Claude Prompt Library

30 Claude Prompts That Build Marketing Dashboards

30 copy-paste prompts

Describe the marketing metrics you track and Claude returns a working dashboard as one self-contained HTML file with realistic sample data and charts. Prompts for acquisition, campaigns, content, funnel, SEO, and spend. Not "give me a chart."

In short: This page contains 30 copy-paste ready prompts, organized into 6 categories with a description and pro tip for each. The first 5 prompts are free instantly, no signup needed. Hand-curated and tested by the AI Academy team.

Louis Corneloup
By Louis Corneloup ยท Founder, Techpresso
Last updated ยทHand-curated & tested by the AI Academy team

Acquisition

5 prompts

Channel Mix Overview Dashboard

1/30

โœจ What it does

Claude builds a channel mix dashboard for [BRAND] with spend, leads, and CAC by source on sample data as a working screen. Click the source bars, then swap the sample numbers with your own.

You are a demand-generation analyst who builds channel mix dashboards marketers can defend in a standup. <context> I need a marketing acquisition channel mix dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic sample data so I can click every chart in the chat. </context> <inputs> - Brand or product name: [BRAND] - Period: [E.G. LAST 30 DAYS] - Channels to include: [E.G. PAID SEARCH, PAID SOCIAL, ORGANIC, EMAIL, REFERRAL] - Goal event: [E.G. SIGNUPS, MQLs, PURCHASES] - Spend available per channel: [AMOUNTS OR "INVENT PLAUSIBLE ONES"] </inputs> <task> Build the dashboard with: a header showing [BRAND] and the period; a KPI row (total spend, total conversions, blended CAC, blended conversion rate); a stacked area chart of conversions over time by channel; a bar chart of spend vs conversions per channel; a channel table with spend, conversions, conversion rate, CAC, and share of conversions; and a best/worst channel callout by CAC. Generate sample data in labeled JS constants at the top and compute CAC, rates, and shares from those constants. </task> <constraints> - One self-contained HTML file; only one charting library from a CDN. - All data lives in clearly labeled JS constants near the top so I can swap in my own. - CAC and rates are computed, not hardcoded. Responsive grid, accessible contrast, green/red change arrows. </constraints> <format> Return the full HTML as an artifact, then list where to paste real channel spend and which chart to make interactive first. </format>

๐Ÿ’ก

Pro tip: Paste last month's spend and lead counts so Claude scales the sample data to your real mix instead of even splits.

Paid vs Organic Acquisition

2/30

โœจ What it does

Claude builds a paid versus organic dashboard for [PRODUCT] with volume, cost, and conversion side by side on sample data as a working screen. Click the split chart, then swap the sample numbers with last month's counts.

You are a growth analyst who compares paid and organic acquisition without mixing the two into one vanity total. <context> I need a paid versus organic acquisition dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic sample data so I can click the split in the chat. </context> <inputs> - Product name: [PRODUCT] - Period: [E.G. LAST 90 DAYS] - Paid channels: [E.G. SEARCH, SOCIAL, DISPLAY] - Organic channels: [E.G. SEO, DIRECT, UNPAID SOCIAL] - Paid spend for the period: [AMOUNT OR "INVENT"] - Goal event: [E.G. TRIAL STARTS] </inputs> <task> Build the dashboard with: a KPI row (paid conversions, organic conversions, paid share %, paid CAC); a dual-line chart of paid vs organic conversions over time; a donut of conversion share; a grouped bar of conversion rate by channel inside each bucket; and a comparison table (volume, spend, CAC, conversion rate) with paid and organic grouped. Seed labeled JS constants and compute share, CAC, and rates. Do not invent an organic CAC; show organic as $0 spend and mark CAC as n/a. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Paid CAC computed from paid spend and paid conversions only. - Clear visual split between paid and organic; responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain how paid share is computed and where to paste last month's counts. </format>

๐Ÿ’ก

Pro tip: Give Claude a real paid budget even if organic is free. The useful view is efficiency, not a fake organic CAC of zero.

Lead Source Quality Dashboard

3/30

โœจ What it does

Claude builds a lead source quality dashboard for [COMPANY] with MQL rate, SQL rate, and cost per SQL on sample data as a working screen. Click the quality table, then swap the sample numbers with your CRM export.

You are a marketing-operations analyst who ranks lead sources by quality, not volume. <context> I need a lead source quality dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic sample leads so I can click the quality table in the chat. </context> <inputs> - Company name: [COMPANY] - Period: [E.G. LAST QUARTER] - Sources: [E.G. PAID SEARCH, WEBINAR, PARTNER, ORGANIC, OUTBOUND] - Stages: [E.G. LEAD, MQL, SQL, OPPORTUNITY] - Spend per source: [AMOUNTS OR "INVENT"] </inputs> <task> Build the dashboard with: a KPI row (leads, MQL rate, SQL rate, cost per SQL); a stacked bar of stage counts by source; a scatter or grouped bar of volume vs SQL rate; a source table with leads, MQL %, SQL %, opportunities, spend, cost per lead, and cost per SQL; and a callout of the highest-volume low-quality source vs the best cost-per-SQL source. Generate sample stage counts in labeled JS constants and compute every rate and cost. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Rates and costs computed from constants. Flag any source with high volume and SQL rate below a threshold constant I can edit. - Responsive and accessible; sortable table. </constraints> <format> Return the full HTML as an artifact, then explain the quality flags and how to paste a CRM source export. </format>

๐Ÿ’ก

Pro tip: Rank sources by cost per SQL, not raw leads. A cheap form-fill source can still be the most expensive SQL.

Signup Cohort Acquisition Dashboard

4/30

โœจ What it does

Claude builds a signup cohort dashboard for [PRODUCT] with weekly cohorts, activation, and paid conversion on sample data as a working screen. Click a cohort row, then swap the sample numbers with your signup dates.

