Let’s start this off by simply asking: why do we want to be alerted? Alerts prevent us from having to check in with our data constantly, most people only want to know when something occurs without having to check constantly. Then again, there are always the ones who love their data, charts and analytics, but even the most data-centric of us need alerts because you can’t be everywhere all the time.

Why use Google Analytics custom alerts

Let’s look at a scenario where alerts are crucial.

An e-commerce manager at ABC company logs into his analytics account to look over the previous day’s performance, only to see that revenue has visibly plummeted. He sees this after it has occurred because he didn’t have his alerts set up to notify him of any issues regarding changes in revenue. Revenue is merely one parent metric that he has to look at- in order to accurately prevent revenue dips, the e-commerce manager needs to look at hundreds of metrics to make sure that all is ok.

Here is where Google Analytics alerts can help. By setting up custom alerts; analytics users can rest assured that if something goes wrong, their static alerts will be able to notify them.

The case for alerts is clear, Google search results exceed  10,000,000 for the search query “custom Google Analytics alerts” – people have shown the demand and need. If you feel overwhelmed and think that it’s hard to be constantly checking and monitoring your metrics, you need to set up Google Analytics Alerts.

We’re going to go over how to set up analytics alerts and the key metrics that you need to set up so that you don’t skip a beat going forward.

How to Create Custom Google Analytics Alerts (Step by Step)

Step 1: Make sure you’re logged into Google Analytics

Step 2: Go to the admin option

Step 3. After clicking the “Admin” option, you’ll see settings columns. Go to View column and go to the Custom Alerts option below Personal Tools and Assets. Click Custom Alerts.

Step 4. After clicking “Custom Alerts”, you’ll see your “custom alerts” overview along with the New Alert button. There will be no items in the list if you haven’t created any custom alerts. Click “New Alert” for a new custom alert.

Step 5. A new page where you can add a new custom alert will be shown.


Step 1: Alert name – Enter an alert name i.e Revenue dropped

Step 2: Apply to – If you have more than 1 Analytics property, choose the account from the apply to option.

Step 3: Period –You can select how often you want to be alerted in the Period option. You can choose by Day, Week, or Month. If you choose Day, it creates alerts based on daily changes in traffic or interactions.

Click Send me an email when this alert triggers. To send alerts to other members of your team add their email addresses from the “also include” option.

Step 4: Alert Conditions – In the Alert Conditions, you can set up the following:

  • This applies to: You can choose for it to apply to All Traffic or specific dimensions.
  • Alert me when: Apply the alert to a specific metric.
  • Condition: Condition options include Is less than, Is greater than, Decreases by more than, % Increases by more than, etc. If you aren’t sure about the average value of any metric, you can choose % Increases/ Decreases by more than option.
  • Value: Enter the condition value.
  • Compared to: If % Increases/ Decreases by more than is chosen, you can then choose to compare it to the previous day, week, or year.

Once you have entered all of the necessary information click Save Alert, and hopefully, by now, you have successfully added an alert for one metric/dimension combination.

Useful Google Analytics custom alerts

Here I am going to map out some useful custom alerts that are essential along with some more advanced alerts.


Page 404 Alert

Alert Name: +10% 404 Increase (WoW) – xyz.com

Apply To: Production (Filtered View) and

Period: Weekly or Daily (depends on your preference, business model, frequency etc)

Alert Conditions

Condition 1:

  • This applies to: Page Title                    
  • Condition: Contains
  • Value: Not Found (Title Tag displayed when 404 error is returned on site)

Condition 2:

  • Alert Me When: Session
  • Condition: % increases more than
  • Value: 10%
  • Compared to: Previous Week

Organic Session Decline Week-Over-Week

Alert Name: Organic Sessions Drop -20% WoW (xyz.com)

Apply To: Production (Filtered View)

Period: Weekly

Alert Conditions

Condition 1:

  • This applies to: Medium
  • Condition: Contains
  • Value: Organic

Condition 2:

  • Alert Me When: Session
  • Condition: % decreases more than
  • Value: 20%
  • Compared to: Previous Week

No Transactions Recorded

Alert Name: No Daily Transactions (xyz.com)

Apply To: Production (Filtered View)

Period: Day

Alert Conditions

Condition 1:

  • This applies to: All Traffic

Condition 2:

  • Alert Me When: Transactions
  • Condition: is less than
  • Value: 1

Social Media Surge

Alert Name: Social Media Surge +100 (xyz.com)

