What’s New in Google Analytics: More Room for Curiosity
Explore how Google Analytics helps teams investigate changes, monitor performance and test budget scenarios before committing to deeper analysis.
Chris Wright
Associate Director, Training and Development
We Can Afford to Be Curious Again
How new capabilities in Google Analytics are making it easier to ask questions before committing to a full analysis
One of the original promises of analytics was pretty simple. Give people good data and useful tools, and they can follow their curiosity.
You notice something. You look into it. You learn something that gives you another question, so you follow that too. Some of those questions lead somewhere useful. Plenty don't. That's fine. The important thing is that asking the question itself doesn't require much commitment.
In practice, analytics became a lot more operationally complicated than that.
Even a fairly casual question can turn into a request. The request gets prioritized. Someone gathers what they need, prepares the data, builds the analysis, validates the results and eventually comes back with an answer.
There's nothing inherently wrong with that process. A lot of analytical questions need that level of care.
The problem is what happens when every question has to enter that process.
Curiosity gets expensive.
Leaders start thinking twice about sending a passing question to an already busy analytics team. Analysts have to protect their capacity. An idea that seemed interesting on Monday morning may be much less interesting by the time someone has room for it two weeks later.
And plenty of questions simply never get asked.
When asking questions is expensive, we learn to ration curiosity.
That's one of the reasons I've found some of the recent changes in Google Analytics so interesting. Alongside existing capabilities, they are starting to make that first step into the data much cheaper.
A faster lane for the first question
Ask Advisor, previously called Analytics Advisor, is probably the clearest example.
Say sessions from Organic Search dropped last week and I want to understand what changed.
I can start with something pretty close to that actual question. From there, what I find can lead me into the next one.
Was the decline concentrated on a few landing pages? Was it mostly mobile? Are we seeing the same pattern somewhere else?
For an analyst, that can remove some of the mechanical work required just to get an investigation moving.
I think there's an equally useful implication for people leading those teams.
If I send someone a message saying, “Can you take a quick look at this? I'm curious whether there's anything there,” that can actually be a small request.
Maybe they spend a few minutes with it and find nothing particularly interesting. Great. We satisfied the curiosity and moved on.
Maybe they find something unexpected and come back with a better question. Now we have something worth discussing.
And maybe that first look gives us enough evidence to decide that we should put real analytical effort behind it.
For an analyst, that can remove some of the mechanical work required just to get an investigation moving. Its responses still need checking, and some explanations will require evidence from outside GA.
There's an equally useful implication for people leading those teams.
If I send someone a message saying, “Can you take a quick look at this? I'm curious whether there's anything there,” that can actually be a small request.
Maybe they spend a few minutes with it and find nothing particularly interesting. We satisfied the curiosity and moved on.
Maybe they find something unexpected and come back with a better question. Now we have something worth discussing.
And maybe that first look gives us enough evidence to decide that we should put real analytical effort behind it.
What if we could find out in two hours whether a question is worth spending two weeks on?
We still have questions that deserve two weeks of work. We just have a better opportunity to figure out which ones they are before starting the clock.
GA can do more of the watching
There is another kind of analytical work that doesn't start with somebody having a question.
A lot of time gets spent watching for the moment when there should be one.
Did revenue move? Did conversions suddenly fall? Is something behaving differently enough that somebody needs to investigate?
GA can take on some of that monitoring.
Custom Insights are an established example. If a weekly change in revenue above a certain threshold matters to me, I can tell GA that. I can choose how frequently it evaluates the condition and who should receive an email when it happens.
That matters because somebody doesn't need to keep opening the same report just to establish that nothing happened.
There is also a version of this where I haven't already decided what I want GA to watch.
Generated Insights can surface changes that appear noteworthy and provide some context around what happened. That gives me somewhere to begin investigating.
This is where the relationship between these features gets more interesting.
I might arrive in GA because I already have a question. I might have a metric I know is important and ask GA to watch it for me. Or GA might surface something I hadn't been thinking about at all.
In each case, we're getting closer to the part where human expertise is particularly valuable: figuring out what happened, whether it matters, what might explain it and what we should do next.
Spend less analytical attention waiting for something interesting to happen. Spend it investigating when it does.
We can start exploring ideas about what happens next
Most analytics questions are naturally retrospective. Something happened, and we want to understand it.
There are also plenty of questions that begin with an idea.
Imagine someone says:
“If we had another $50,000 to spend, where would you put it?”
You'll probably get several answers pretty quickly.
The question I find more useful is:
“What would you expect to happen?”
Now we have something we can explore.
If we had another $50,000, where would you put it?
And what would you expect to happen?
Scenario Planning gives eligible GA properties a way to work with questions like this, using historical data and campaign data from linked integrations or imports. You can explore different budget levels, review suggested channel allocations and see how the modeled outcome responds.
A modeled scenario isn't a decision on its own. The output is an estimate, not a guarantee of performance.
What I like about it is that it gives us somewhere to put the idea.
Instead of debating a hypothetical entirely in the abstract, we can explore it, change it and see whether there's enough there to justify taking it further.