You are a growth analyst who reads acquisition through signup cohorts, not a single lifetime conversion rate. <context> I need a signup cohort acquisition dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic weekly cohorts so I can click a cohort row in the chat. </context> <inputs> - Product name: [PRODUCT] - Cohort grouping: [E.G. SIGNUP WEEK] - Number of cohorts: [E.G. LAST 8 WEEKS] - Activation event: [E.G. COMPLETED ONBOARDING] - Paid event: [E.G. STARTED PAID PLAN] - Acquisition channel mix per cohort if known: [NOTES OR "INVENT"] </inputs> <task> Build the dashboard with: a KPI row (newest cohort size, average activation %, average paid conversion %, best cohort); a cohort table (signups, activated, paid, activation %, paid %) with the newest row first; a line chart of cohort size over time; a small multiples or grouped bar of activation vs paid by cohort; and a short callout comparing the newest cohort to the oldest. Generate decaying-but-plausible cohort numbers in labeled JS constants and compute the rates. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Rates computed from counts. Do not invent revenue. Mark unknown facts as comments in the constants. - Responsive and accessible; newest cohort easy to spot. </constraints> <format> Return the full HTML as an artifact, then explain how to replace the cohort matrix with real signup dates. </format>

๐Ÿ’ก

Pro tip: Tell Claude your real activation event. A cohort heatmap is noise if week 1 means something you do not actually track.

Partner and Referral Dashboard

5/30

โœจ What it does

Claude builds a partner and referral dashboard for [BRAND] with partner volume, close rate, and payout on sample data as a working screen. Click a partner row, then swap the sample numbers with your affiliate report.

You are a partnership analyst who builds referral and affiliate dashboards. <context> I need a partner and referral acquisition dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic partner rows so I can click a partner in the chat. </context> <inputs> - Brand name: [BRAND] - Period: [E.G. LAST 90 DAYS] - Partner types: [E.G. AFFILIATE, AGENCY, INTEGRATION, CUSTOMER REFERRAL] - Partners to list: [6-12 NAMES OR "INVENT REALISTIC ONES"] - Payout model: [E.G. % OF FIRST INVOICE, FLAT PER QUALIFIED LEAD] </inputs> <task> Build the dashboard with: a KPI row (referred leads, close rate, revenue attributed, payout owed); a bar chart of leads by partner; a table of partners with type, leads, SQLs, closed-won, close rate, attributed revenue, payout, and net after payout; a type-mix donut; and a callout of the partner with the best net after payout. Generate sample partner data in labeled JS constants and compute close rate, payout, and net. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Payout and net computed from an editable formula constant. - Do not invent contract terms I did not give; mark those [VERIFY]. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the payout math and how to paste an affiliate report. </format>

๐Ÿ’ก

Pro tip: Include payout or rev-share if you have it. Volume without payout hides partners that look busy and lose money.

Campaigns

5 prompts

Multi-Channel Campaign Dashboard

6/30

โœจ What it does

Claude builds a multi-channel campaign dashboard for [BRAND] with spend, conversions, CAC, and ROAS on sample data as a working screen. Click the channel table, then swap the sample numbers with this week's spend.

You are a marketing analytics specialist who builds campaign performance dashboards. <context> I need a multi-channel campaign dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic sample campaigns so I can click the channel table in the chat. </context> <inputs> - Brand name: [BRAND] - Channels: [E.G. PAID SEARCH, PAID SOCIAL, EMAIL, SEO] - Goal metric: [E.G. SIGNUPS, PURCHASES, LEADS] - Budget for the period: [AMOUNT] - Period: [E.G. LAST 30 DAYS] - Campaign names if I have them: [LIST OR "INVENT"] </inputs> <task> Build the dashboard with: a KPI row (total spend, conversions, blended CAC, ROAS, % of budget used); a stacked area of conversions over time by channel; a bar chart of spend vs conversions per channel; a campaign table with channel, spend, impressions or sends, clicks, conversions, CTR, conversion rate, CAC, and ROAS; and a best/worst channel callout. Seed labeled JS constants and compute CAC, ROAS, CTR, and budget used. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - ROAS and CAC computed from constants, not hardcoded. - Top performer highlighted, underperformer flagged; responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then say which channel the sample data would scale and where to paste this week's spend. </format>

๐Ÿ’ก

Pro tip: Name the one goal metric (leads, trials, or purchases). Mixed goals in one ROAS column make every channel look fine.

Paid Social Campaign Dashboard

7/30

โœจ What it does

Claude builds a paid social dashboard for [BRAND] with CPM, CTR, frequency, and creative results on sample data as a working screen. Click a creative row, then swap the sample numbers with your ads export.

You are a paid social analyst who builds creative and campaign dashboards for ads managers. <context> I need a paid social campaign dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic sample ads so I can click a creative row in the chat. </context> <inputs> - Brand name: [BRAND] - Platforms: [E.G. META, LINKEDIN, TIKTOK] - Period: [E.G. LAST 14 DAYS] - Goal: [E.G. LEADS, PURCHASES, TRAFFIC] - Campaigns or creatives: [6-10 NAMES OR "INVENT REALISTIC ONES"] - Budget: [AMOUNT] </inputs> <task> Build the dashboard with: a KPI row (spend, impressions, CPM, CTR, frequency, cost per result); per-platform cards with spend and results; a line chart of spend and results over time; a creative table with platform, format, spend, impressions, CTR, frequency, results, and cost per result; and a callout of the creative to kill vs the creative to scale. Generate sample ad rows in labeled JS constants and compute CPM, CTR, frequency, and cost per result. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Metrics computed from impressions, clicks, spend, and results. - Flag frequency above an editable threshold. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the kill-or-scale flags and how to paste an ads manager export. </format>

๐Ÿ’ก

Pro tip: Ask Claude to show frequency next to CTR. High frequency and a falling CTR is usually the creative, not the audience.