Apply To: Production (Filtered View)

Period: Day

Alert Conditions

Condition 1:

  • This applies to: Medium
  • Condition: Matches Regular Expression
  • Value: twitter|facebook|instagram|reddit (this is the regular expression format for social media sources add according to the ones you want to track)

Condition 2:

  • Alert Me When: Session
  • Condition: increases more than
  • Value: 100
  • Compared to: Previous Day

Advanced Alerts

High Bounce Rate on Paid Traffic

Alert Name: Bounce rate on Paid traffic > 10%

Apply To: xyz.com – Master

Period: Day

Alert Conditions

Condition 1:

  • This applies to:  Google Ads: Ad Group
  • Condition: Matches Exactly
  • Value: “ad group name”

Condition 2:

  • Alert Me When: Bounce Rate
  • Condition: % increases more than
  • Value: 10
  • Compared to: Previous Day

Low Revenue Alert

Alert Name: Low Revenue

Apply To: Master

Period: Day

Alert Conditions

Condition 1:

  • This applies to:  All Traffic

Condition 2:

  • Alert Me When: Revenue
  • Condition: is less than
  • Value: 1000 (or your business’ low revenue threshold)

Low Conversion Rate

Note: In order to trigger alerts for conversion rates, you must have set up a Google Analytics Goal with its goal being either a purchase or what you consider to be a conversion in your funnel.

Alert Name: Low Conversion Rate

Apply To: xyz.com – Master

Period: Day

Alert Conditions

Condition 1:

  • This applies to:  All Traffic

Condition 2:

  • Alert Me When: Purchase (Goal 3 Conversion Rate)
  • Condition: is less than
  • Value: 1 (i.e your business’ low conversion threshold)

Google Analytics Alerts: Drawbacks & Limitations

Static Google Analytics alerts most definitely make our life easier, as we saw from the above, there is a multitude of static alerts that can be set for a variety of analytics metrics.

BUT, is your business static?

I’m assuming the answer is most definitely no.

As your business or organization evolves, so do its metrics. This causes static alerts to quickly become irrelevant to the current business structure and thus, must be adjusted constantly to prevent false alerts. A good example of this is again with revenue, let’s say month 1, your business is new and your revenue is low – your alerts will be set accordingly. But, as your business grows you might set your low revenue alert at a higher point in comparison to the past, and this leads to the main drawback of static alerts.

If you run a relatively low-growth, low-impact business with a finite number of metrics (under 6) to track; that are relatively stable and static then yes, by all means, set up alerts.

But if you are part of a data-centric, fast-growing, and metrically dense business, then you need your own team of analysts just to deal with the constant adjusting and tweaking of the analytics alert rules.  This can get very pricey and unrealistic very quickly.

On top of all that cost, even if you have the team of analysts to work with and constantly tweak the alerts; what do you do with the data you’ve been alerted to? You still need to chart and visualize the data and try to make sense of it all.  Even after visualizing everything, doing metric tracking and analysis with manpower opens you up to a host of biases and human errors.

A solution to this madness

Static metric monitoring with static analytics alerts manually done by manpower is becoming increasingly inefficient. As more data flows in, there are more metrics, and therefore more rules, more alerts, more analysts, more costs. Is there a better way?


Today with machine-learning algorithms, automated and constant metric tracking is a reality.

That’s what companies like Millimetric.ai have set out to do.

One of Millimetric’s main features is that it can automatically alert you to pertinent changes in your metrics. More importantly, these automatic alerts aren’t for a specific group of metrics you predefined, the alerts are for all your metrics across multiple data sources.  

Say good-bye to the team of analysts you’ve employed to tweak rules, analyze your data and derive insight from it. Millimetric does all of that for you.

Millimetric is not human, therefore, the types biases and human errors that would exist when you employ a team of analysts instantly disappear.

What are you waiting for?

Instead of hiring a slew of analysts and buying thousands of dollars worth of BI Tools – try Millimetric – you’ll reap the benefits of being alerted without the hassle of creating them.

Millimetric.ai creates thresholds and rules automatically from your business’ historical data and automatically alerts you to any anomalous activity. To put it simply, Millimetric is the opposite of static, it’s variable, just like your business.

Get Started with Millimetric Today!

