How To Plan Google Analytics Alerts To Improve Your Site or App Traffic

Analysis can be fun, discovering new marketing ideas from the data collected, but in many instances it is counterproductive to check metric changes daily. While advertising campaigns need it to make sure budgets are not exceeded, daily checks can lead to overlooking trends that develop over a long arch. Trends are the main indicators of behavior behind the data, and they can be best viewed through periodic checks and smart comparisons.

The key to making smart comparisons and avoid wasting time on analytics is to examine when certain trend changes occur. Measuring metric lifts or declines is the best way to decide when analytic reports need reviews. With so much data entering a site or app today, it’s better to let analytic alert you when something unusual happens rather than spend time trying to find potential differences.

Alerts are sent as email notifications. Google Analytics calls these alerts intelligence events. In Google Analytics you can set alerts via the admin panel, click on the custom alerts, and then set the dimensions, metric, and segment categories that will trigger the alert.

Three choices for metrics can be used for an alert:

  • changes in volume (increase by 10, decrease by 15 visits)
  • a threshold percentage (%)
  • a threshold amount (less than or more than).

So, once at the alert panel, what settings are worth your attention? Well, here are a few basic settings ideas that you can plan from the start.

Traffic spikes

Google Analytics Alert

This falls under the “duh” category, but basic traffic spikes are what we want to see. The issue is how much of a lift should be monitored. The best starting point is to set a 10% increase as an arbitrary starting point.

Now, low volume (about 600 sessions or below) will probably trigger notices every other day for small spikes. That has happened for Zimana at the beginning. But using a 10% setting will trigger less often as for a site that is developing traffic over time. Consider revising as the overall volume grows to a steady amount per month.

Let your knowledge of the X visitors per day decide if a larger percentage is more reasonable. But before you set a number, just ask about the value expected when a number doubles or even triple.

Traffic drop

When something goes up, something can come down. Every decline may not be worthwhile, but consider a percentage decline the same way you would for the traffic spikes. Again adjust as volume increases. A significant drop will reflect a coding issue with the analytics, site, or app.

Changes within known traffic segments

Google Analytics Alert Sources

Alerts can also work for specific traffic segments. Examples include traffic associated with an AdWords campaign or a custom segment set in Google Analytics. Segmentation by traffic source, media, and referral path are other ways to highlight a given set of traffic.

Spike in goal completions

The idea behind this is similar to the traffic spike, replaced with a focus on goals completed by site visitors or app users during a session. Create an alert for raw number of completions, but also set alerts for intended demographics, such as users that arrive from a specific city, region, or with language. Goals should be triggered but mainly for granularity that would be of interest in a business operation, like the percentage that resulted in a conversions or the number of people who started conversion activity but didn’t complete it.

Drop in goal completions

Just as a goal spike is analogous to a traffic spike, a traffic drop is analogous to goal drops.

But goal alerts do not just have to be for web site or app metrics. There are other metrics available in an analytic solutions that can be monitored. E-commerce is exactly that “other” that typically comes to mind to most analysts.

Ecommerce Trends

Google_Analytics_E-Commerce
Using E-commerce alerts can help your team understand what products are being viewed frequently, leading to what might be selling quickly online.

For an e-commerce site you can set alerts for a number of financial metrics related to a purchase – unique purchases, product, revenue, shipping charges, and tax charges. Monitoring these can help understand how a site or app contribute to cost. These can also lead to understanding how frequently these costs are occurring.

Alerts in AdWords and Bing Ad campaigns

Google_Adwords_Manager_Alerts
There are also AdWords Alerts in the AdWords Manager as well, but these alerts are meant more for campaign changes rather than website related behavior.

Alerts can be helpful in deterring upswings in digital ad performance, from increase clicks to increase conversion. Campaign managers such as those for Bing and Adwords do have stop campaign alerts, to let users know when a campaign has been suspended because a preset limit has occurred.

Using alerts can make monitoring your data less time consuming and allows you to focus on other aspects of your marketing or within your business.

IoT Primer: Beacon Basics For Retail Analytics

Maybe the joyous reaction of marketers to beacons, low-cost sensors that can communicate with tablets and smartphones, comes from the fact that the word’s spelling is close to bacon – seems to be a lot of memes suggesting that bacon is a favorite food.  But let’s stick to a more straightforward reason:  Marketers are as excited about beacons as a child is excited about Christmas day.

So what is the beacon excitement about? Here’s the run down:

  • The idea of using sensors in commerce started with radio frequency identification (RFID) tags. These tags were used in retail and operations mainly for inventory management and shipping identification.
  • Beacons represent the next “evolution” of sensor usage because they can communicate with apps on smart devices.  The apps are activated by shoppers, allowing the device to be picked up by the proximity beacon.  This establishes proximity broadcasting, which means objects can continuously “communicate” information. The devices and sensors communicate in a data broadcasting environment that pairs mobile devices with beacon sensors on a physical item – a display, a store shelf, or in a store location.
  • A network of beacons are popular because mobile-savvy customers have opted-in through using an app provided by a retailer which has beacons placed in a retail space. Smartphones, tablets and laptops already use Bluetooth, so there is a natural usage already primed for beacons acceptance.
  • The latest generation of Bluetooth, Bluetooth Low Energy (BLE) 4.0, expands the breadth of beacon sensor capability. It includes metadata association to the objects, allowing objects to essentially broadcast their identity, location, and additional information.  The data shared in this environment gives users valuable information, from on-the-spot announcement of a sales offer to simple assistance in finding the right product selection.
  • All of this plays into the Internet of Things (IoT) environment, in which machines are now communicating signals to each other in a coordinated effort.

