Executives don't need more dashboards — they need answers. "How are we doing?" shouldn't require logging into four platforms and cross-referencing spreadsheets. This recipe builds an AI-powered daily briefing that pulls from every critical system, compares against historical trends stored in persistent memory, and delivers a concise summary that highlights what changed and what needs attention.
What You'll Need
- Google Analytics 4 connected to your website
- Google Ads with active campaigns
- Stripe processing payments
- Salesforce with pipeline data
- Memory integration enabled (for trend tracking)
- All five integrations connected in Pipeworks
The Workflow
Pull website traffic and conversion data
The agent queries Google Analytics for yesterday's metrics: total sessions, unique users, traffic by source/medium, top landing pages, conversion rate, and goal completions. It also pulls real-time data to show current active users and trending pages. Key segments are broken out — organic vs. paid vs. direct vs. referral — so you can see where growth is coming from.
Retrieve advertising performance
From Google Ads, the agent pulls yesterday's campaign-level data: total spend, impressions, clicks, CTR, conversions, and cost-per-conversion. It identifies the top 3 performing campaigns and the bottom 3 by ROAS. Any campaigns that exceeded their daily budget or saw a significant CPC increase get flagged.
Aggregate revenue and payment metrics
The agent pulls Stripe data for the same period: gross revenue, net revenue (after fees and refunds), new subscriptions, churned subscriptions, average transaction value, and refund rate. It calculates MRR (monthly recurring revenue) changes and identifies any large transactions or unusual refund patterns that deserve attention.
Assess pipeline and deal health
From Salesforce, the agent retrieves pipeline metrics: total pipeline value, deals by stage, new opportunities created, deals closed (won and lost), and average days in each stage. It highlights deals that are stuck (no stage change in 14+ days), deals closing this week, and any opportunities with amounts above your threshold that need executive attention.
Compare against historical trends
This is where the recipe gets powerful. The agent reads previous daily reports stored in Memory, comparing today's numbers against the 7-day average, 30-day average, and same day last week. It calculates trend direction and magnitude for every metric — is traffic up 15% week-over-week? Is conversion rate declining for the third day in a row? Is pipeline velocity improving? Trends that deviate more than 20% from the rolling average get flagged as anomalies.
Generate cross-platform insights
The agent connects dots across platforms that no single dashboard can show. It correlates Google Ads spend increases with Stripe revenue changes to calculate true advertising ROI. It maps GA conversion spikes to Salesforce lead creation to measure marketing-to-sales handoff speed. It identifies mismatches — like ad spend going up while pipeline stays flat, or traffic growing but conversion rates falling — and surfaces them as action items.
Build the briefing and store trends
The agent produces a structured briefing with four sections: Key Metrics (the numbers), Trends (what's changing), Anomalies (what needs attention), and Recommendations (what to do about it). The briefing is conversational, not a data dump — it reads like a report from a sharp analyst, not a spreadsheet export. Finally, it saves today's metrics to Memory so tomorrow's report can build on the trend data.
What Happens
Every morning, you get a comprehensive business briefing without opening a single dashboard. The reports get smarter over time because the agent remembers historical data and can spot emerging trends before they become obvious. Instead of reacting to last month's numbers, you're catching changes as they happen.
Run this recipe at the same time every morning. Consistency in timing makes the trend comparisons more accurate since you're always comparing the same business hours.
The Memory integration stores your daily snapshots in a structured format. After 30 days of daily runs, the agent can generate monthly trend reports comparing week-over-week and month-over-month performance with genuine historical context, not just two data points.