Revenue, if I had a dollar for every time someone told me that this was their most important metric, I’d be a millionaire. “Let’s increase revenue, and decrease costs”, doesn’t seem very helpful. That’s because it isn’t. We know that Revenue = conversion rate x sessions x average order value, so we can use this formula to lead us to metrics that we can use. Once we have all the building blocks of revenue tracked we can now find the factors impacting revenue with more granularity and accordingly can take action on those metrics.

Let’s break down this formula into its parts and look at them individually along with some other metrics that will help understand revenue more holistically.

Conversion Rate

The conversion rate is a ratio that indicates how many transactions you make per 100 visitors. It’s a great indicator of marketing efforts and the traffic those efforts attract. If your conversion rate is slipping, you should start by spending time diagnose potential reasons as to why. Here are some, amongst many, potential problems that affect conversion rate;

Technical Problems

  • Site speed
  • SEO
  • Mobile compatibility
  • Payment problems

UX Issues

  • Distracted objects
  • Low-resolution product photo
  • Cart problems

Product Issues

  • Price
  • Inventory
  • Product variations

Marketing Problems

  • Wrong audience
  • Unrelated landing pages

As you can see from the bulleted list above there are a host of areas that link to conversion rate performance. Merely tracking conversion rate as a sole metric is too general, it’s important to track all metrics that impact conversion rate. Accordingly, when you have a conversion rate drop you can identify the affecting metrics and take action.

Average Order Value

Average Order Value (AOV) is a metric that helps you understand your customers purchasing habits. The better you know your customers’ purchasing habits, the better you can deploy a more effective marketing campaign. For example, if you were to run a marketing campaign for items on your website where your customers’ AOV is 120 USD, but advertise an item priced at 200 USD, you may not get as much return on that campaign.

AOV sheds light on which products should be kept in inventory and the overall sales tactic of those items. Just like with campaigns you don’t want to be keeping items in inventory that are much higher than your AOV, these items could be difficult to sell if you don’t have a strong upsell plan in place. Accordingly, you can use AOV to help you focus on your cross-selling and upselling efforts.

Bounce Rate

Bounce rate measures the rate at which a visitor comes to your site and leaves with no interactions. Businesses want to monitor this metric because it’s a good indication of their websites “stickiness”. Stickiness refers to the overall appeal of your website to the user, this could be (amongst many) the relevance of information provided, a clear website UX/UI, page load time, incoherent terminology. If your bounce rate is going up you immediately need to start diagnosing potential issues, but as shown in the previous example there can be a wide array of problems spanning multiple departments, where would one start?

Page / Sessions

Pages / Session is key in determining the overall “flow” of your site, are people having trouble navigating through your website? Is your interface slow and laggy? This metric works closely with your conversion sales funnel, the closer pages/session is to the number 1 the less likely it is your customers are going along your conversion path. Ideally, this ratio should be closest to the number of pages you need for a conversion. Navigating through your website with ease is key for SEO, page/session is a good indicator of how easily users can navigate through your site once they have landed.

Tracking pages/session can help tackle issues such as how to categorize products so that the customer reaches the desired product more quickly. All site improvements on the user experience or user interface front will encourage the user to stay on your site and increase the odds of a user conversion.

Automated Analytics

Revenue is a general goal that all businesses look at, by following the metrics above you can look at sales driving metrics with more granularity which leads to actionability. Breakdown revenue into conversion rate, traffic, and AOV and now you can focus on those metrics individually and make a decision on how to improve them. But as you drill down deeper into the equation you realize that you need to track more and more metrics which can be overwhelming.

This creates the need to automate the analysis process to alert you to specific issues relating to conversion rate drops or spikes in bounce rates but more importantly providing the root cause for those issues so the user can take action by sharing the anomaly with its root cause to the necessary people to fix the problem.

Get Started with Millimetric Today!

Business Intelligence – Then and Now

Howard Dresner, President of Dresner Advisory is partially known for readapting the term “Business Intelligence” to “describe making better business decisions through searching, gathering, and analyzing the accumulated data saved by an organization.” He adapted that umbrella term in 1989; it can easily be said that data along with business intelligence has come a long way since then.

In the 1980’s we saw the growth of data warehouses, this was the first sign of the breadth in which data was headed. As technology progressed (Moore’s law) and hard drives consequentially got cheaper, so did data storage. The architecture of data warehouses was developed to help transform data coming from operational systems into decision-making support systems. For example, enterprises were now able to store sales data along with the respective time through a data warehouse, which enabled them to plot and analyze data in order to make decisions.