Paid Search Campaign Dashboard

8/30

โœจ What it does

Claude builds a paid search dashboard for [BRAND] with clicks, CPC, conversion rate, and impression share on sample data as a working screen. Click a campaign row, then swap the sample numbers with your Google Ads export.

You are a paid search analyst who builds Google Ads style campaign dashboards. <context> I need a paid search campaign dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic search campaigns so I can click a campaign row in the chat. </context> <inputs> - Brand name: [BRAND] - Period: [E.G. LAST 30 DAYS] - Campaign types: [E.G. BRAND, NON-BRAND, COMPETITOR, PMAX] - Goal: [E.G. LEADS, PURCHASES] - Campaigns: [5-10 NAMES OR "INVENT"] - Budget: [AMOUNT] </inputs> <task> Build the dashboard with: a KPI row (spend, clicks, avg CPC, conversions, conversion rate, impression share); a spend-vs-conversions bar by campaign; a line chart of CPC and conversion rate over time; a campaign table with type, spend, clicks, CPC, conversions, conv. rate, cost per conversion, and impression share; and a callout of wasted spend (high spend, weak conversion) vs a campaign that is limited by impression share. Seed labeled JS constants and compute CPC, rates, and cost per conversion. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Brand vs non-brand visually grouped. Computed metrics only. - Responsive and accessible; impression share shown as a percent bar. </constraints> <format> Return the full HTML as an artifact, then explain wasted-spend logic and how to paste a Google Ads campaign export. </format>

๐Ÿ’ก

Pro tip: Include impression share if you have it. A cheap CPC on a campaign that cannot win the auction is not a win.

Email Campaign Dashboard

9/30

โœจ What it does

Claude builds an email campaign dashboard for [NEWSLETTER] with opens, clicks, unsubscribes, and list growth on sample data as a working screen. Click a campaign row, then swap the sample numbers with your last send.

You are an email-marketing analyst who builds campaign and list-health dashboards. <context> I need an email campaign dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic sends so I can click a campaign row in the chat. </context> <inputs> - Newsletter or brand name: [NEWSLETTER] - List size: [NUMBER OR "INVENT"] - Recent campaigns: [4-8 NAMES OR "INVENT"] - Period: [E.G. LAST QUARTER] - Metrics I track: [E.G. OPEN RATE, CTR, UNSUBSCRIBE RATE, LIST GROWTH] </inputs> <task> Build the dashboard with: a KPI row (subscribers, avg open rate, avg CTR, unsubscribe rate, net list growth); a list-growth line (gained vs lost); a per-campaign table with sends, delivered, opens, clicks, unsubscribes, open rate, CTR, and unsub rate; a bar chart comparing open and click rates; and a best-send callout plus an underperformer flag. Generate sample campaign rows in labeled JS constants and compute every rate. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Rates computed from delivered, opens, clicks, and unsubscribes. - Underperforming campaigns flagged against editable thresholds. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then name the winning send and where to paste an ESP CSV. </format>

๐Ÿ’ก

Pro tip: Put unsubscribe rate next to CTR. A high-click send that also burns the list is not the winner.

Launch Campaign War Room

10/30

โœจ What it does

Claude builds a launch war-room dashboard for [LAUNCH NAME] with daily signups, spend, and channel mix on sample data as a working screen. Click the daily chart, then swap the sample numbers with launch-week counts.

You are a launch marketer who builds a one-screen war room for a finite campaign window. <context> I need a launch campaign war-room dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic daily launch data so I can click the daily chart in the chat. </context> <inputs> - Launch name: [LAUNCH NAME] - Launch window: [START DATE] to [END DATE] - Goal and target: [E.G. 500 TRIALS] - Channels in the launch: [E.G. PAID SOCIAL, EMAIL, PARTNERS, ORGANIC] - Daily budget: [AMOUNT OR "INVENT"] - Today inside the window: [DAY NUMBER OR DATE] </inputs> <task> Build the dashboard with: a header with [LAUNCH NAME], days remaining, and percent of target; a KPI row (signups to date, spend to date, pace vs target, CAC so far); a daily bar/line of signups vs a target pace line; a channel mix stacked bar for the window; a today strip (yesterday vs today); and a short risks list (behind pace, over daily budget, one channel carrying everything). Generate daily sample series in labeled JS constants and compute pace, remaining, and CAC. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Pace compares cumulative actuals to an even or custom target curve defined in a constant. - Behind-pace and over-budget states flagged. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the pace math and where to paste launch-week daily counts. </format>

๐Ÿ’ก

Pro tip: Lock the launch window and the one goal number before you ask for the board. A war room with five goals is a status page.

Content

5 prompts

Content Performance Dashboard

11/30

โœจ What it does

Claude builds a content performance dashboard for [BLOG] with views, time on page, and conversions on sample data as a working screen. Click the top article, then swap the sample numbers with your analytics export.

You are a content-marketing analyst who ranks articles by the job they do, not vanity views. <context> I need a content performance dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic article rows so I can click the top piece in the chat. </context> <inputs> - Blog or brand name: [BLOG] - Period: [E.G. LAST 6 MONTHS] - Articles: [6-12 TITLES OR "INVENT REALISTIC ONES"] - Content types: [E.G. GUIDE, NEWS, CASE STUDY] - Conversion event: [E.G. EMAIL SIGNUP, DEMO CLICK] </inputs> <task> Build the dashboard with: a KPI row (total pageviews, avg time on page, content conversions, conversion rate); a pageviews-over-time line; a sortable article table with type, views, avg time, conversions, conversion rate, and traffic source; a bar chart of performance by content type; and a callout of the single best converting article vs the highest-view low-converting piece. Generate sample content data in labeled JS constants and compute conversion rates. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Conversion rate computed from conversions over views. - Sortable table; best converter highlighted. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then say which content type to double down on and how to paste an analytics export. </format>

๐Ÿ’ก

Pro tip: Ask Claude to rank by conversion rate, not pageviews, so a quiet high-converting guide does not hide behind a viral post.

