Analytics page explanations

The Analytics page answers three questions: how many people came to your store, where they came from, and what they did once they arrived. This guide walks through every number on that page — what it counts, what it does not count, and how to read it — one card at a time.

Where to find it

Open Analytics from the sidebar. The page is a single scroll, and it always shows the same six blocks in the same order:

1. The header, with the live now pill and the period selector
2. Four overview cards — Total Views, Orders, Revenue, Conversion
3. The Page Views chart
4. The Revenue chart
5. Traffic Sources
6. Top Pages and Top Viewed Products

Everything below the header follows whichever period you pick, so start there.

The period selector

Three buttons at the top right: 7 Days, 30 Days, 90 Days. The default is 7 days. Whichever you choose, every card, chart and table on the page reloads for that window — there is no separate date filter anywhere else on the page.

The period also decides what the little green or red badge on each overview card compares against. It is always the same length of time immediately before the window you are looking at:

Period selected	The badge compares	Against
7 Days	the last 7 days	the 7 days before those
30 Days	the last 30 days	the 30 days before those
90 Days	the last 90 days	the 90 days before those

So a green +40% on the Revenue card with 30 Days selected means: this month earned 40% more than the month before it. Green is up, red is down. If the earlier period was completely empty and this one has any activity at all, the badge simply reads +100% — there is no other sensible way to compare against zero.

Note: The current day is still in progress, so the newest period always includes a partial day. When you check at 10am, today’s bar looks small next to yesterday’s. That is expected, not a drop.

The “live now” pill

The green pill beside the period selector counts the shoppers who are on your storefront right now — more precisely, distinct visitors seen in the last 5 minutes. It refreshes by itself every 15 seconds, so you can leave the page open while you post a link and watch people arrive.

A few things worth knowing:

• Visitors are counted anonymously. The count comes from a one-way fingerprint of the connection, and no personal information is stored for it.
• It is a live count, not a history. It is not affected by the period selector and it is not saved anywhere — there is no way to look up “how many were live last Friday”.
• If the pill is missing entirely, the live counter is temporarily unavailable. A count of zero shows as 0 live now, so a hidden pill and an empty store are not the same thing.

This is the fastest feedback loop you have. Post to Facebook, watch the number rise, and you know the link works.

The four overview cards

Card	What it counts	Cancelled orders
Total Views	Every storefront page view in the period	—
Orders	Every order placed in the period	Included
Revenue	Order totals added up	Excluded
Conversion	Orders ÷ product page views, as a percentage	Orders included

Total Views

Every page load on your storefront during the period, added together across four kinds of page: your homepage, product pages, category pages and landing pages.

This counts views, not people. One shopper who opens your homepage, looks at three products and refreshes one of them adds 5 views, not 1. That is normal for this kind of counter, and it is why Total Views is best read as a trend line rather than an audience size — if it doubles week over week, twice as much browsing is happening.

Not counted here: your content pages (About, Privacy, Terms), the cart, checkout and search pages, and anything you do inside your own admin dashboard. Visiting your own storefront in a normal browser tab, though, does count — so check your store in the same browser all day and you will inflate your own numbers a little.

Orders

The number of orders placed during the period, straight up. This one includes every order, cancelled ones as well, because it is measuring how many times someone completed checkout — not how much of it survived.

Revenue

The order totals for the period added together, excluding cancelled orders. Each order’s full total is used, which includes the delivery charge. An order counts on the day it was placed, not the day it was delivered or paid for, so a COD order that is still in transit is already in this number.

Note: Because the Orders card includes cancellations and Revenue does not, these two cards can move in opposite directions. Orders up but revenue flat usually means you are cancelling more — worth checking the fake-order and delivery-history tools on the order page.

Conversion

Orders divided by product page views, shown as a percentage. If 500 people opened product pages and 15 orders came in, conversion reads 3%.

Two details make this number easier to trust:

• The bottom of the fraction is product views only — not Total Views. Homepage and category browsing does not dilute it.
• The top is every order in the period, including any that came through a landing page without a product-page visit first. On a store that sells mostly through landing pages, this can push conversion above what a strict funnel would show.

Read it as a health indicator rather than an exact funnel measurement. What matters is the direction over time and how it compares between your own products, and for that it is very reliable. Somewhere between 1% and 3% is ordinary for a Bangladeshi storefront; above 5% is strong.

The Page Views chart

A line chart with one point per day and five lines: Total, Homepage, Product, Category and Landing. The big number at the top right is the total for the whole period.

The five lines are what make this chart useful, because the shape of the gap between them tells you where attention is going:

• Homepage high, Product low — people arrive and leave without opening anything. Your homepage is not putting products in front of them fast enough.
• Product high, Orders low — interest is there but something at the last step is not working. Price, delivery charge, photos, or the checkout itself.
• Landing climbing — your ad campaigns are landing where you sent them.

Days with no traffic are drawn as zero rather than skipped, so a flat line at the bottom really does mean nobody came that day.

The Revenue chart

A bar chart with two bars per day: Revenue against the left axis (in ৳) and Orders against the right axis (a plain count). The totals for the period sit above the chart.

Cancelled orders are excluded from both bars here. That is the one place this chart differs from the Orders card above it, and it is the most common reason the two order counts on this page do not match. If the card says 48 and the chart says 44, four orders were cancelled.

