Analytics FAQs

See some of the frequently asked question about Analytics in Turtl

Maruša Barle avatar
Written by Maruša Barle
Updated this week

This article will cover the following frequently asked analytics questions:

What is Turtl's definition of a read?

The total number of times a Turtl Doc was opened and the reader explored beyond the first page. If they don’t turn the first page within 30 minutes, they are counted as a bounce. If they come back after 30 minutes and take action, this is counted separately as a read.

What constitutes a reader when there is a reader personalization form?

Anyone who completes the personalization form and lands on the doc itself, then explores beyond the first page.

How does Turtl calculate bounce rate?

If you want to be able to calculate the bounce rate yourself, here is how you can arrive at the bounce rate number as it appears on the dashboard from the reads analytics report.

As explained here, the formula we use is this one:

Math.round(((allReaders - readers) * 100) / allReaders) || 0;

allReaders can be obtained by applying the following filters to reads analytics report:

  • Filter out (remove) any rows that have value 0 in is_initialized column.

  • Filter out (remove) any rows that have value 1 in is_banned_ip column.

  • Only keep rows that have either both values in source_iframe and source_lightbox 0 (meaning Doc is not embedded) or both values 1 (meaning Doc is embedded, but opens in a lightbox rather than in an iframe.). The reason is that we don’t include reads on Docs that are embedded in an iframe when calculating the bounce rates.

  • After applying the above filters, identify the number of unique values in reader_id column to get allReaders.

To get readers, the steps are the same, but you would need to additionally remove any rows where count_read is 0 before counting the unique values.

Note: The readers no. on the dashboard may be higher because it includes readers that have read the Doc embedded in the iframe. These reads are excluded from bounce rate calculation.


Can Turtl block or filter certain IP addresses from the analytics?

Turtl can block analytics data for certain IPs or CIDR IP groups. This means that the data from blocked IPs is not displayed on the dashboard for the time during which IPs are blocked. By blocking IPs, we also filter the analytics data that was generated by the same IPs before they were blocked. The maximum number of IPs that can be blocked is 525,000.

Note: A high number of banned IPs may affect dashboard analytics performance.

Please turn to support@turtl.co with a list of IPs you wish to block and they will arrange it for you.


If I unblock certain IP addresses, will I be able to retrieve the analytics data?

Unblocking IPs will retrieve the data that was generated before blocking.


I have engaged with the Doc. Why is my IP not showing in Reads raw data download?

If you set the timeframe to the last 30 days and download the reads report CSV file you will get reads for today minus 30 days (so, excluding today).

If you also need today's reads (up to 5 minutes ago) you can select a custom timeframe and set today as an end date and you will have today's reads included in the downloaded CSV file.


Does Turtl collect analytics from logged-in in Turtl users?

We do not collect analytics events for logged-in Turtl users.

If you log into Turtl via SSO and we match your email address with a corresponding one in the tenant, you will be logged in just like any other user and analytics will not be collected.

If we aren't able to match your email address with any of the users in the tenant, you will log in via SSO as a Guest. We do collect analytics events for Guests.

If you submit your email address via a form in a Turtl Doc, we will register and count you as a sign-up, regardless of whether we are collecting your analytics events or not.


How does Turtl process known readers and unknown readers (guests)?

Both known readers and guests create unique reader ids (reader_id) when they access a Turtl Doc. If they access a Doc from the same browser/device, they will be matched against the existing id, but if they access it from a different browser/device, a new reader_id will be created, so the same reader will actually be counted twice.

If some readers are counted twice, the total number of (unique) readers as displayed on the dashboard will be slightly higher than it actually is. This is usually not a problem, as readers don’t usually view Docs in different browsers/devices.

Known readers are additionally identified by their email address (reader_email), so when they log in with their email (which is the case in the above access settings), a reader_email entry is created. If they happen to log in from a different browser/device and they enter the same email, it will be matched against the existing reader_email, so the total number of unique known readers will be accurate.


Where does the no. of signups on the dashboard come from?

The sign-ups can either come from a native Turtl form in the back cover or in the immerse section or a third-party form anywhere on the doc. If a sign-up is created with the same email address multiple times, each sign-up will add up to this number. For a more in-depth analysis of sign-ups, please download Reads raw data.


What is the relationship between sign-ups and known readers? Is the no. of known readers equal to the number of sign-ups?

Known readers can be identified through:

  • Sign-up form from Turtl or any third-party form integrated with Turtl

  • Lead capture URL

  • Email authentication (SSO or ‘Specific people only’ access settings)

This means that the no. of known readers may be higher than the no. of sign-ups.

It is also worth noting that once we have identified a known reader through a sign-up form, lead-capture URL, etc., we will be able to collect known reader analytics across other Turtl Docs they read, too, even if they weren't identified again through one of the above methods.

However, if the reader clears their browser cookies, we won’t be able to track known reader analytics until they complete a new sign-up form, or open a new lead capture URL.


Our doc is internal, so why is the number of known readers smaller than the number of all
readers?

User may create several unique reader_ids, which count towards readers if they log in from different browsers or devices, but we only display unique email addresses as known readers. This can be checked if you count unique reader_id entries against unique reader_email entries in Reads raw data report.


I can see X number of sign-ups, but no known readers. Why is that?

Being logged into Turtl will exclude readers from being shown in the Known reader's section.

If a signup comes from a list of banned IPs, they will not add to the number signups.


What is a lead store?

This is where your known reader information is stored. If your CRM is integrated with Turtl, your CRM will show here. Turtl will appear if you don't have one set up. Internal means the reader had access to the Doc either through SSO or email authentication (Source will show as 'Authentication' in this case).


What happens if a reader was identified via multiple channels, what will appear under ‘Store’?

If a person completes the form when they’re already identified (a known reader), the store that identified them first will be written in the “Known readers” section.

Example:

First, a company sends Turtl docs with lead capture URLs from their CRM, which already identify a reader.

Then, the reader is prompted to fill out a Turtl form to sign up for a webinar or newsletter, etc.

In this case, the known reader's lead store is a CRM (Hubspot, Eloqua, etc. ) and not Turtl.

If a personalization or a Doc is deleted, does this remove their analytics from analytics dashboards?

Yes, it does! We don’t show analytics of removed personalizations and Doc analytics dashboards. If a Doc is deleted, we don’t show its analytics on team dashboard.


Why is the Avg. Immerse time showing as n/a in the analytics?

You may encounter a situation where you can see a reader clicking on an URL in an Immerse page but the analytics are showing the page performance time as n/a for an individual reader.

The Avg. Immerse reader time is showing as n/a if a reader lands on an Immerse page and does not flip that Immerse page or go to the Surf page.

Did this answer your question?