The chart above published by Mckinsey&Co shows the level of change in multiple industries’ core business practices brought on by data and analytics for the past three years. Published Jan. 2018 1

Following the onslaught of data warehouses, the 1990s were the era of data mining. Businesses were now able to house much more data, and technology allowed this to be done faster and more efficiently. Data mining focused on discovering patterns in large data sets, and this enabled companies to do predictive analysis through historical patterns. Businesses now began focusing on pattern recognition in data in order to make more educated forecasting decisions.

By 2005, the data collected on average was so large that business intelligence tools simply could not cope with the volume – requiring the invention of new data analysis technologies such as Hadoop. Accordingly, the term “big data” was coined by Roger Magoulas in 2005 to describe data of this size.

Data-Driven Initiatives – What do businesses want from their data?

According to a data analytics survey by IDG, the top three challenges organizations hope to solve through data-driven initiatives are: finding correlations across disparate data sources (60%), predicting customer behaviour (47%), and predict product or service sales (42%)2. Businesses are looking to make data-driven investments in order to bring further insight into their operations and segments, helping them predict trends. Businesses can take a few avenues in order to reach these goals.


The analysis of big data today is reaching a point of inefficiency from a personnel standpoint. You can always hire more analysts but the marginal return of every analyst you add to your team will decrease with time and will open you up to a host of personnel related issues. Entry-level data analysts have a yearly national average salary of $40K3, after a certain point, scalability is no longer viable through manpower. The insight deduced from the data will not justify the cost.


Pioneers in the technology industry such as Netflix and Twitter have decided to tackle the big data analysis issue by developing their own AI-powered Anomaly Detection and Automated Analytics platform. The investment made by these tech giants shows us a few things. Firstly, if Netflix and Twitter are both willing to invest the time and money it requires to develop an automated analytics tool, then clearly they see value in this tool. Secondly, Netflix and Twitter have $111B and $21B dollar valuations, making the in-house investment for anomaly detection plausible and fundable.  Thirdly, they are best practice creators, their investment into automated analytics shows where the industry is headed in terms of data management. But is automated analytics just a game for giants?

Automated Analytics

Data analysis is forcibly moving toward automated analysis. Automated analysis is done through machine learning; the software constantly scales through data to find insights that are important to your business. A business that aims to reach their data-driven initiative goals (mentioned above) will find that automated analytics through Software as a Service (SaaS) providers is the quickest route to take. Automated analysis tools through SaaS providers are quickly scalable and can generate insights more quickly at a fraction of the cost. As shown above, we saw that the in-house coding of these tools may be too costly for SMBs.

Looking back at the growth of data and accordingly the expenditure on data-driven initiatives it is fair to say big data is here to stay!

What Millimetric Offers

Millimetric is a SaaS automated analytics provider that offers you what the tech giants are

developing in-house and what you can’t scale with manpower. We provide you with:

  • Anomaly detection email alerts
  • Root cause analysis for technical issues
  • 24H Anomaly monitoring
  • Anomaly Reports and analysis tools
  • Anomaly Share & Collaborate

Don’t take our word for it, though, give it a shot and let us know what you think!

1 https://www.mckinsey.it/idee/fueling-growth-through-data-monetization
2 https://cdn2.hubspot.net/hubfs/1624046/IDGE_Data_Analysis_2016_final.pdf
3 https://www.ziprecruiter.com/Salaries/Entry-Level-Data-Analyst-Salary

Get Started with Millimetric Today!

When it comes to business analytics it is crucial to keep your metrics in clear sight. The metrics that we keep an eye on are specific to the relative departments, or if you are a business owner your metrics might differ from operational sector to sector. We see that in today’s world data is king, and accordingly what we do with that data and how we analyze it could be the difference between a company that goes boom versus bust.

Where data is today?

Businesses today are striving to become more data-driven in hopes to get a clear picture of what is going on in their business, their sector and the overall economy. Over the last 15 or so years, we have seen businesses (especially online businesses) segway into using some form of dimension and metric-based data analysis software. As data compiling got more and more specific and holistic the data load for analysis has grown tremendously. Today we see that using dashboards are coming to a glass ceiling.  Not only has it gotten harder to keep track of all the necessary metrics, but it has also gotten even more difficult to derive meaning from the current metrics we have.

