagram_profile`), 20 × ₹0.20.
2. **Do people actually engage?** Reads their recent posts to work out real engagement, posting rhythm, how much is sponsored and what looks bought. Uses Instagram posts (`get_social_posts`), 20 × ₹0.28.

**What you get:** A ranked shortlist with engagement rate and posting rhythm per creator, and the ones to skip.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Vet these Instagram creators for a paid collaboration. Handles: [twenty handles, without the @]. For each one, read the profile (followers, following, post count, bio, verification) and the recent posts (likes, comments, captions, dates). Work out engagement rate on the last posts, how often they post, how much is sponsored, and anything that looks bought (comments out of proportion, sudden gaps). Give me a shortlist of the five best fits for [your product and audience], with one line each on why, and the ones to avoid. Tell me the total cost.
```

## What people say

This week's conversation about your category across Reddit, YouTube comments, X and the news, by theme.

**About ₹6.40 a run** · 27 searches · About five minutes · [Open it step by step](https://parel.si/playbooks/what-people-say)

1. **What do people complain about?** Reads recent posts in the subreddits where your buyers talk. Uses Reddit posts (`get_reddit`), 5 × ₹0.28.
2. **What do reviewers’ audiences ask?** Reads the comments under review and comparison videos in your category. Uses YouTube comments (`get_youtube_comments`), 10 × ₹0.28.
3. **What are the loud voices saying?** Reads the recent posts of competitors, reviewers and vocal customers on X. Uses X posts (`get_x_posts`), 10 × ₹0.18.
4. **Is anything happening in the news?** Searches the latest news on your category in India. Uses News search (`search_news`), 2 × ₹0.20.

**What you get:** Themes with real quotes, where each came from, and one recommended action.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Tell me what people are saying about [your category or brand] this week. Read the recent posts in these subreddits: [five subreddits]. Read the comments on these YouTube videos: [ten video URLs, reviews or comparisons in your category]. Read the recent posts of these X accounts: [ten handles: competitors, reviewers, loud customers]. Search the news for [your category] in India, newest first. Group what you find into themes (complaints, wishes, praise, price talk, competitor mentions), quote two or three real lines per theme, and tell me the one thing I should act on. Tell me the total cost.
```

## Competitor teardown

Five competitors side by side: what they sell and charge, how they talk, who links to them, how big they are on Instagram.

**About ₹21.25 a run** · 25 searches · About five minutes · [Open it step by step](https://parel.si/playbooks/competitor-teardown)

1. **What do they sell, and for how much?** Reads each competitor’s home page and pricing or best-sellers page. Uses Read a page (`read_page`), 10 × ₹0.57.
2. **How big are they on Instagram?** Reads each competitor’s Instagram profile. Uses Instagram profile (`get_instagram_profile`), 5 × ₹0.20.
3. **Who links to them?** Pulls each domain’s backlink summary: how many sites link to it and how strong. Uses Backlinks (`backlinks`), 5 × ₹2.71.
4. **What have they announced?** Searches recent news for each brand. Uses News search (`search_news`), 5 × ₹0.20.