Social Content Dashboard

12/30

โœจ What it does

Claude builds a social content dashboard for [BRAND] with reach, engagement, and top posts by platform on sample data as a working screen. Click a platform card, then swap the sample numbers with your native stats.

You are a social-media analyst who builds multi-platform content dashboards. <context> I need a social content dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic posts so I can click a platform card in the chat. </context> <inputs> - Brand name: [BRAND] - Platforms: [E.G. LINKEDIN, INSTAGRAM, X, TIKTOK] - Period: [E.G. LAST 30 DAYS] - Top posts: [6-10 OR "INVENT REALISTIC ONES"] - Metric I care about most: [E.G. ENGAGEMENT RATE, SAVES, PROFILE VISITS] </inputs> <task> Build the dashboard with: a KPI row (total reach, engagements, avg engagement rate, posts published); per-platform cards with followers, growth %, and engagement rate; a follower-growth line per platform; an engagement-rate bar by platform; and a top-posts table with platform, format, reach, engagements, and engagement rate. Generate sample data in labeled JS constants and compute engagement rate from interactions over reach. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Consistent platform colors. Growth arrows green/red. - Responsive and accessible. Do not invent follower counts I marked unknown. </constraints> <format> Return the full HTML as an artifact, then name the platform that deserves more posts and how to paste native stats. </format>

๐Ÿ’ก

Pro tip: Lead the layout with the one platform you actually post on. Equal-weight cards hide the account that pays the bills.

Newsletter Content Dashboard

13/30

โœจ What it does

Claude builds a newsletter content dashboard for [NEWSLETTER] with per-issue opens, clicks, and forwards on sample data as a working screen. Click an issue row, then swap the sample numbers with your ESP export.

You are a newsletter editor-analyst who judges issues by clicks and forwards, not open rate alone. <context> I need a newsletter content dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic issues so I can click an issue row in the chat. </context> <inputs> - Newsletter name: [NEWSLETTER] - Cadence: [E.G. WEEKLY] - Recent issues: [6-10 SUBJECT LINES OR "INVENT"] - List size: [NUMBER OR "INVENT"] - Period: [E.G. LAST QUARTER] </inputs> <task> Build the dashboard with: a KPI row (issues sent, avg open rate, avg CTR, avg forwards or shares if provided, net unsubscribes); a per-issue table with date, subject, delivered, opens, clicks, CTR, unsubscribes, and a topic tag; a line of CTR over time; a topic-mix bar if topics exist; and a callout of the issue to study (high CTR) vs the issue to retire (high unsub or low CTR). Generate sample issues in labeled JS constants and compute rates. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Rates computed from delivered/opens/clicks. If forwards are unknown, omit that KPI rather than inventing it. - Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain which issue to study and how to paste an ESP export. </format>

๐Ÿ’ก

Pro tip: Compare issues that share a topic, not only the latest send. One lucky subject line should not rewrite the editorial calendar.

Video Content Dashboard

14/30

โœจ What it does

Claude builds a video content dashboard for [CHANNEL] with views, watch time, and CTA clicks on sample data as a working screen. Click a video row, then swap the sample numbers with your video stats.

You are a video-marketing analyst who builds content dashboards for YouTube, webinars, or short-form. <context> I need a video content dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic videos so I can click a video row in the chat. </context> <inputs> - Channel or series name: [CHANNEL] - Platforms: [E.G. YOUTUBE, WEBINAR REPLAY, LINKEDIN VIDEO] - Videos: [6-10 TITLES OR "INVENT"] - CTA I track: [E.G. SIGNUP CLICKS, DEMO REQUESTS] - Period: [E.G. LAST 90 DAYS] </inputs> <task> Build the dashboard with: a KPI row (views, watch time hours, avg percent viewed, CTA clicks, CTR to CTA); a views-over-time line; a video table with title, platform, views, avg percent viewed, CTA clicks, and CTA rate; a bar of CTA rate by video; and a callout of the video that converts vs the video that only gets views. Generate sample video rows in labeled JS constants and compute CTA rate and percent viewed from inputs you invent consistently. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - CTA rate computed from CTA clicks over views. - Do not invent subscriber or RPM figures. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain how to paste platform analytics into the video array. </format>

๐Ÿ’ก

Pro tip: Include the CTA you actually ask for (signup, demo, next video). Views without a CTA column turn the board into a vanity reel.

Content-to-Lead Attribution Dashboard

15/30

โœจ What it does

Claude builds a content-to-lead dashboard for [BRAND] with first touch, last touch, and assisted conversions on sample data as a working screen. Click an attribution row, then swap the sample numbers with your assisted-conversion export.

You are a content-attribution analyst who refuses to credit only the last page before a form. <context> I need a content-to-lead attribution dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic content touches so I can click an attribution row in the chat. </context> <inputs> - Brand name: [BRAND] - Period: [E.G. LAST 90 DAYS] - Content pieces: [8-12 TITLES OR "INVENT"] - Models to show: [FIRST TOUCH, LAST TOUCH, ASSISTED] - Conversion: [E.G. MQL, TRIAL] </inputs> <task> Build the dashboard with: a KPI row (conversions, first-touch leaders, last-touch leaders, assisted total); a table of content pieces with first-touch conversions, last-touch conversions, assisted conversions, and a simple influence score (first + last + 0.5 * assisted, editable); a grouped bar of first vs last vs assisted for the top pieces; and a callout of a high-assist, low-last-touch URL that last-touch reporting would cut. Generate sample attribution counts in labeled JS constants and compute the influence score. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Influence formula in an editable constant with a comment. - Do not claim a true multi-touch model beyond the three columns I asked for. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the influence score and how to paste an assisted-conversion export. </format>

๐Ÿ’ก

Pro tip: Show assisted conversions next to last touch. A comparison post that never sits last still earns a place on the calendar.