That's the same curiosity loop again, just applied to a bigger question.
Quick questions need to be allowed to stay quick
The technology only gets us part of the way here.
There's an operating habit that needs to change with it.
If every question from a senior person gets interpreted as a request for a polished analysis and a deck, then making the first few steps faster isn't going to help very much.
We need to be comfortable with exploratory answers.
“I took a look and I don't see much there.”
“There might be something here, but we'd need to investigate it properly.”
“I didn't find what you expected, but I did find something else that might be worth looking at.”
All three of those can be perfectly useful answers to a quick question.
Let exploratory answers stay exploratory.
A quick look can miss something, though. The stakes matter too. A question about a major revenue loss may deserve careful investigation even if the first exploration doesn't reveal much.
Once something looks promising, or the risk warrants it, we can increase the level of effort.
That's when I want the careful analysis. Gather what we need. Prepare it properly. Validate what we're seeing. Bring in the people with the right expertise and give them the time to do the work properly.
The quick exploration helped us decide where to put that effort.
We can afford to be curious again
The important story is that the distance between having a question and getting the first useful piece of evidence is getting shorter.
That creates room for more questions.
Careful analytical work is still going to take time. We still need reliable measurement, clean data, people who know how to analyze it and enough judgment to understand what the output actually means.
But that level of investment doesn't have to be the entry price for every question.
We can take a look first. We can let GA watch some of the things we care about. We can explore an idea and see whether it holds enough promise to keep going.
Have a question? Explore it.
Something changed? Investigate it.
Have an idea? Try it.
Then put the bigger analytical investment behind the questions that earn it.
Then we can put the bigger analytical investment behind the questions that earn it.
For a long time, organizations haven't suffered from a shortage of questions. The constraint has been how much work it takes to do something useful with them.
If that first step keeps getting cheaper, we can afford to ask more of them.
And I think that's one of the more exciting things happening in analytics right now.
More Insights
Building Scalable Serverless Data Pipelines on Google Cloud
Cem Bakar
Cloud Architect
Oct 7, 2026
Read More
Understanding the IAB TCF v2.3 Update: New Vendor Disclosure Requirements Explained
Rob English
Director, Data Solutions
Jun 17, 2026
Read More
CM360 Path to Conversion Reporting: Full Customer Journey Attribution
Monika Boldak
Associate Director, Marketing
Jun 3, 2026
Read More
More Insights
Sign Up For Our Newsletter
Napkyn Inc.
204-78 George Street, Ottawa, Ontario, K1N 5W1, Canada
Napkyn US
6 East 32nd Street, 9th Floor, New York, NY 10016, USA
212-247-0800 | info@napkyn.com
What’s New in Google Analytics: More Room for Curiosity
Explore how Google Analytics helps teams investigate changes, monitor performance and test budget scenarios before committing to deeper analysis.
Chris Wright
Associate Director, Training and Development
October 28, 2026
We Can Afford to Be Curious Again
How new capabilities in Google Analytics are making it easier to ask questions before committing to a full analysis
One of the original promises of analytics was pretty simple. Give people good data and useful tools, and they can follow their curiosity.
You notice something. You look into it. You learn something that gives you another question, so you follow that too. Some of those questions lead somewhere useful. Plenty don't. That's fine. The important thing is that asking the question itself doesn't require much commitment.
In practice, analytics became a lot more operationally complicated than that.
Even a fairly casual question can turn into a request. The request gets prioritized. Someone gathers what they need, prepares the data, builds the analysis, validates the results and eventually comes back with an answer.
There's nothing inherently wrong with that process. A lot of analytical questions need that level of care.
The problem is what happens when every question has to enter that process.
Curiosity gets expensive.
Leaders start thinking twice about sending a passing question to an already busy analytics team. Analysts have to protect their capacity. An idea that seemed interesting on Monday morning may be much less interesting by the time someone has room for it two weeks later.
And plenty of questions simply never get asked.
When asking questions is expensive, we learn to ration curiosity.
That's one of the reasons I've found some of the recent changes in Google Analytics so interesting. Alongside existing capabilities, they are starting to make that first step into the data much cheaper.
A faster lane for the first question
Ask Advisor, previously called Analytics Advisor, is probably the clearest example.
Say sessions from Organic Search dropped last week and I want to understand what changed.
I can start with something pretty close to that actual question. From there, what I find can lead me into the next one.
Was the decline concentrated on a few landing pages? Was it mostly mobile? Are we seeing the same pattern somewhere else?
For an analyst, that can remove some of the mechanical work required just to get an investigation moving.
I think there's an equally useful implication for people leading those teams.
If I send someone a message saying, “Can you take a quick look at this? I'm curious whether there's anything there,” that can actually be a small request.
Maybe they spend a few minutes with it and find nothing particularly interesting. Great. We satisfied the curiosity and moved on.
Maybe they find something unexpected and come back with a better question. Now we have something worth discussing.
And maybe that first look gives us enough evidence to decide that we should put real analytical effort behind it.
