Why numbers differ between pages

#At a glance

Each place counts slightly differently, on purpose. Here’s how they compare:

#UTC days and ranges

• Teacup Quiz counts in UTC days. Shopify’s reports use your store’s time zone, so a day’s numbers differ by the hours in between. A store in Los Angeles sees its evening orders on the next day in Teacup Quiz.
• The 7, 30 and 90 day ranges include today, which is only partly over. Looking at 7 days today and again tomorrow moves the window by a day. Custom dates on a quiz’s page don’t move.
• Starts, finishes and orders each count on the day they happen. At the edges of a range, a run can start inside it and finish or order outside it.
• The quiz editor’s line is always the last 30 days, whatever range you used in Analytics.
• The weekly report covers last Monday to Sunday, so it never lines up exactly with 7 days in the app.
• A/B results count from 00:00 UTC on the day after the test started.

#Calendar month and billing cycle

Completed quizzes on the Plans page are what your plan counts, and two different periods are in play:

• Pausing, usage emails and paid plans’ AI drafts follow the calendar month (UTC), starting over on the 1st. Free’s one AI draft never starts over.
• On a paid plan with usage charges, the Plans meter and Home’s usage banner follow your 30-day Shopify billing cycle, which rarely starts on the 1st.

So the count in a usage email (calendar month) can differ from the Plans page (billing cycle). Both count every completed quiz across all your quizzes, including test versions and quizzes you’ve since deleted. See Plans.

#Orders with products from two quizzes

A shopper who takes your skin quiz and your body quiz may buy from both in one order. Each quiz counts that order with its own products, so it shows in both quizzes’ Orders in the Analytics list, on each quiz’s page, and under By quiz in the weekly report. Home and the weekly report’s tiles count it once.

So the Orders in the Analytics list can add up to more than Ordered on Home. Sales aren’t doubled anywhere: each quiz only adds its own products.

#Revenue per answer and per result

• Every answer a shopper picked gets their run’s whole order. On a question where shoppers pick several answers, the answers’ revenue adds up to more than the quiz’s.
• What shoppers answer, Result pages, Sources, Where the quiz is shown, Where shoppers are and Devices count orders by the day the shopper started the quiz, not the day they ordered, so they don’t match Revenue at the top of the page.
• They count your main currency only.
• On a very busy quiz, they read the 20,000 most recent runs with orders in the range.
• Products shares each order’s quiz value between its products by price, and lists only the top 12.

#Currencies

Nothing is converted between currencies:

• Home’s Sales from quizzes lists every currency, such as “$1,240.00 · €80.00”. Average order uses the currency most orders came in.
• A quiz’s page uses its main currency (the one most orders came in, or your store’s currency) for Revenue, Average order value and the cards, with other currencies under Revenue, such as “Plus €80.00”. Conversion counts orders in every currency.
• The weekly report shows the biggest currency without cents, then “+ 1 more” for the others.
• A/B results show sales in the currency most of the test’s orders came in.

#Views and start rates

• Views started being counted later than starts. For days before that, Viewed and Views show “–”, and a note says when counting began.
• Start rates only use days that have views, so Started ÷ Viewed doesn’t always equal the rate shown.
• Views count page loads, not people: a shopper who sees the quiz on three pages counts three times. A/B results count visitors the same way.

#Activity that isn’t counted

Some activity is left out of every number: your own test runs, the editor’s preview, shoppers who decline analytics, runs too fast for a person, and unusual bursts of finishes. It can still show elsewhere:

• Emails from those runs are still saved, so Email rate and the A/B Left their email row compare all emails with counted finishes only. They can look high, and in rare cases pass 100%.
• Shopify still records their orders, so Shopify shows more sales than Teacup Quiz.
• Runs that weren’t counted as finished show Didn’t finish in Responses.

#Deleted quizzes and test versions

• Home’s totals include every quiz’s numbers, including quizzes you’ve deleted and A/B test versions. The Analytics list shows only the quizzes you have now, so its rows can add up to less than Home.
• Home’s quiz cards hide test versions, while Analytics lists them as quizzes of their own.
• When you use a test version, it’s deleted. Its numbers stay in Home’s totals but no longer have a page of their own. See Reading A/B results.

#Teacup Quiz and Shopify’s reports

Use Shopify for your store’s sales, and Teacup Quiz for what the quiz brings in and how to improve it. When a particular order is missing, see Why an order doesn’t show.