These prompts give you the what. Tutorials give you the why.

Learn when to use extended thinking, how to build Claude Projects, and workflows that compound. 300+ tutorials and growing.

Try AI Academy Free

Funnel

5 prompts

Marketing Funnel Dashboard

16/30

โœจ What it does

Claude builds a marketing funnel dashboard for [PRODUCT] with visit, lead, MQL, SQL, and opportunity steps on sample data as a working screen. Click the biggest drop, then swap the sample numbers with your real step counts.

You are a conversion-rate analyst who builds marketing funnel dashboards. <context> I need a marketing funnel dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic step counts so I can click the biggest drop in the chat. </context> <inputs> - Product name: [PRODUCT] - Funnel steps in order: [E.G. VISIT, LEAD, MQL, SQL, OPPORTUNITY] - Period: [E.G. THIS QUARTER] - Segments if any: [E.G. PAID VS ORGANIC OR "NONE"] </inputs> <task> Build the dashboard with: a header with [PRODUCT] and the period; a horizontal funnel showing count and step-to-step conversion at each step; per-step drop-off callouts with the largest leak highlighted; a line of overall visit-to-opportunity rate over time; and a step table with count, conversion from previous, and conversion from top. Seed labeled JS constants and compute every rate. Flag the largest drop-off. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Each step shows absolute count and step-to-step percentage. - Largest drop-off visually flagged. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then point out which drop-off to investigate first and how to paste real step counts. </format>

๐Ÿ’ก

Pro tip: List the steps in the order they actually happen. A pretty funnel on the wrong sequence points at the wrong leak.

Signup to Paid Funnel

17/30

โœจ What it does

Claude builds a signup-to-paid funnel for [PRODUCT] with activation and conversion rates on sample data as a working screen. Click a funnel step, then swap the sample numbers with your product events.

You are a product-led growth analyst who builds signup-to-paid funnels. <context> I need a signup-to-paid funnel dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic product events so I can click a funnel step in the chat. </context> <inputs> - Product name: [PRODUCT] - Steps in order: [E.G. SIGNUP, ACTIVATED, STARTED TRIAL, PAID] - Activation event: [E.G. CREATED FIRST PROJECT] - Period: [E.G. LAST 8 WEEKS] - Plan or segment split if useful: [E.G. SELF-SERVE VS SALES-ASSIST] </inputs> <task> Build the dashboard with: a KPI row (signups, activation rate, trial rate, paid conversion, signup-to-paid); a funnel chart of the steps; a weekly trend of signup-to-paid; a table of step counts and rates; and a callout of the weakest step plus one diagnostic question for that step. Generate sample weekly series in labeled JS constants and compute all rates from counts. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Rates computed from the step counts. Do not invent revenue. - Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain how to replace the event counts with a product-analytics export. </format>

๐Ÿ’ก

Pro tip: Name the activation event in the inputs. Signup-to-paid without activation hides whether people never started or started and quit.

Landing Page Conversion Dashboard

18/30

โœจ What it does

Claude builds a landing page dashboard for [SITE] with sessions, bounce, and conversion rate by URL on sample data as a working screen. Click a page row, then swap the sample numbers with your landing-page report.

You are a CRO analyst who builds landing page conversion dashboards. <context> I need a landing page conversion dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic URLs so I can click a page row in the chat. </context> <inputs> - Site name: [SITE] - Period: [E.G. LAST 30 DAYS] - Landing pages: [6-12 PATHS OR "INVENT REALISTIC ONES"] - Conversion event: [E.G. FORM SUBMIT, TRIAL START] - Traffic split if known: [PAID / ORGANIC / BOTH] </inputs> <task> Build the dashboard with: a KPI row (sessions, conversions, site-wide CVR, avg bounce); a bar of conversion rate by page; a sortable page table with sessions, bounce rate, conversions, CVR, and traffic split if provided; a scatter or grouped view of sessions vs CVR so high-traffic low-CVR pages stand out; and a callout of the page to fix first (traffic times weak CVR). Generate sample page rows in labeled JS constants and compute CVR and bounce from counts where you have them. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - CVR computed from conversions over sessions. - High-traffic low-CVR pages flagged. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then name the first page to fix and how to paste a landing-page report. </format>

๐Ÿ’ก

Pro tip: Group paid and organic sessions if you have them. A page that converts paid traffic and bounces organic is two pages, not one.

Trial Activation Funnel

19/30

โœจ What it does

Claude builds a trial activation funnel for [PRODUCT] with each onboarding step on sample data as a working screen. Click the leak step, then swap the sample numbers with your activation events.

You are an onboarding analyst who builds trial activation funnels. <context> I need a trial activation funnel dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic onboarding steps so I can click the leak in the chat. </context> <inputs> - Product name: [PRODUCT] - Activation steps in order: [4-6 STEPS, E.G. SIGNUP, CONNECT DATA, INVITE TEAMMATE, FIRST VALUE] - Trial length: [E.G. 14 DAYS] - Period: [E.G. LAST 30 DAYS OF TRIAL STARTS] - Definition of activated: [ONE LINE] </inputs> <task> Build the dashboard with: a KPI row (trials started, activated, activation rate, median time to activate if you invent a consistent sample); a step funnel with counts and drop-off; a bar of drop-off at each step; a simple day-0 to day-N activation curve; and a callout of the leak step with a one-line hypothesis labeled as a hypothesis, not a fact. Generate sample step counts in labeled JS constants and compute rates. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Do not invent survey quotes or support tickets. - Leak step flagged. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain how to paste activation event counts into the constants. </format>

๐Ÿ’ก

Pro tip: Keep the steps to the five actions a new user really takes. Extra micro-steps make every leak look tiny.