For an analyst, that can remove some of the mechanical work required just to get an investigation moving. Its responses still need checking, and some explanations will require evidence from outside GA.
There's an equally useful implication for people leading those teams.
If I send someone a message saying, “Can you take a quick look at this? I'm curious whether there's anything there,” that can actually be a small request.
Maybe they spend a few minutes with it and find nothing particularly interesting. We satisfied the curiosity and moved on.
Maybe they find something unexpected and come back with a better question. Now we have something worth discussing.
And maybe that first look gives us enough evidence to decide that we should put real analytical effort behind it.
What if we could find out in two hours whether a question is worth spending two weeks on?
We still have questions that deserve two weeks of work. We just have a better opportunity to figure out which ones they are before starting the clock.
GA can do more of the watching
There is another kind of analytical work that doesn't start with somebody having a question.
A lot of time gets spent watching for the moment when there should be one.
Did revenue move? Did conversions suddenly fall? Is something behaving differently enough that somebody needs to investigate?
GA can take on some of that monitoring.
Custom Insights are an established example. If a weekly change in revenue above a certain threshold matters to me, I can tell GA that. I can choose how frequently it evaluates the condition and who should receive an email when it happens.
That matters because somebody doesn't need to keep opening the same report just to establish that nothing happened.
There is also a version of this where I haven't already decided what I want GA to watch.
Generated Insights can surface changes that appear noteworthy and provide some context around what happened. That gives me somewhere to begin investigating.
This is where the relationship between these features gets more interesting.
I might arrive in GA because I already have a question. I might have a metric I know is important and ask GA to watch it for me. Or GA might surface something I hadn't been thinking about at all.
In each case, we're getting closer to the part where human expertise is particularly valuable: figuring out what happened, whether it matters, what might explain it and what we should do next.
Spend less analytical attention waiting for something interesting to happen. Spend it investigating when it does.
We can start exploring ideas about what happens next
Most analytics questions are naturally retrospective. Something happened, and we want to understand it.
There are also plenty of questions that begin with an idea.
Imagine someone says:
“If we had another $50,000 to spend, where would you put it?”
You'll probably get several answers pretty quickly.
The question I find more useful is:
“What would you expect to happen?”
Now we have something we can explore.
If we had another $50,000, where would you put it?
And what would you expect to happen?
Scenario Planning gives eligible GA properties a way to work with questions like this, using historical data and campaign data from linked integrations or imports. You can explore different budget levels, review suggested channel allocations and see how the modeled outcome responds.
A modeled scenario isn't a decision on its own. The output is an estimate, not a guarantee of performance.
What I like about it is that it gives us somewhere to put the idea.
Instead of debating a hypothetical entirely in the abstract, we can explore it, change it and see whether there's enough there to justify taking it further.
That's the same curiosity loop again, just applied to a bigger question.
Quick questions need to be allowed to stay quick
The technology only gets us part of the way here.
There's an operating habit that needs to change with it.
If every question from a senior person gets interpreted as a request for a polished analysis and a deck, then making the first few steps faster isn't going to help very much.
We need to be comfortable with exploratory answers.
“I took a look and I don't see much there.”
“There might be something here, but we'd need to investigate it properly.”
“I didn't find what you expected, but I did find something else that might be worth looking at.”
All three of those can be perfectly useful answers to a quick question.
Let exploratory answers stay exploratory.
A quick look can miss something, though. The stakes matter too. A question about a major revenue loss may deserve careful investigation even if the first exploration doesn't reveal much.
Once something looks promising, or the risk warrants it, we can increase the level of effort.
That's when I want the careful analysis. Gather what we need. Prepare it properly. Validate what we're seeing. Bring in the people with the right expertise and give them the time to do the work properly.
The quick exploration helped us decide where to put that effort.
We can afford to be curious again
The important story is that the distance between having a question and getting the first useful piece of evidence is getting shorter.
That creates room for more questions.
Careful analytical work is still going to take time. We still need reliable measurement, clean data, people who know how to analyze it and enough judgment to understand what the output actually means.
But that level of investment doesn't have to be the entry price for every question.
We can take a look first. We can let GA watch some of the things we care about. We can explore an idea and see whether it holds enough promise to keep going.
Have a question? Explore it.
Something changed? Investigate it.
Have an idea? Try it.
Then put the bigger analytical investment behind the questions that earn it.
Then we can put the bigger analytical investment behind the questions that earn it.
For a long time, organizations haven't suffered from a shortage of questions. The constraint has been how much work it takes to do something useful with them.
If that first step keeps getting cheaper, we can afford to ask more of them.
And I think that's one of the more exciting things happening in analytics right now.
More Insights
What’s New in Google Analytics: More Room for Curiosity
Chris Wright
Associate Director, Training and Development
Oct 28, 2026
Read More
Building Scalable Serverless Data Pipelines on Google Cloud
Cem Bakar
Cloud Architect
Oct 7, 2026
Read More
The Most Durable AI Investment: A Strong Data Foundation
Jasmine Libert
Senior VP Growth and GTM
Sep 28, 2026
Read More
More Insights
Sign Up For Our Newsletter