Building post-call data extraction variables in Retell AI
This article explains how to build reliable post-call data extraction for voice AI agents. By the end, you'll know how to define
Insights on Voice AI, GA4, and data analytics
This article explains how to build reliable post-call data extraction for voice AI agents. By the end, you'll know how to define
A multilingual voice agent can understand and reply in more than one language during a single call. The multilingual voice agent below can reply in
Use the following SQL (automatically generated via a text prompt in ChatGPT) to create a Google Analytics 4 Exit Page report in BigQuery: -- Calculating Exit
Use the following SQL (automatically generated via a text prompt in ChatGPT) to create a Google Analytics 4 landing page report in BigQuery: WITH base_
Follow the steps below to create a Google Analytics Scroll Tracking Report in BigQuery: Step-1: By default, GA4 tracks only 90% of vertical scroll
What are Environment Tags in Retell AI? An environment tag is a named pointer that you attach to a version of an agent. An Environment
Versioning lets you update a voice agent while keeping other versions unchanged. This means you can keep improving an agent without ever breaking a live
Many AI content developers will tell you that you don't really need to know n8n anymore. Just use Claude to build and deploy
The biggest mistake companies make in AI workflow development is trying to automate 100% of their processes, and that too without a human in the
Google has moved the Measurement Protocol into maintenance mode. No new features are planned for it. If you send server-side events to GA4 this
For Voice AI solutions, a well-established business with high call volume is an ideal client. But one big problem that I encountered is that
I want you to know that no solution (custom or third-party) can recover or reconstruct the full, row-by-row hit data for GA4
Over the past few years, I’ve tested both native GCP (Google Cloud Platform) setups and third‑party “easy” platforms (e.g. Stape, Taggrs) across
When you conduct a downward analysis, mixing dimensions and metrics of different scopes is not problematic in GA4 reports. It is problematic when you conduct
When you schedule data deletion in GA4 (via Admin > Data Collection and modification > Data deletion requests), Google give the following three warnings: However,
Should you wait for 24 hrs, 3 days, 7 days or 12 days before analysing and reporting on your GA4 data? The wait time largely
The Advertising Reports section of the GA4 property is getting a major revamp: You can see now there is a clear distinction between Key Events
The 'Days to key events' metric in the GA4 Attribution Paths report is not what you think. This metric is unreliable for understanding
When your GA4 property uses a reporting identity setting other than "Blended", it can underreport conversions, transactions, and revenue. This underreporting also affects
By default, GA4 counts users as engaged if they have generated one or more engaged sessions. An engaged session in GA4 is defined as a
The default Key Event counting method used by GA4 properties makes your conversion tracking setup inaccurate by default. The majority of users have set up
You can calculate 'Active Users' in GA4 BigQuery by counting the number of unique users who triggered at least one event during the
Google recently launched two new audience templates in GA4 that directly target your best customers and revenue leakage. These two new audience templates are: #1
You should not configure Google Tag’s settings simultaneously via GTM, Google Ads, and GA4. The Google documentation regarding this is not very clear, and
You should not be splitting nested event tables of GA4 BigQuery into separate tables (like sessions, transactions and users) just for the sake of simplicity
Follow the steps below to check for any change in the GA4 BigQuery schema: Step-1: Make sure you have connected your GA4 property with
If, for whatever reason, you can not reduce the token size of your voice agent prompt but still want to reduce its latency, experiment with
Check your voice agent prompt for zero width character immediately after creating the first draft of the prompt. What are zero-width characters in Voice
I’ve been tracking my time for over a decade using a time-tracking tool called 'Hravest'. So naturally, I also log the
As I crossed over 1000 hours working with AI, I discovered 4 LLM failure modes and 1 user failure mode. The five failure modes aren&