When you need to (and can) track so many metrics it is that much more crucial to hypothesise about the relationship between said metrics. This is where the analysis of the data comes into play, however, this is highly prone to human error. Not only are humans limited in their ability to be aware of all metrics, but their analysis of said relationships may also be flawed, due to improper conclusions or lack of data. This is where a missed metric can be costly for your business. We are going to dive deeper into this notion with an example.

Let’s look at an e-commerce business that produces clothing and sells it on their website. Some of the metrics it would be driven by are revenue, conversion rate, and shopping cart abandonment rate. When looking at these generalized metrics it is important to understand the thousands of factors that impact these e-commerce metrics. For example, if on any given day your shopping cart abandonment rate increased dramatically, and if it is due to a metric that you have not been following it could take you a very long time to find the source of the issue. Here you can see how any overlooked metric can be costing your business money.

Where to go from here?

The solution to this is anomaly detection. Millimetric can analyze and scale through data at a rate that cannot be done through manpower. Most importantly, Millimetric will make sure no key metric goes overlooked and accordingly uncover relationships you never knew existed. With this added insight Millimetric is able to find the root cause of issues so that you can not only see the affecting metric, you can go to the source of that metrics issue. Accordingly, you can take the necessary steps to fix the issue.

Anomaly detection is the only logical path to help businesses steer their ship through the sea of data. It is important to have a system that alerts you when you need to be alerted, data needs to be constantly scaled and analyzed for insights. The need to do this poses a problem with execution. The data load is simply too much to be analyzed at such great detail that anomaly detection tools such as Millimetric are a must.

Get Started with Millimetric Today!

Michael is the director of e-commerce for a leading retailer in the UK. He generally starts his days off by checking his key e-commerce metrics, looking over revenue charts, keeping a close eye on conversion rates simply to analyse them and make sure everything is ok. But what happens when something isn’t ok? What kind of steps does Michael have to take? Let’s take a look.

It’s another day at the office and Michael spots something. Conversion rates are down in comparison to last week. In a hurry, Michael decides to email his marketing director Kyle:

“Kyle, conversion rates are down, what going on?” Kyle writes back,

“Not sure, I am looking into it”

Kyle gets back to Michael a few hours later with a reply stating:

“There is a problem with checkout form submission for Apple iPhone users, you need to talk to IT”

Michael forwards the email to IT only to receive a reply stating “We’re looking into it”

The following day, IT gets back to Michael to let him know the issue is resolved.

What Happened?

It’s important to take a step back and understand why this is inefficient. As with the case above you can see that most of the time was spent diagnosing the problem, emailing back and forth only to fix the issue the following business day. Michael was relieved they sorted out the issue yet upset about the revenue lost in the meantime. Could have Michael solved this issue more quickly? Let’s look at the same scenario had Michael used Millimetric’s anomaly detection software.

A Different Type of Day

Another typical day, Michael wakes up to an email in his inbox titled “Critical Alert: E-commerce Conversion Rate 27% drop”, Michael clicks on the anomaly and is redirected to Millimetric’s website where he clicks on “root cause” to see the source of the issue. He sees the mobile device related breakdown of conversion rates and sees “iPhone down 27%” he quickly shares this anomaly directly with IT, bypassing Marketing because he now knows it’s an IT issue. IT replies “That was a quick fix, thanks for the insight”

By simply using Millimetric Michael could have bypassed a whole day worth of lost iPhone-sourced revenue.

This is just one of the thousands of aspects where Millimetric’s anomaly detection diagnosed the problem, triaged it and delivered an actionable insight within minutes!

Great story, but what is Anomaly Detection?

An anomaly, or outlier, is defined as something that does not fit the expected pattern. In other words, an anomaly is a deviation from usual data characteristics. An example of this could be an unexpected increase in advertising costs or a sudden drop in sales. At its core anomaly detection is about sorting data into groups of regular and irregular data. A data point that falls in the irregular data group is called an anomaly or outlier.

How Can Millimetric Improve Your business?

Millimetric automatically notifies you if an anomaly has been detected within the provided data. Simply add a data source (e.g. Google Analytics, Facebook Ads etc.) and our Machine Learning Algorithm starts to understand the normal, by analyzing data from the past. Then Millimetric detects unexpected changes, incidents or problems and reports the must-see data on a daily basis.

The ability to quickly find and resolve business incidents enables you to prevent revenue-loss and keep costs at acceptable levels. But, its not only about resolving issues that arise, Millimetric also provide positive anomalies. The idea is to create levers within your company to help you lessen the bad and increase the good.

Get Started with Millimetric Today!