**What you get:** One comparison table across the five, and two concrete moves.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Do a competitor teardown. Competitors: [five domains]. Instagram handles if you know them: [handles]. For each competitor, read the home page and the pricing or best-sellers page, read the Instagram profile, pull the backlink summary of the domain, and search the news for the brand. Then compare them in one table: positioning line, hero products and prices, offers, follower count, referring domains and rank, latest news. Finish with where I am stronger, where I am weaker, and two moves to make this month. Tell me the total cost.
```

## Stockist list

Fifty stores, gyms, salons or cafés in ten cities that could stock you, with phone numbers, and verified emails for the top ten.

**About ₹31.50 a run** · 30 searches · About five minutes · [Open it step by step](https://parel.si/playbooks/stockist-list)

1. **Who could stock you?** Searches Google Maps for the kind of place you sell to, about twenty per city. Uses Maps search (`search_maps`), 10 × ₹0.13.
2. **Who runs them?** For the top ten with a website, finds the people with emails at that domain, preferring a buyer, owner or manager. Uses Emails at a company (`domain_search`), 10 × ₹2.77.
3. **Will the email reach them?** Checks each email before you get it. Uses Verify an email (`verify_email`), 10 × ₹0.25.

**What you get:** Fifty places with phone numbers, and ten verified contacts ready to write to. Indian store domains are covered unevenly; expect emails for some of the ten, not all. A domain with nothing is free.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Build me a stockist list. I sell [your product]. Find [the kind of place: premium grocery stores, gyms, salons...] in these ten cities: [ten cities], about twenty places per city from Google Maps, with name, address, rating, review count, phone and website. Rank them by rating and reviews and keep the fifty best. For the top ten that have a website, find the people with emails at that domain, prefer a buyer, owner or manager, and verify the email before you give it to me. Give me a spreadsheet-style table and a short note on how to approach each of the ten. Tell me the total cost.
```

## Reels that work in your category

The best-performing Reels in your niche and from your competitors, their scripts word for word, what viewers ask for, and ten ideas to film next.

**About ₹9.45 a run** · 35 searches · About five minutes · [Open it step by step](https://parel.si/playbooks/reels-that-work)

1. **What is working in your category?** Searches Reels for your category and keeps the top performers by views. Uses Instagram reels search (`search_instagram_reels`), 5 × ₹0.25.
2. **What are your competitors posting?** Pulls the recent Reels of each competitor account with plays, likes and comments. Uses Instagram reels (`get_instagram_reels`), 5 × ₹0.29.
3. **What exactly do they say?** Gets the spoken words of the fifteen best Reels, so the hooks and scripts can be studied line by line. Uses Instagram reel transcript (`get_instagram_transcript`), 15 × ₹0.29.
4. **What do viewers want next?** Reads the comments on the ten best to find the questions people keep asking. Uses Instagram comments (`get_instagram_comments`), 10 × ₹0.24.

**What you get:** The five strongest hooks, three scripts adapted for your brand, and ten ranked ideas, each linked to the Reel behind it.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Research Reels for my next month of content. My category: [your category, e.g. "skincare for oily skin"]. Competitor accounts: [three to five Instagram handles, without the @]. Search Instagram Reels for my category and keep the top performers by views from the last 60 days. Pull the recent Reels of each competitor account. For the fifteen best Reels overall, get the transcript, and break each one down: the hook in the first three seconds, the structure, the call to action, length, and views compared with the account's followers. Read the comments on the ten best to find what viewers keep asking for. Then give me: the five hooks that work best, three scripts I can adapt (rewritten for my brand, not copied), and ten content ideas ranked by how likely they are to perform, each with the Reel that inspired it. Tell me the total cost.
```

## Creators who punch above their weight

Creators in your niche whose posts reach far more people than their follower count suggests: the cheapest reach you can buy.

**About ₹22.36 a run** · 84 searches · About five minutes · [Open it step by step](https://parel.si/playbooks/creators-who-outperform)

1. **Who makes content in your niche?** Finds creators by what they make, not just by name, within the follower range you set. Uses Creator search (`search_creators`), 4 × ₹0.79.
2. **How big are they, really?** Reads each profile: followers, following, post count and verification. Uses Instagram profile (`get_instagram_profile`), 40 × ₹0.20.
3. **Who reaches more than their size?** Reads the last twelve posts to work out engagement, the outlier ratio and how often a post breaks out. Uses Instagram posts (`get_social_posts`), 40 × ₹0.28.

**What you get:** Ten creators ranked by how far they reach beyond their follower count, with engagement, best post and a fair price range.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Find creators who outperform their size. Niche: [your niche, e.g. "home workouts for busy women in India"]. Platform: Instagram. Find forty creators in this niche between [5,000] and [200,000] followers. For each one, read the profile and the last twelve posts. Work out: engagement rate (likes plus comments over followers), the average views or likes per post compared with the follower count (the outlier ratio), how many posts beat three times their usual numbers, posting rhythm, and how much is sponsored. Flag anything that looks bought. Rank them by outlier ratio and engagement, and give me the top ten with one line each on why, their best post, and a rough idea of what they might charge for their size. Tell me the total cost.
```

## Your competitors’ creator deals

Which creators your competitors pay, which of those posts actually worked, and lookalike creators they haven’t found yet.