Segment Funnel Compare

20/30

โœจ What it does

Claude builds a two-segment funnel compare for [FUNNEL NAME] so [SEGMENT A] sits next to [SEGMENT B] on sample data as a working screen. Click the segment toggle, then swap the sample numbers with both segments.

You are a conversion analyst who compares the same funnel across two segments. <context> I need a two-segment funnel comparison dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with two realistic segment series so I can click the segment toggle in the chat. </context> <inputs> - Funnel name: [FUNNEL NAME] - Steps in order: [4-7 STEPS] - Segment A: [SEGMENT A] - Segment B: [SEGMENT B] - Period: [E.G. THIS QUARTER] </inputs> <task> Build the dashboard with: a header and a control that focuses Segment A, Segment B, or both; side-by-side funnels or a grouped bar per step; a comparison table with counts and step-to-step rates for both segments plus a delta column; a line of overall conversion for each segment over time; and a callout of the step where the two segments diverge most. Generate sample data for both segments in labeled JS constants and compute rates and deltas. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Toggle or tabs update the focused view without a page reload. - Deltas computed, largest gap flagged. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the delta math and where to paste both segment counts. </format>

๐Ÿ’ก

Pro tip: Compare two segments you can actually change (device, channel, plan). A compare nobody can act on is a trivia chart.

SEO

5 prompts

Organic Search Overview

21/30

โœจ What it does

Claude builds an organic search dashboard for [SITE] with clicks, impressions, CTR, and average position on sample data as a working screen. Click the trend chart, then swap the sample numbers with your Search Console export.

You are an SEO analyst who builds organic-search overview dashboards from Search Console style metrics. <context> I need an organic search overview dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic search data so I can click the trend chart in the chat. </context> <inputs> - Site name: [SITE] - Period: [E.G. LAST 90 DAYS] - Metrics: [CLICKS, IMPRESSIONS, AVG POSITION, CTR] - Top queries: [6-12 OR "INVENT REALISTIC ONES"] - Top landing pages: [4-8 PATHS] </inputs> <task> Build the dashboard with: a KPI row (clicks, impressions, average position, average CTR) with vs-prior-period change; a dual-axis line of clicks and impressions; a keyword table with query, clicks, impressions, CTR, position, and position change; a top-pages table by organic clicks; and a position-distribution bar (how many queries sit 1-3, 4-10, 11+). Generate sample SEO data in labeled JS constants and compute CTR and position buckets. Color a better (lower) position green. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - CTR computed from clicks over impressions. - Position improvements green even when the number falls. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the position-bucket logic and how to import a Search Console export. </format>

๐Ÿ’ก

Pro tip: Color a rank improvement green even though the position number drops. A lower position is the win.

Keyword Ranking Tracker

22/30

โœจ What it does

Claude builds a keyword ranking dashboard for [SITE] with position, change, volume, and CTR on sample data as a working screen. Click a keyword row, then swap the sample numbers with your rank tracker.

You are an SEO analyst who builds keyword ranking trackers marketers can scan on Monday. <context> I need a keyword ranking dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic keywords so I can click a keyword row in the chat. </context> <inputs> - Site name: [SITE] - Keywords: [10-20 TERMS OR "INVENT REALISTIC ONES"] - Brand vs non-brand if known: [TAG EACH OR "INVENT A SPLIT"] - Period or last crawl: [E.G. THIS WEEK VS LAST WEEK] - Search volume if I have it: [NUMBERS OR "INVENT PLAUSIBLE"] </inputs> <task> Build the dashboard with: a KPI row (keywords tracked, average position, keywords in top 3, keywords that moved 3+ spots); a sortable keyword table with term, brand/non-brand, volume, current position, prior position, change, CTR if provided; a movement strip (gainers / losers); a position-distribution bar; and a callout of the biggest non-brand mover. Generate sample ranks in labeled JS constants and compute change and bucket counts. Lower position number is better and must show as a gain. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Do not invent rankings as facts about the live web; keep them clearly sample data. - Responsive and accessible; brand/non-brand filter if you add a simple control. </constraints> <format> Return the full HTML as an artifact, then explain how to paste a rank-tracker CSV into the keywords array. </format>

๐Ÿ’ก

Pro tip: Group brand and non-brand. A brand-term table that always sits at position 1 will hide the terms you are actually fighting for.

Landing Page SEO Dashboard

23/30

โœจ What it does

Claude builds a landing page SEO dashboard for [SITE] with organic clicks, position, and CTR by URL on sample data as a working screen. Click a page row, then swap the sample numbers with your top pages export.

You are an SEO analyst who judges landing pages by organic clicks and CTR, not only average position. <context> I need a landing page SEO dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic URLs so I can click a page row in the chat. </context> <inputs> - Site name: [SITE] - Period: [E.G. LAST 28 DAYS] - Landing pages: [8-15 PATHS OR "INVENT"] - Metrics per page: [CLICKS, IMPRESSIONS, CTR, AVG POSITION] </inputs> <task> Build the dashboard with: a KPI row (organic clicks, impressions, avg CTR, pages losing clicks); a pages table with path, clicks, impressions, CTR, position, and click change vs prior; a bar of clicks by page; a view of high-impression low-CTR pages; and a callout of the URL that needs a title/meta pass vs the URL that needs content. Generate sample page rows in labeled JS constants and compute CTR and a simple opportunity flag (high impressions, CTR below a threshold). </task> <constraints> - One self-contained HTML file; one CDN charting library only. - CTR computed. Opportunity threshold in an editable constant. - Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the opportunity flag and how to paste a top-pages export. </format>

๐Ÿ’ก

Pro tip: Sort by clicks first, then CTR. A page with impressions and no clicks is a title and meta job, not a new article.

SEO Opportunity Dashboard

24/30

โœจ What it does

Claude builds an SEO opportunity dashboard for [SITE] with rising queries, decaying pages, and gap keywords on sample data as a working screen. Click a gap row, then swap the sample numbers with your query export.

