The ecommerce weekly routine: 30 minutes that pay off

Most stores do not fail for lack of tools. They fail because nobody looks regularly. There is Analytics, Search Console, an ads account and a Merchant Center, but no fixed appointment at which the numbers turn into a decision. This article gives you exactly that: an ecommerce weekly routine of 30 minutes with a fixed order, clear decision rules and a template you can copy. Plus the monthly and quarterly loops, so the big topics do not sit untouched.

Why a routine beats another tool

Three patterns you see again and again.

Numbers without a rhythm do not produce decisions. If you open Analytics every six weeks, you will always find something odd and never know whether it is new.

Looking daily means reacting to noise. At small order volumes, conversion rate swings from day to day without anything having changed. Measures derived from that kind of swing do damage, because they overwrite effects before you can see them.

Without a fixed slot, the urgent always wins. Returns, supplier emails, a broken image. The work that produces growth is rarely urgent, so it is the first thing to get dropped.

The fix is unspectacular: a recurring appointment, always the same order, always a note at the end.

The ecommerce weekly routine, minute by minute

Put the appointment on a fixed weekday, ideally Monday or Tuesday morning, and block 30 minutes. More is allowed, less is not enough. The timings below are guide values from practice, not a rule.

Minutes 0–5: the four base numbers

Compare last week against the week before and, if you have the data, against the same week last year:

• Revenue
• Number of orders
• Conversion rate
• Average order value

Write the values down even when nothing stands out. The value of the note appears after two months, when you have a series. What these numbers mean and how they interact is in the overview of the ecommerce metrics that actually matter.

Minutes 5–12: where visitors came from

Sessions per channel: organic search, paid search, direct, email, social, referral. And revenue for each, not just sessions.

Two questions are enough. Did a channel drop noticeably? Did a channel grow without you knowing why? Both are reasons to look closer; everything else is normal operation. Use the same channel grouping every week, otherwise you are comparing weeks with different definitions.

Minutes 12–20: advertising

If you run Google Ads or Meta Ads:

1. Spend for the week against the week before.
2. Revenue from advertising and ROAS per campaign.
3. Flag campaigns with spend and no conversions.
4. For search campaigns: go through the search terms list once and exclude the obviously irrelevant ones.
5. Merchant Center: are there disapproved products? A disapproval left sitting for three weeks costs revenue quietly.

Two things about budgets that save confusion here. For most campaign types Google Ads allows a daily spending limit of twice the average daily budget and a monthly spending limit of 30.4 times it — so a single expensive day is not a reason to intervene. And a campaign flagged as limited by budget is telling you about lost impressions, not about a mistake; what limited by budget actually means is worth reading once so you can skip that decision every week afterwards.

If a campaign spends and converts nothing across a full evaluation period, that is a different situation — the causes are sorted through in clicks but no conversions in Google Ads. How to calculate the return properly, and what it hides, is in how to calculate ROAS. Disapprovals get their own short process in product disapproved in Merchant Center.

Minutes 20–26: visibility

In Search Console, last 28 days against the previous 28 days:

• Clicks and impressions as a trend
• The five pages with the biggest changes
• New indexing errors
• Queries in positions 8 to 20 with a lot of impressions

The last point is the most productive, because that is where a small amount of work produces movement. Queries that collect impressions but almost no clicks are a separate pattern with its own fixes — see impressions but no clicks.

Minutes 26–30: one decision and one note

At the end you write down two things: what stood out this week, and which single measure you will implement this week. One, not five. Plus a date on which you will look at the result.

If you have to collect these numbers from several interfaces, the collecting alone takes longer than the analysis. A weekly email report, like the one ecomando sends, removes that step: the values already sit next to each other and the routine starts with reading.

What gets checked at which rhythm

The daily check is deliberately tiny. Its only job is to notice a fault within 24 hours, not to spot trends.

Decision rules: when to act, when to wait

This is where most routines fail. Without rules, every fluctuation becomes a measure.

Act when

• Something is broken. Error pages, tracking stopped, disapproved products, a checkout that does not complete. Immediately, no discussion.
• A channel sits clearly below the level of previous months for two weeks running and there is no known explanation.
• Money flows without result. A campaign that spends across a whole evaluation period and delivers no conversion gets paused or rebuilt.
• A concrete opportunity is visible. A query with many impressions at position 9 whose landing page has no real copy.