**About ₹5.79 a run** · 24 searches · About five minutes · [Open it step by step](https://parel.si/playbooks/competitor-creator-collabs)

1. **Who are they working with?** Pulls each competitor’s recent Reels and spots the ones made with creators. Uses Instagram reels (`get_instagram_reels`), 6 × ₹0.29.
2. **Which ones are they paying to boost?** Reads the ads each competitor runs on Facebook and Instagram, so you see which creator content they put money behind. Uses Meta ads of a brand (`get_meta_ads`), 3 × ₹0.28.
3. **Did it work?** Reads each creator’s profile to compare the collab with the brand’s usual numbers. Uses Instagram profile (`get_instagram_profile`), 12 × ₹0.20.
4. **Who haven’t they found yet?** Finds the accounts Instagram recommends as similar to the best performers. Uses Similar Instagram accounts (`get_instagram_similar_profiles`), 3 × ₹0.27.

**What you get:** Every creator deal your competitors ran, which worked, and ten untapped lookalikes to approach first.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Map my competitors' creator collaborations. Competitor accounts: [three Instagram handles, without the @]. For each, pull their recent Reels and posts and find the ones made with creators (tagged, collab posts, or a creator speaking on camera), and the paid ads they are running in the Meta Ad Library. For each creator you find, read the profile and work out how the collab performed against the competitor's usual numbers. Then, for the three best-performing creators, find similar accounts Instagram recommends that have not worked with any of my competitors. Give me: a table of competitor, creator, post, views and how it did against the usual; the three creators to approach first; and ten lookalikes nobody has signed yet. Tell me the total cost.
```

## What X is saying about you

Every post on X about your brand and your rivals this week, the threads worth reading, and the accounts worth replying to.

**About ₹8.28 a run** · 36 searches · A few minutes · [Open it step by step](https://parel.si/playbooks/x-listening)

1. **Who is talking about you?** Searches X for your brand, your competitors and your category, keeping posts with real engagement. Uses X search (`search_x`), 6 × ₹0.18.
2. **What is the conversation really about?** Reads the full thread of the ten busiest conversations. Uses X thread (`get_x_thread`), 10 × ₹0.46.
3. **Who is worth replying to?** Looks up the accounts behind the most-engaged posts: followers, bio and who they are. Uses X profile (`get_x_profile`), 20 × ₹0.13.

**What you get:** This week on X in themes with real posts, five replies to send today, and the accounts worth knowing.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Listen to X for me. My brand: [your brand]. Competitors: [two or three names]. Search X for posts this week mentioning my brand, my competitors, and [two or three category phrases]. Keep the ones with real engagement. Read the full thread for the ten busiest conversations. Look up the accounts behind the twenty most-engaged posts (followers, bio, who they are). Group everything into praise, complaints, questions and comparisons, quote real posts, and tell me: the five posts I should reply to today (with a suggested reply in my voice), the accounts worth building a relationship with, and anything that looks like a problem starting. Tell me the total cost.
```

## Map a market in one go

Describe a market in plain words. Get the companies in it, what each has been doing this year, and where the gaps are, with sources.

**About ₹27.32 a run** · 27 searches · A few minutes · [Open it step by step](https://parel.si/playbooks/market-map)

1. **Who is in this market?** Finds companies by describing them, not by name, from four angles. Uses Describe a company (`search_companies_by_description`), 4 × ₹0.79.
2. **What have they done this year?** Finds each company's funding, launches and moves in the news by meaning, not keyword. Uses News by meaning (`search_news_semantic`), 20 × ₹0.79.
3. **How is the market changing?** Runs a reasoning search over several phrasings and returns the passages that answer it. Uses Agentic search (`search_deep_agent`), 2 × ₹1.36.
4. **How big is it, and how fast is it growing?** Writes a short brief on size and growth from many sources, each figure with its source. Uses Research brief (`research_brief`), 1 × ₹5.64 (standard).