You are an SEO strategist who turns query exports into an opportunity board, not a keyword fantasy list. <context> I need an SEO opportunity dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic queries and pages so I can click a gap row in the chat. </context> <inputs> - Site name: [SITE] - Period: [E.G. LAST 90 DAYS VS PRIOR 90] - Rising queries: [OR "INVENT PLAUSIBLE ONES"] - Decaying pages: [OR "INVENT"] - Gap keywords I already rank 11-30 for: [OR "INVENT"] </inputs> <task> Build the dashboard with: three boards on one screen: rising queries (query, click change, current position), decaying pages (path, click change, position), and gap keywords (query, position 11-30, impressions, suggested existing URL); KPI chips for each board (count of risers, decliners, gaps); and a callout of the single next action (refresh a decaying URL, or retarget a gap query on a page you already have). Generate sample rows in labeled JS constants. Mark invented mappings as sample, not research. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Do not invent competitor domain metrics or backlink counts. - Responsive and accessible. Each board sortable. </constraints> <format> Return the full HTML as an artifact, then explain how to paste query and page exports into the three arrays. </format>

๐Ÿ’ก

Pro tip: Keep gap keywords tied to pages you already have. A keyword with no honest URL is a brief, not a dashboard win.

Technical SEO Health Dashboard

25/30

โœจ What it does

Claude builds a technical SEO health dashboard for [SITE] with index coverage, Core Web Vitals, and errors on sample data as a working screen. Click an error row, then swap the sample numbers with your crawl report.

You are a technical SEO analyst who builds a health dashboard from crawl and Search Console coverage style data. <context> I need a technical SEO health dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic crawl issues so I can click an error row in the chat. </context> <inputs> - Site name: [SITE] - Index coverage snapshot: [INDEXED, EXCLUDED, ERROR COUNTS OR "INVENT PLAUSIBLE"] - Core Web Vitals: [LCP, INP, CLS OR "INVENT PLAUSIBLE"] - Error samples: [4-8 URL ISSUES OR "INVENT"] - Crawl date: [DATE] </inputs> <task> Build the dashboard with: a KPI row (indexed pages, coverage errors, valid CWV % if you can compute it, open issues); a coverage donut or stacked bar (indexed / excluded / error); CWV cards with pass/fail against common thresholds in editable constants; an errors table with URL, type (4xx, 5xx, noindex, canonical), and severity; and a callout of the first fix (blocking errors before CWV polish). Generate sample issues in labeled JS constants. Thresholds must be constants I can edit. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Do not invent live crawl results for my real domain. Keep numbers in the sample constants. - Blocking errors visually above CWV. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the CWV thresholds and how to paste a crawl report. </format>

๐Ÿ’ก

Pro tip: Lead with errors that block indexing. A pretty Core Web Vitals card will not save pages Google cannot see.

Most people use 10% of Claude. Tutorials unlock the rest.

AI Academy: 300+ hands-on tutorials on Claude, ChatGPT, Midjourney, and 50+ AI tools. New tutorials added every week.

Start Your Free Trial

Spend

5 prompts

Marketing Spend vs Budget

26/30

โœจ What it does

Claude builds a marketing budget dashboard for [BRAND] with planned spend, actuals, and variance by channel on sample data as a working screen. Click the reddest line, then swap the sample numbers with this month's ledger.

You are a marketing-finance analyst who builds spend versus budget dashboards. <context> I need a marketing spend vs budget dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic planned and actual figures so I can click the variance table in the chat. </context> <inputs> - Brand name: [BRAND] - Period: [E.G. YEAR TO DATE BY MONTH] - Channels or cost centers: [E.G. PAID SEARCH, PAID SOCIAL, CONTENT, TOOLS, AGENCY] - Total marketing budget: [AMOUNT] - Currency: [E.G. USD] </inputs> <task> Build the dashboard with: a KPI row (budget, actual, variance, % of budget used); a grouped bar of budget vs actual by channel; a variance table with over/under in red/green and percent variance; a cumulative spend vs budget line; and a channel donut of actual spend. Generate sample budget and actuals in labeled JS constants and compute variance and % used. Flag over-budget rows. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Variance and % used computed from constants. - Currency formatting consistent. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then point out the biggest variance and how to paste this month's ledger totals. </format>

๐Ÿ’ก

Pro tip: Use the channel names your finance team already uses. A pretty mix that does not match the ledger will not survive the review.

Channel CAC and ROAS Dashboard

27/30

โœจ What it does

Claude builds a CAC and ROAS dashboard for [BRAND] with spend, conversions, and payback by channel on sample data as a working screen. Click the efficiency table, then swap the sample numbers with your channel costs.

You are a performance marketer who builds CAC and ROAS dashboards from spend and conversion value. <context> I need a channel CAC and ROAS dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic channel economics so I can click the efficiency table in the chat. </context> <inputs> - Brand name: [BRAND] - Period: [E.G. LAST 30 DAYS] - Channels: [E.G. PAID SEARCH, PAID SOCIAL, AFFILIATE, EMAIL] - Spend and conversions per channel: [OR "INVENT PLAUSIBLE"] - Revenue or assigned value per conversion: [AMOUNT OR "INVENT"] </inputs> <task> Build the dashboard with: a KPI row (spend, conversions, blended CAC, blended ROAS); a bar of CAC by channel; a bar of ROAS by channel with a target line from an editable constant; a table with spend, conversions, CAC, revenue, ROAS, and a scale/hold/cut label from thresholds; and a callout of the channel below the ROAS target. Generate sample channel rows in labeled JS constants and compute CAC, revenue, ROAS, and labels. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - CAC = spend / conversions. ROAS = revenue / spend. Both computed. - Thresholds editable. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the scale/hold/cut rules and how to paste channel costs. </format>

๐Ÿ’ก

Pro tip: Put the revenue or goal value you actually assign to a conversion in the inputs. ROAS on a $1 lead value is fiction.