Two axes on one chart means the bar heights are not comparable to each other — only each bar against its own history. What you are looking for is the two series moving apart: orders steady while revenue climbs means your average order got bigger, which is usually a bundle or an upsell working.

Traffic Sources

A horizontal bar chart showing where your visitors came from, built entirely from the UTM tags in the links you share. Up to eight tracked sources are shown, plus one bar for Direct / None.

How a visit gets a source

When someone opens a link that carries utm_source, utm_medium or utm_campaign, your store remembers those tags in the shopper’s browser and attaches them to every page they view in that visit. Two consequences are worth remembering:

• The first campaign wins. Once tags are stored for that browser, a later link does not overwrite them for the rest of the visit.
• The whole visit is attributed, not just the page they landed on. Someone who arrives from a Facebook ad and then browses eight pages contributes eight views to that campaign.

Direct / None

Everything with no UTM tags at all. That includes people who typed your address, used a bookmark, came back later — and, importantly, anyone who clicked a plain untagged link you posted anywhere. A Facebook post with a bare link lands in Direct / None, not under Facebook.

A large Direct bar is not a problem in itself, but it is usually a sign that your links are untagged rather than that your customers are loyal.

Tagging your links

Add the tags to the end of any link you share:

https://yourstore.sohojdokan.com/?utm_source=facebook&utm_medium=cpc&utm_campaign=eid-sale

• utm_source — where the link lives: facebook, instagram, whatsapp, tiktok, sms
• utm_medium — the kind of link: cpc for paid, post for organic, bio for a profile link
• utm_campaign — which push it belongs to: eid-sale, winter-launch

Only utm_source is really necessary. Use lowercase and keep spelling consistent — Facebook and facebook are stored as two separate bars.

The chart works for product and landing-page links too, not just your homepage: put the tags after the product URL and that campaign’s traffic shows up in the same place.

Top Pages

Your ten most-viewed pages for the period, each with a coloured badge showing what kind of page it is — Homepage, Product Pages, Category Pages or Landing Pages — and a small bar showing its size against the busiest page.

Products and categories appear by their real names, not their web addresses, so you can read the list without decoding slugs.

Note: The number at the top right of this table is the total of the rows shown, not your whole store. It will be smaller than Total Views, and on a store with many products it will be much smaller. That is not a discrepancy.

Top Viewed Products

Your ten most-viewed products, with four columns:

Column	What it means
Product	Name and thumbnail — click to open the product for editing
Views	Product page views during the period
Orders	How many orders in the period included this product
Conv.	Orders ÷ Views for that product

The Conv. figure is colour-coded so problems stand out without doing arithmetic: strong at 5% and above, fair between 2% and 5%, and greyed out below that.

This is the most directly useful table on the page. A product with high views and a low conversion has already done the hard part — people want to look at it. Something is stopping them at the end, and it is almost always one of four things: the price, the delivery charge, weak photos, or a description that leaves a question unanswered. Fix one and check the same row a week later.

The opposite pattern is worth acting on too. A product with few views but a high conversion is being hidden, not rejected — put it on the homepage or in your next campaign and it will likely repay the placement.

How the numbers are counted

Knowing the mechanics saves you from chasing differences that are not real:

• A small tracking beacon fires once per page load on your storefront. A refresh counts again. Nothing is de-duplicated per person, so “views” always exceed “visitors”.
• Counts are stored per calendar day, Bangladesh time. Today’s totals keep climbing until midnight.
• Views are recorded for four page types only: homepage, product, category and landing pages.
• Nothing that identifies a shopper is stored for these counts.
• If a visitor blocks scripts or leaves before the page finishes loading, their view is not recorded.
• The Analytics page reads only your own store’s data, and it always shows the store you are currently signed in to.

When two numbers don’t match

What you notice	Why
Orders card ≠ orders in the Revenue chart	The card includes cancelled orders; the chart does not
Conversion looks higher than your own maths	It is divided by product views only, not total views
Top Pages total is far below Total Views	The table totals only the ten rows shown
These numbers differ from Google Analytics or Meta	Different counting rules, and blockers stop each tool differently — a 10–30% gap is normal
The live now pill is missing	The live counter is temporarily unavailable; zero would still display as 0 live now
Traffic Sources is nearly all Direct	Your shared links have no UTM tags on them
A view you made yourself appears	Your own storefront visits count; only admin dashboard activity is excluded

In the mobile app

The merchant app has Analytics in the bottom navigation bar. It shows the same period selector and the same four cards — Views, Orders, Revenue, Conversion — plus the page-views chart, revenue, top products and top pages, calculated exactly the same way.

Traffic Sources and the live now pill are currently available only in the web dashboard.

Tips

• Tag every link you post with at least utm_source. It costs a few seconds and it is the only way Traffic Sources can tell you anything.
• Check 7 Days for what to do this week and 30 Days to judge whether it worked. Day-to-day swings on a small store are mostly noise.
• Work down the Conv. column of Top Viewed Products before you spend more on ads. Fixing a product that already gets traffic is cheaper than buying more traffic.
• Leave the page open with the live now pill visible for ten minutes after posting a campaign — you will know immediately whether the link is reaching anyone.
• Before you compare two months, remember the newest period includes today, which is not over yet.

What’s next

For a quick walkthrough of the page, see the analytics dashboard. For deeper store metrics — order pipeline, repeat-customer rate, low-stock alerts — switch your home dashboard to the Advanced view. To measure ads outside your store, connect Google Analytics or the Meta Pixel.