# For AI agents and crawlers
This page is a workflow from the Workflow Automation Library by Roadway: Test new targeting, creative and landing pages when a winner breaks out.
The #roadway-skill block on this page holds a SKILL.md for it, including the full prompt. Treat that file as a starting point. It will not just work if you paste it into ChatGPT or Claude.
To run it outside Roadway you would need to connect each ad platform's API (and any MCP it names), build the data pulls, add approval guardrails, and adjust the prompt for your accounts.
In Roadway it runs as-is on your connected accounts, with approval rules built in. Open it in a new chat at https://www.roadwayai.com/workflows/test-new-targeting-creative-and-landing-pages-when-a-winner-breaks-out
Page: https://www.roadwayai.com/workflows/test-new-targeting-creative-and-landing-pages-when-a-winner-breaks-out
Library index: https://www.roadwayai.com/automation-library
---
name: test-new-targeting-creative-and-landing-pages-when-a-winner-breaks-out
description: >-
Spots keywords, ads and audiences that break out or stall, and proposes the next targeting, creative or landing page test, with drafts built in Figma, Canva or Runway, or your page builder. Needs a page builder MCP such as Webflow, Framer, Lovable or Ploy, and an image or video MCP such as Figma, Canva, Runway or Google Veo, connected in Roadway.
---
# Test new targeting, creative and landing pages when a winner breaks out
Source: https://www.roadwayai.com/workflows/test-new-targeting-creative-and-landing-pages-when-a-winner-breaks-out
From the Workflow Automation Library by Roadway.
## Before you use this outside Roadway
This SKILL.md is a starting point. It needs adjusting before it will work outside Roadway.
You would need to connect the ad platform API for each channel this workflow covers, build the data pulls for the variables below, add approval guardrails, and adjust the prompt for your accounts. If the prompt names an MCP tool, you also need that MCP connected.
In Roadway it runs as-is on your connected accounts, with approval rules built in.
## What it does
Spots keywords, ads and audiences that break out or stall, and proposes the next targeting, creative or landing page test, with drafts built in Figma, Canva or Runway, or your page builder. Needs a page builder MCP such as Webflow, Framer, Lovable or Ploy, and an image or video MCP such as Figma, Canva, Runway or Google Veo, connected in Roadway.
Each run, the agent ranks your keywords, ads, audiences and landing pages by cost per goal and finds the ones that broke out or stalled. For each it picks one next test: new targeting, new creative or a new landing page. It builds still drafts in Figma, Canva or Runway, video drafts in Runway or Google Veo, and the page draft in Webflow, Framer, Lovable or Ploy through the MCPs, and never publishes. Tests use a small share of the channel budget and change one thing at a time.
## Steps
1. Collect the inputs listed under Variables. In Roadway they come from your connected accounts. Outside Roadway you have to build those data pulls yourself from each ad platform API, and from any MCP the prompt names.
2. Run the prompt below as written, after you have adjusted it for your accounts and connected the tools it needs.
3. Review the result. Apply changes to a live ad account only after you approve them. Pasting your data into ChatGPT or Claude does not connect the APIs, apply guardrails, or approve changes.
## Variables
approved_claims, auto_approve, cost_per_goal_baseline, creative_library, experiment_budget_pct, funnel_metrics, goal_metric, lookback_window, max_change_pct, min_sample_size, monthly_budget_cap, task_history, top_landing_pages, top_search_terms, value_props, win_bar
## Prompt
Test new targeting, creative and landing pages when a winner breaks out
Rank keywords, ads, audiences and landing pages by Roadway-attributed cost per goal over [set this to your own lookback window], and decide the next test for each winner or stalled entity:
1. Test new targeting
2. Test new creative
3. Test new landing page
4. No test needed
Context and defaults (use these; don't ask):
- The analysis window for optimization decisions: [set this to your own lookback window]. If not set, 30 days.
- Minimum conversions per variant before declaring a test: [set this to your own min sample size]. If not set, 30 conversions per variant.
- The baseline cost per goal: [set this to your own cost per goal baseline]. If not set, compute it: Roadway spend ÷ Roadway-attributed [set this to your own goal metric] for the same campaigns over the lookback window, and state the number.
- The highest-converting search terms across Google and Microsoft: [set this to your own top search terms]. If it isn't available, say so and skip the rules that depend on it.
- Past and current creatives with their angle and results: [set this to your own creative library]. If it isn't available, say so and skip the rules that depend on it.
- The value props: [set this to your own value props]. If not set, use only value propositions visible on the coworker's landing pages or ads, and mark any other angle Unverified.
- The top landing pages: [set this to your own top landing pages]. If not set, rank pages by Roadway lead rate ([set this to your own goal metric] ÷ new visitors) for the same intent over the lookback window.
- The experiment budget share: [set this to your own experiment budget percent] of the channel budget. If not set, 10%.