Paid Media Pacing Dashboard

28/30

โœจ What it does

Claude builds a paid media pacing dashboard for [BRAND] with daily spend versus the period budget on sample data as a working screen. Click a day bar, then swap the sample numbers with your ads pacing export.

You are a media buyer who builds daily pacing dashboards so a budget does not dump in the last three days. <context> I need a paid media pacing dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic daily spend so I can click a day bar in the chat. </context> <inputs> - Brand name: [BRAND] - Period start and end: [DATES] - Period budget: [AMOUNT] - Channels: [E.G. SEARCH, SOCIAL] - Today: [DATE] - Daily spend so far: [OR "INVENT A SERIES"] </inputs> <task> Build the dashboard with: a KPI row (budget, spent, remaining, expected spent by today, pace status); a daily spend bar against a daily ideal pace line; a cumulative spend vs cumulative budget line; a channel split of spend to date; and a callout (ahead of pace / on pace / behind) with remaining daily run-rate needed. Generate a daily series in labeled JS constants and compute expected spent, remaining, and required daily run-rate from today to period end. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Pace math computed from budget, days elapsed, and days left. No hardcoded status labels. - Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the run-rate formula and how to paste a daily spend export. </format>

๐Ÿ’ก

Pro tip: Set the period end date. Pacing without a last day is just a spend chart.

LTV to CAC by Channel

29/30

โœจ What it does

Claude builds an LTV to CAC dashboard for [COMPANY] with ratio, payback months, and channel flags on sample data as a working screen. Click a channel bar, then swap the sample numbers with your unit economics.

You are a unit-economics analyst who builds LTV to CAC dashboards for marketing channels. <context> I need an LTV:CAC dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic channel economics so I can click a channel bar in the chat. </context> <inputs> - Company name: [COMPANY] - Channels and CAC: [E.G. PAID, ORGANIC, PARTNER WITH COSTS] - Average revenue per account: [AMOUNT OR "INVENT"] - Gross margin: [E.G. 80%] - Average customer lifespan: [E.G. 24 MONTHS] - Target LTV:CAC: [E.G. 3] </inputs> <task> Build the dashboard with: a KPI row (blended LTV, blended CAC, LTV:CAC, CAC payback months); a bar of LTV:CAC by channel with the target line; a table of channels with CAC, LTV, ratio, payback months, and a below-target flag; a payback trend if you invent a short consistent series; and a callout of any channel below the target ratio. Generate sample channel data in labeled JS constants. Compute LTV from ARPA, margin, and lifespan in a commented formula. Compute ratio and payback from those outputs. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Formula commented at the top. Ratios below target flagged red. - Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain the LTV and payback formulas so I can plug in real margins and CAC. </format>

๐Ÿ’ก

Pro tip: Use your real gross margin and a lifespan you can defend. LTV swings hard on those two inputs, so guesses make the ratio useless.

Agency and Vendor Spend Dashboard

30/30

โœจ What it does

Claude builds an agency and vendor spend dashboard for [BRAND] with retainers, media, and tools against budget on sample data as a working screen. Click a vendor row, then swap the sample numbers with your invoices.

You are a marketing-operations lead who tracks agency, freelancer, and tool spend against a budget. <context> I need an agency and vendor spend dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic vendors so I can click a vendor row in the chat. </context> <inputs> - Brand name: [BRAND] - Period: [E.G. THIS QUARTER] - Vendors: [6-12 NAMES WITH TYPE AGENCY / FREELANCER / TOOL OR "INVENT"] - Cost types: [RETAINER, MEDIA, TOOLS, OTHER] - Vendor budget: [AMOUNT] - Currency: [E.G. USD] </inputs> <task> Build the dashboard with: a KPI row (vendor budget, actual, variance, % in retainers vs media vs tools); a stacked bar of cost type by vendor; a vendor table with type, retainer, media, tools, total, and share of budget; a type-mix donut; and a callout of the vendor that is quietly the largest line. Generate sample invoices in labeled JS constants and compute totals, mix, and share. Flag any vendor over an editable share threshold. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Totals computed from line types. Currency consistent. - Do not invent contract terms I did not give. Responsive and accessible. </constraints> <format> Return the full HTML as an artifact, then explain how to paste invoice lines into the vendor array. </format>

๐Ÿ’ก

Pro tip: Split retainer, media, and tools. A blended vendor total hides a cheap agency sitting on an expensive media line.

Free tool

Prompt Optimizer

Turn a rough idea into a structured, professional AI prompt.

Try it free โ†’

Frequently Asked Questions

Yes. Each prompt asks Claude to return one self-contained HTML file with inline CSS, vanilla JS, and a charting library from a CDN, pre-seeded with realistic sample data. You get a live screen you can click in the chat, with charts and tables already populated.
Every prompt tells Claude to keep sample data in clearly labeled JS constants near the top of the file. Replace those arrays with your channel spend, Search Console rows, or ads export. The closing note also says how to wire a CSV or an API. KPIs such as CAC and ROAS recompute from the constants.
The generic dashboards page covers analytics, sales, finance, SaaS, and ops in one library. This page is only marketing: acquisition mix, campaigns, content, funnels, SEO, and spend. Use this set when the screen has to survive a marketing standup. Use the generic page when you want a kanban, a P&L, or a product heatmap.
The sample numbers are invented so the screen renders. The formulas are real. Each prompt asks Claude to compute CAC, ROAS, CTR, step-to-step conversion, or LTV:CAC from editable constants and to explain the math at the end, so you can check the logic before you paste your own rows.
Yes. It is one HTML file, so you can copy it out of the chat, save it as dashboard.html, and open it in a browser or put it on any static host. To make it live, follow Claude's closing note on swapping the sample constants for a fetch or a CSV import.

Prompts are the starting line. Tutorials are the finish.

A growing library of 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ AI tools. New tutorials added every week.

7-day free trial. Cancel anytime.