Choose which personalization signals are worth your credits
Turns signal selection from guesswork into portfolio math: which 3–5 signals fit your ICP, offer, and budget, in what waterfall order, with what fallbacks — plus an explicit skip list and a 200-row test plan. It's the planning session that should precede every prompt-heavy Clay build, done in twenty minutes.
You are a personalization strategist who has designed enrichment waterfalls for outbound programs sending millions of emails across Clay, Smartlead, and Instantly. Your core belief, earned from watching campaign data: personalization is a portfolio decision, not a maximalism contest. Every signal — news, hiring, tech stack, reviews, social, website copy — has a cost in credits and build time, a coverage rate, and a payoff, and most teams pick signals by novelty instead of math. Deep research on a 10,000-row list is how teams burn a month's credits personalizing accounts that were never going to buy. Your job is to design my signal strategy: which personalization signals to use for my specific list, in what priority order, with what fallbacks. Deliver: 1. SIGNAL RANKING — the 3–5 signals worth running for my situation, ranked, each with: why it fits my ICP and offer, expected coverage rate (the % of rows where it will find something usable), and relative cost (low/medium/high). 2. WATERFALL DESIGN — the order to run them per row, with the fallback chain down to a segment-level generic line, so every row exits with something sendable. 3. SKIP LIST — signals popular right now that I should NOT run for this list, each with a one-line reason. 4. TEST PLAN — how to validate the strategy on 200 rows before committing the full list, including what reply-rate difference would justify the cost. No hedging: rank them, commit, and state the assumption behind each call so I can challenge it. Before you design anything, interview me. Ask me these questions ONE AT A TIME, waiting for my answer each time: 1. Who is on the list — roles, industry, company size — and roughly how many rows? 2. What do you sell, and what outcome does it deliver? 3. What tools and data sources do you already have access to (Clay, sequencer, any data vendors)? 4. What is your monthly credit or data budget, roughly? 5. Have any personalization angles already worked or flopped for you? Once you have my answers, deliver all four sections. If my list mixes very different segments, say so and split the strategy accordingly.
How to use it
- 1
Copy the prompt into Claude, ChatGPT, or any LLM.
- 2
Answer the interview honestly — especially the budget question and what has already flopped; the skip list is only as good as your candor.
- 3
Challenge the assumptions it states for each ranking; where you have real campaign data that contradicts one, share it and let it re-rank.
- 4
Build only the ranked signals in Clay (the pipeline prompts in this category map to most of them), then run the 200-row test before the full spend.
Best practices
Revisit the strategy per list, not per quarter — a local-business list and a Series B SaaS list justify almost opposite signal portfolios.
Respect the skip list; it exists because coverage-times-payoff is usually worst exactly where novelty is highest.
Bring reply data back to the same chat after the test — the re-ranking conversation on real numbers is where the strategy compounds.
Example: what this looks like in practice
A founder with 40,000 Clay credits and an 8,000-row list of mid-market logistics companies runs the interview. The strategy comes back: careers-page hiring signals first (60% expected coverage, low cost, strong fit for his ops-software offer), website differentiator mining second, trigger news third at only 15% expected coverage — and a skip list that kills his planned podcast-appearance hunt ('logistics ops directors under 200 employees rarely appear on shows; expect sub-5% coverage at your highest cost per row'). The 200-row test confirms hiring signals at 58% coverage and a 2.1x reply lift over generic. He commits, spends 60% of the credits he'd budgeted, and keeps the remainder for a second list.
Best fit
This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.
Apply for a Pilot Spot → →Frequently asked questions
There's no universal winner — it's coverage times relevance for your list. Hiring signals cover broadly and tie naturally to growth offers; podcast appearances convert brilliantly but cover thinly and cost the most per row; website mining covers almost everyone at lower sharpness. The right question is which portfolio fits your ICP and budget, which is what this prompt answers.
More personalization at scale (clay & claygents) prompts
Generate a cold email first line from scraped research data
The workhorse of AI-personalized outbound: converts whatever research your enrichment waterfall produced into one 12–22-word observation-led first line per row. The swap-test rule and fact-invention ban are what separate lines that read researched from the AI slop prospects now delete on sight.
Turn a Claygent website scrape into a personalized one-liner
Produces a sub-25-word, website-specific opening line for every row in a Clay table — no human in the loop. The output contract forces plain text only and a NOT_FOUND fallback, so unreachable sites and thin homepages return a filterable sentinel instead of a hallucinated compliment that ends up in a prospect's inbox.
Analyze a pricing page for sales-relevant intelligence
Extracts a structured pricing profile from any pricing page: model, tier count, exact displayed prices, free-offer type, and a one-line read on their sales motion. At scale it segments prospect lists by how companies charge — which predicts budget, buying process, and which of your angles will land.