Wait when

• It has only been one week. A weekly figure is a data point, not a trend.
• The case count is small. At 20 orders a week, conversion rate moves without anything having changed. Then you compare months, not weeks.
• A change cannot have taken effect yet. SEO changes often need one to three months to show in rankings.
• You just tested something else. Two changes at once make the result unreadable.

Never

• Change several levers in the same week, if you want to know what worked.
• Adjust budgets daily. Campaign learning phases get reset.
• Derive a measure from a screenshot without a comparison period.

A constructed example on case counts: a store has 1,000 sessions and 20 orders per week, so a conversion rate of 2.0 percent. The following week it is 16 orders at the same sessions, so 1.6 percent. That looks like a 20 percent collapse and is a difference of four orders. At those magnitudes, patience beats action. How to calculate and interpret the rate cleanly is in ecommerce conversion rate.

The monthly loop

Once a month, two to three hours, ideally on the first working day. Fixed order:

1. Review the measures. Take your four weekly notes and answer per measure: implemented, effect visible, keep or roll back? Measures without a review are the most common reason stores go in circles.
2. Contribution margin instead of revenue. Revenue per product group minus cost of goods, shipping, payment fees and attributed ad spend. Priorities often shift considerably against the revenue list.
3. Range. What is selling, what has been sitting for months, where is something missing that customers asked for?
4. Returns. Group the reasons. Recurring reasons are almost always a copy problem on the product page, not a customer problem.
5. Checkout. Once a month, walk your own checkout on a phone, as a new customer, with each payment method. The recurring causes are collected in reducing cart abandonment.
6. Competitors. Prices, ads, visibility, new products.
7. Next month. Set two or three intentions, no more, each with a date.

The quarterly loop

Four times a year, for the questions a weekly cadence cannot answer:

• Retention. How many customers buy a second time? How long until that second purchase? What is a customer worth over time?
• Channel mix. Are you still splitting budget the way the last few months' results justify? This is the moment to move money between channels, not in week three.
• Technology. Load time, error pages, indexing, mobile rendering as a whole.
• Prices and free shipping thresholds. Moving a free shipping threshold affects average order value and deserves to be planned, not done on impulse.

Template for the weekly note

Copy this block into a spreadsheet or a notes document, one row per week.

Week (no. / date):

Revenue:                ____   (prev week: ____ / prev year: ____)
Orders:                 ____   (prev week: ____)
Conversion rate:        ____   (prev week: ____)
Avg. order value:       ____   (prev week: ____)

Sessions organic:       ____   Revenue: ____
Sessions paid:          ____   Revenue: ____
Sessions direct/email:  ____   Revenue: ____

Ad spend:               ____   ROAS: ____
Campaigns with no conversions: ____________________
Disapproved products in Merchant Center: ____

Search Console: clicks ____ / impressions ____
New errors: ____________________
Opportunity of the week (query + page): ____________________

Stood out this week:
____________________

Measure this week (exactly one):
____________________
Check the result on: ________

Four filled rows are a monthly overview. Twelve are a quarterly report. Keep this note for six months and you have something no tool supplies: the link between the numbers and your own decisions.

Common mistakes in the first weeks

• Too many metrics. Start with 20 values and you stop after three weeks. Five to seven is enough.
• No fixed appointment. Without a calendar entry the routine does not happen.
• No note. Without a written trail you will run the same analysis again next quarter.
• Several measures at once. Then you will not know at the end what worked.
• Analysis without checking the measurement. If tracking counts wrongly, every conclusion is wrong. Check it once, properly, before you base decisions on it: the walkthrough is in GA4 ecommerce tracking not working, and for Shopify stores specifically, why Shopify revenue does not match GA4 explains which gaps are expected.

Checklist for getting started

• Weekly appointment in the calendar, 30 minutes, recurring
• Template for the weekly note created
• Tracking checked once: orders and revenue agree with the store
• Search Console and Analytics set up, access working
• Ad accounts and Merchant Center to hand, or merged into one view
• Decision rules written down and visible
• Monthly appointment in the calendar, two to three hours
• Quarterly appointment in the calendar
• After four weeks: first monthly loop done, measures evaluated

The routine feels like extra work in week one and like an obvious habit after two months. Its real benefit is not that you know more. It is that decisions happen at a fixed time instead of getting postponed.