Every Shopify operator has the same morning ritual. Open Shopify, glance at yesterday's revenue. Open Klaviyo, check the campaign you sent at 2pm. Open Meta Ads Manager, see if anything broke overnight. Maybe open Gorgias to clear tickets. Open Slack to tell the team what you found.
Forty-five minutes later, you have a vague feeling about how the store is doing. You haven't actually written anything down. By the afternoon you can't remember what yesterday's revenue was. By the end of the week, the team is in slightly different versions of reality.
A daily ops report fixes this. The good news is the report itself is dead simple — what made it expensive was the human time gathering the data. An AI agent flattens that to one scheduled run. This piece walks through the shape of a useful report, the prompt that produces it, and how to land it somewhere the team will actually read.
Why the daily report is usually broken
Three failure modes:
- It's only one app. Shopify-only reports miss the ad-spend context. Klaviyo-only reports miss the revenue context. A pulse needs both.
- It's a dashboard. Dashboards require you to go to them. Reports come to you.
- It's noise-heavy. Most automated reports dump twenty metrics. A useful pulse has four or five.
The fix is a written summary, posted to where the team already is (Slack, Telegram, email), pulling from multiple apps, scheduled, and short.
The shape of a useful 30-second pulse
A good daily report fits in a Slack message and answers four questions:
- Did we make money yesterday? Revenue, orders, AOV, vs the trailing 7-day average.
- What worked? Top one or two campaigns or flows by attributed revenue.
- What's broken? Anything anomalous — spike in refunds, drop in conversion, an underperforming paid campaign, a flow with an open-rate cliff.
- What needs a human today? One line. The single decision the team should pay attention to.
That's it. No graphs, no twenty metrics, no “here's everything we could possibly measure.” If you want graphs you have a BI tool for that. The pulse is for orientation, not analysis.
The prompt that produces it
The whole thing — the agent dispatching to Shopify, Klaviyo, and Meta in parallel, synthesizing, formatting, posting — is one sentence:
What the agent actually does under the hood:
- Shopify specialist pulls yesterday's orders, revenue, refunds, AOV, conversion rate; compares to trailing 7-day average.
- Klaviyo specialist pulls campaign sends + flow performance for the prior 24 hours; identifies top one or two by attributed revenue.
- Meta specialist pulls campaign-level spend, ROAS, and any active campaigns that crossed a threshold (CPA spike, ROAS drop, budget cap hit).
- Lead orchestrator takes the three responses, picks what's noteworthy, drafts the message, posts it to Slack with a thread for the supporting numbers.
Total elapsed time: 20–30 seconds. You read it on your phone before you open the laptop. The supporting detail is in the Slack thread if you want it.
Sample output
What lands in Slack:
◆ Daily ops pulse — Apr 23
Revenue: $42,180 (108 orders · AOV $390)
↑ 14% vs 7-day avg
What worked
• Klaviyo "Spring drop · day 2" — $8.4k, 38% of total
• Meta "New visitor lookalike" — ROAS 4.2, doubled budget
What's noisy
• Refund rate 4.1% (vs 1.8% avg) — 3 returns on SKU SHIRT-OLV-M;
Gorgias tickets cite fit. May need to update sizing copy.
Human needed
• The fit-feedback above is repeating. Worth a 10-min call with
the design team this week.That's the entire report. You know in 30 seconds what happened and what to do.
Where to land it (and where not to)
- Slack. Default for teams. The thread for supporting data, the top-level message for the pulse.
- Telegram. If you're a solo operator or your team prefers it. Same shape.
- Email. Works but easier to ignore.
- Not a dashboard. Dashboards don't come to you. The whole point is that this lands without you opening anything.
Extending it: the weekly, the monthly
The same pattern composes upward. A weekly version of this pulse is a different prompt to the same orchestrator:
The monthly version adds cohort retention, LTV by acquisition source, and inventory burn. Each new report is a prompt, not a development project.
Try it on your stack
If your morning ritual is opening four browser tabs and forgetting half of what you find, the daily pulse is the highest-leverage first thing to set up with an AI ops manager. The math is straightforward: 45 minutes a day → 30 seconds a day → 3.5 hours a week → a half-time week per month back.
Read the broader category framing in What is an AI ops manager, or jump to 7 Klaviyo plays you can hand off to an AI agent if email is the loudest item in your backlog. Install free and the daily pulse can be running by tomorrow morning.