# Prepare an evidence-based social agency pitch. Make the proposed work specific Compare a prospect with one competitor on a common platform and period, then read its current offer and recent public news. Build a written scope of work grounded in evidence, with success measures to validate using the prospect's own data. - For: Brand, social and creative teams - Providers: 3 (Adspy.sh, Firecrawl and Serper), 6 tools - Example (Two brand lookups, one two-brand TikTok comparison, one social report, one product page and one news search. Deliverable is a written pitch brief): at most $0.38, computed from the live catalog - Updated: 2026-10-24 ## Each provider does one part No single provider covers this job. Through Agentik your AI calls each one for its part, from one connection and one prepaid balance. - **Adspy.sh** (4 tools, at most $0.37): A same-platform comparison and the observed organic evidence behind the pitch. - **Firecrawl** (1 tool, at most $0.0025): The brand's current product offer, positioning and factual constraints from its website. - **Serper** (1 tool, at most $0.01): Recent public news with dates and original sources to contextualise the proposed work. ## Give this to your agent ```text Use my connected Agentik MCP server for this job: social agency pitch. Prepare a social agency pitch for the brand and competitor I name. Compare their collected accounts on one platform over the same period and open the prospect's report for the evidence. Read the product page and one recent news search, following dates and original source references. Propose a limited scope of work tied to observed findings, with success measures that require the prospect's own data to validate. Keep uncertain comparisons uncertain. Return a written brief and tables; do not promise a native AdSpy PDF export or send outreach. Use these Agentik tools for the written steps: adspy/get-context, adspy/search-brands, adspy/compare-social, adspy/get-brand-social, firecrawl/scrape, serper/news. Reuse identifiers from actual earlier results, never the illustrative page inputs. When collection is needed, quote it separately and wait for my approval. Use the account-collection playbook. Do not invent a handle, move an account attached to another brand, or treat a waiting job as failed. Keep platform, period, observation dates, coverage, sample size, denominator, uncertainty and source links. Preserve the provider's no_clear_gap and insufficient-data conclusions. Do not infer spend, reach, conversions, ROAS or causal performance from public posts. Before running anything, inspect each tool with inspect_tool, ask me for any input you don’t have instead of inventing it, and tell me the total price. Wait for my go. At the end, tell me what Agentik charged. ``` ## You provide - Prospect and one competitor - One platform and shared period - Product-page URL, market and pitch objective ## Before this job - The brand's account is already collected, or the account-collection playbook is completed first. - Brand reports require the actual brand key and its account association; an account ID alone is not a brand key. ## What you get: a pitch brief with benchmark evidence and proposed next steps - A benchmark with sample sizes and supported verdicts - The actual offer and current sourced context - A proposed scope tied to specific evidence - Measurable next steps and unresolved questions ## How to prepare a sourced social agency pitch in Claude or ChatGPT Prices are the most each call can reserve for this example: Two brand lookups, one two-brand TikTok comparison, one social report, one product page and one news search. Deliverable is a written pitch brief. ### 1. Check the brand and available data Read the current coverage, resolve the brand and keep accounts known only through organic posts. Use the returned brand key and the platforms actually collected. - [Adspy.sh ad and organic coverage](/tools/adspy/get-context.md) `adspy/get-context` (Adspy.sh): free - [Find brands in Adspy.sh](/tools/adspy/search-brands.md) `adspy/search-brands` (Adspy.sh): at most $0.001 - [Find brands in Adspy.sh](/tools/adspy/search-brands.md) `adspy/search-brands` (Adspy.sh): at most $0.001 ### 2. Build the platform-specific benchmark Keep sample size, interval and verdict visible. The prospect's individual report is reusable when opened by the comparison under the provider contract. - [Compare brands on one social platform](/tools/adspy/compare-social.md) `adspy/compare-social` (Adspy.sh): at most $0.24 - [Adspy.sh brand organic social report](/tools/adspy/get-brand-social.md) `adspy/get-brand-social` (Adspy.sh): at most $0.12 ### 3. Read the current offer Use the brand's actual page for product facts and proposed scope constraints. - [Scrape a web page to markdown](/tools/firecrawl/scrape.md) `firecrawl/scrape` (Firecrawl): at most $0.0025 ### 4. Add sourced current context Keep publication dates and original news sources. Unverified snippets do not become established claims. - [Google News search](/tools/serper/news.md) `serper/news` (Serper): at most $0.01 Example total: at most $0.38. Each call is billed for measured usage; unused reservations are released. ## Ask it your way - Keep the research to one platform and show the evidence behind each conclusion. - Explain what the available sample cannot establish. ## Why through Agentik - Use one Agentik connection and the existing prepaid balance. - Read the report and the posts behind it in the same conversation. - Inspect the tools, approve the quote and see the amount actually charged. ## Good to know - These are observed public posts. Missing counts stay unknown; rounded counts and incomplete history stay labelled. - TikTok and YouTube engagement is per view; Instagram engagement is per follower. Keep those bases separate. - Suspected paid-reach posts are shown separately and excluded from the organic figures as the provider reports them. - Post tags may come from video, a cover and caption, or a linked ad. Keep that provenance and do not invent a missing transcript. ## When to use something else For private reach, conversions or paid account performance, use your connected account's reporting tools. Public posts provide a different and incomplete set of observations. ## Social agency pitch: questions ### What does the example cost? At most $0.38 for the example described. New account collection and any larger volume are quoted separately. You pay the settled usage; repeated or included items can cost less. ### Are AI tags billed separately? No. Available post tags are included with the post. A missing analysis stays missing, and a cover-based analysis is identified as such. ### Can the result prove business performance? No. Public engagement, posting activity and creative reuse are observed signals. They do not establish spend, conversions, profitability or causal lift. ## Keep going - [Organic social benchmark](/organic-social-benchmark.md): Two to five brands on one platform over the same period, with medians, intervals, sample sizes and the differences the data supports. - [Paid and organic brand strategy](/paid-organic-brand-strategy.md): A brand’s public ad strategy and organic activity read together, with their evidence kept separate and the observed links between posts and ads. - [Account research](/account-research.md): Their site, latest news, official filings and the profiles of the people you’ll meet, in a one-page brief with sources. - [Social tools for your agent](/tools/social.md) - [Advertising tools for your agent](/tools/advertising.md) [All playbooks](/playbooks.md) · [Connection guides](/docs/clients.md) · [Billing](/docs/billing.md) --- [HTML page](/social-agency-pitch) · [Agent index](/llms.txt)