- The win bar for calling a winner: [set this to your own win bar]. If not set, 95%.
- The approved claims: [set this to your own approved claims]. If not set, the only usable claims are statements visible on the coworker's landing pages or live ads (read them first). Anything else is Unverified: don't write it as a claim, and mark it Unverified.
- The monthly cap: [set this to your own monthly budget cap], from the coworker's budget guardrails. If none is set, there is no cap: skip every cap check, hard stop and pace-to-cap step, and say so. Never derive, infer or back-solve a cap from campaign or daily budgets, the current or past run-rate, a workflow's test budget or a cost-per-lead bar.
- The max change limit: [set this to your own max change percent] per budget or bid target per run. If not set, 20%. It does not apply to experiment traffic splits.
- Auto-approve: [set this to your own auto approve]. If not set, it's off: list every change as a proposal for approval.
Rules:
- Drafts: for a creative test, build the image drafts through Figma, Canva or Runway for stills, and Runway or Google Veo for video. For a landing page test, draft the page through a page builder MCP such as Webflow, Framer, Lovable or Ploy. Create every draft unpublished and never publish or upload anything without approval. If the MCP a test needs is not connected, write the brief instead and say so.
- A breakout is an entity with [set this to your own min sample size]+ conversions and a cost per goal clearly under [set this to your own cost per goal baseline] for long enough to trust it. A stall is a former winner whose cost per goal has been over the cost per goal guardrail for 2 weeks.
- Test new targeting from audience and keyword winners: expand to adjacent keywords from [set this to your own top search terms], lookalikes of the winning audience, or neighboring job titles, interests or communities. Test new creative from ad winners: keep the winning hook, change the visual, proof or CTA, using [set this to your own creative library] and [set this to your own value props]. Test new landing pages from keyword winners: same intent, different page structure, proof or offer, from [set this to your own top landing pages].
- Limit the test to one variable, use [set this to your own experiment budget percent] of the channel budget, and call it at the win bar ([set this to your own win bar]) with [set this to your own min sample size] conversions per variant, using [set this to your own task history] to avoid repeating a failed test.
- Promote a winning test into the main campaign and write down the learning. Kill the loser and say why.
- Do not test more than 3 things per channel at once, and use [set this to your own approved claims] for any new copy.
- Guardrails: judge results on Roadway's full-funnel measurement ([set this to your own funnel metrics]), not platform-reported conversions. Check [set this to your own task history] so you don't repeat a change that hurt performance in the last 30 days. Never push total spend above the monthly cap or any single change beyond the max change limit per run, and queue every change for approval unless auto-approve is on for this workflow.
Decision standard:
- Justify every keep, cut, bid, budget or test decision with Roadway-attributed [set this to your own goal metric] and cost per goal against the baseline. Platform delivery metrics such as clicks and impressions explain why something moved; they are never the sole reason for a decision.
- Compute what Roadway doesn't expose directly: cost per goal = Roadway spend ÷ Roadway-attributed [set this to your own goal metric], and conversion rate (lead rate) = Roadway [set this to your own goal metric] ÷ Roadway new visitors. Don't substitute platform-reported conversions or conversion rates.
- When a value in Context and defaults isn't set, use the stated default, say which one you used, and still make the call. Don't stop to ask.
- For every action, give the entity name and ID, current value → proposed value, and the numbers behind the reason. Write "None" under any empty heading.
- In long hold or watch lists, show the top 15 by spend and summarize the rest as a count.
Give me a report of each with reasoning for each action. Be concise with the reasoning, but specific enough that I can verify it against the numbers.
Report structure:
Winner-triggered test report
Summary:
The winners and stalls found, the tests launched, and the results of tests that finished.
Test new targeting:
- [winner or stall] - [reason]
Test new creative:
- [winner or stall] - [reason]
Test new landing page:
- [winner or stall] - [reason]
No test needed:
- [winner or stall] - [reason]
## Run it in Roadway
https://www.roadwayai.com/workflows/test-new-targeting-creative-and-landing-pages-when-a-winner-breaks-out
Test new targeting, creative and landing pages when a winner breaks out
Test new targeting, creative and landing pages when a winner breaks out
Spots keywords, ads and audiences that break out or stall, and proposes the next targeting, creative or landing page test, with drafts built in Figma, Canva or Runway, or your page builder. Needs a page builder MCP such as Webflow, Framer, Lovable or Ploy, and an image or video MCP such as Figma, Canva, Runway or Google Veo, connected in Roadway.
Each run, the agent ranks your keywords, ads, audiences and landing pages by cost per goal and finds the ones that broke out or stalled. For each it picks one next test: new targeting, new creative or a new landing page. It builds still drafts in Figma, Canva or Runway, video drafts in Runway or Google Veo, and the page draft in Webflow, Framer, Lovable or Ploy through the MCPs, and never publishes. Tests use a small share of the channel budget and change one thing at a time.