**What you get:** A table of the market's companies with their latest news, the three trends that matter and two openings, each with sources.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Map this market for me: [describe it, e.g. "protein snacks sold online in India"]. Find the companies that match by describing them, in four searches from different angles (big brands, new brands, private labels, imports), and keep the forty most relevant. For the twenty that matter most, find their news from the last twelve months by meaning: funding, launches, price changes, new channels. Run two deep searches on how the market is changing and who is gaining. Then write a standard research brief on the market's size and growth, with sources. Give me a table (company, what they sell, stage, latest news with link), then the three trends that matter and two places a new brand could win. Tell me the total cost at the end.
```

## Find the people worth hearing from

The people who know a topic best: who spoke about it on podcasts, who wrote the papers, who makes content on it, and where they work.

**About ₹7.90 a run** · 10 searches · About two minutes · [Open it step by step](https://parel.si/playbooks/find-the-experts)

1. **Who has talked about it?** Searches podcast episodes by what was said and notes the guests and the passage that matches. Uses Podcast search (`search_podcasts`), 3 × ₹0.79.
2. **Who has written about it?** Finds research papers on the topic and their authors. Uses Research papers (`search_scholar`), 2 × ₹0.79.
3. **Who does this for a living?** Finds people who match a description of the role, seniority and place. Uses Describe a person (`search_people_by_description`), 3 × ₹0.79.
4. **Who makes content on it?** Finds creators by what they make, with their profile and the lines that matched. Uses Creator search (`search_creators`), 2 × ₹0.79.

**What you get:** Ten people ranked by how worth hearing they are, each with the evidence linked, and three to invite first.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Find the ten people most worth hearing from on [your topic, e.g. "reducing returns in Indian e-commerce"]. Search podcast episodes on the topic and note who spoke and what they said. Find research papers on it and note the authors. Find people by describing them ([the role, seniority and place you want, e.g. "heads of operations at Indian D2C brands"]). Find creators who make content on it. Merge everyone into one list of ten, with: name, why they are worth hearing (the episode, paper or post, linked), where they work, and how to reach them publicly. Suggest which three to invite to a podcast or a panel, and why. Tell me the total cost at the end.
```

## Pick the right spots, at the right hour

Where to run a sampling stall, a pop-up or a flyer drop: places that match your crowd, how busy each one is hour by hour, and what their visitors say.

**About ₹3.57 a run** · 13 searches · About two minutes · [Open it step by step](https://parel.si/playbooks/pick-the-right-spots)

1. **Where does your crowd go?** Finds places that match a description of your buyers' haunts, with address, rating, hours and hour-by-hour busy times. Uses Describe a place (`search_places_by_description`), 3 × ₹0.79.
2. **Who actually goes there?** Reads the recent reviews of the ten best to see who visits and what they like. Uses Maps reviews (`get_maps_reviews`), 10 × ₹0.12.

**What you get:** Ten spots ranked for your crowd, the best day and hour for each, and a one-day route through the top three.

**The prompt to paste** (replace the parts in [square brackets]):

```text
Find the best spots for [what you are doing, e.g. "a weekend sampling stall for our cold brew"] in [area and city, e.g. "Bandra and Khar, Mumbai"]. Describe the places my buyers go ([e.g. "specialty cafés, brunch places and gyms popular with people in their twenties"]) and search for them by description in three searches from different angles, keeping thirty places. For each, note the address, rating and number of reviews, opening hours and its popular times: which hours on [the days you can do it] it is busiest. For the ten best, read their recent reviews to see who goes there and what they like. Give me a ranked table (place, why it fits, best day and hour, rating, address) and a plan for one day that visits the top three at their busiest hours. Tell me the total cost at the end.
```