Where are beacons being used in retail?

Luxury brands are becoming the first movers in experimenting with beacons as part of an augmented reality strategy. Sephora, a make up provider, announced beta testing beacons in its San Francisco store.  And Burberry uses augmented reality technology in its flagship store in the UK.

So what does beacons mean for analytics?

Beacon-generated data can then be pulled up into a corporate cloud application for tracking and analysis, influencing the business intelligence and the analytics associated with it. Real time analytics that include beacon-sourced referral traffic must consider digital activity over time.  Additional analytic tools highlighting foot traffic patterns can aid where message are being triggered, and facilitate inventory planning.   Integrating the associated metrics mean increased tagging complexity, so marketers must consider what capabilities lie within their tools.

Adoption is still early – eMarketer reports that 29% of retailers have incorporated beacons into their stores.

But as the IoT environment emerges from the budding stages, retailers will learn how to use beacons to determine how well their business model aligns with the customer experience needs.  Customers are demanding sophisticated experiences. Beacons provide the the building block for understanding those experiences.

Good Starter Questions for Organizing an A/B Test

A/B testing and optimization does not always sound like a sexy subject to talk with someone (if that someone is not an analytics practitioner). Neither is going to the dentist for some people.
But just as people do make an appointment with a dentist, managers need to make an appointment with their analytics teams to understand what content and app features are working to attract customers.  Thus managers need to have discussions about testing and optimization plans.
In starting discussions about testing and optimization,  a few simple questions are worth asking.
  • What testing is needed in the overall digital strategy? What website elements, qualities or user actions are repeatable to warrant testing and optimization?  These questions highlight what the test is meant to accomplish – essential asking “What is the test for?” The questions also highlight what element in a website or app is being questioned.
  • Which internal organization owns the Testing / Optimization process? In short, who is running the test and responsible for the reporting?
  • How can this testing support the analytical needs of the marketing campaign?  In other words, how does this support a goal for the organization?
  • What obstacles hinder testing?  What resources are needed?
Consider the following as those thought-starter questions that can lead to a healthy discussion and decisions on how to best approach a digital presence for an organization.
Polymer.js example

How Semantic Search Relates Online Content to Customers

Polymer.js example
This HTML mark up is a technical example of how semantic search works. It calls out specific parts in the code, like this callout designed to address a version of JavaScript called Polymer.  Although JavaScript case like this is not an influence on SEO, most semantic search is meant to address highlighting specialty HTML elements in a search query.

Search engines have subtly changed their methodologies over the past few years. One methodology that has yet to see widespread adoption has already experienced its most significant change yet.

Back in 2013, an ontology library site, Good Relations, announced an alignment of its markup structure definitions with those used on schema.org, a metatag library. The end result is increased consistency of definition usage among businesses, and a wider shared usage of structure markup among web developers and search engine optimization practitioners.

This merger occurred thanks to increasing search discovery needs for digital media. From music to webinar presentations, businesses have added numerous content to appear when potential customers research product and service information online.

The content has led marketing managers to give a refreshed look at their optimization strategies through apply semantic search. Semantic search involves organizing keywords and content with website element protocols and structure markup language. The organization makes the pages and site content more visible to nuanced search engine queries.

Good Relations and schema.org support separate protocols for semantic search. Schema.org contains metadata meant for HTML5, an update of the venerable website structure code language positioned for future website development. Good Relations contains RDF – resources description framework that has proven utility for current retailers and E-commerce sites.

One strategic benefit for managers is learning enhanced ways to translate potential client language to its digital properties. If businesses within a given industry agreed to ontology for services and needs, those businesses can adjust their content tags to position its content to potential query results from those businesses. Imagine a video on better financing for construction projects – With a metadata protocol, now imagine that video section appearing in a search query run by a construction firm.

That exact example lies at the heart of HTML5. HTML5 added video- and music-related tag elements, developed to increase media exposure to relevant search engine queries. Other tagging protocols can help search engines recognized a group of authors – an aid to marketing teams leveraging personal brands of its members online (You can learn about what Google accepts in structured markup here – Bing also has a structured markup guide). The fundamentals of digital marketing is increasingly shifting toward strategic data ownership, which is supported through content marketing and semantic search.

With the schema.org – Good Relations alliance, digital marketers and website developers can optimize metadata and RDF information across varied content. The success of such an effort will create a true application of semantic search’s definition – the science of actual customer’s language.

The “James Bond” of programming: Javascript And Its Importance To Business Development

James Bond Skyfall Image
If James Bond can be rebooted, as Daniel Craig has done back in 2008, then so can programming languages.  That is the backstory for JavaScript, which has grow in capability and importance since its pop up window days.

JavaScript has been rising in importance for web and app design over the past few years. It’s an old programming language, much like James Bond is an old movie series. But just as Daniel Craig revitalized the series back in 2008 with Casino Royale, culminating in 2015’s Spectre,  JavaScript is being revitalized for a new era of mobile devices and IoT that was not in existence in its 1996 debut in Netscape.

Part of the reason stems from the canvas element in HTML5. Because it is called out in a container, called <canvas>, Javascript can be used to manage the elements. This gives Javascript the capability to manage elements in an unprecedented way, such as data visualization.

Another source is a growing number of Javascript libraries for interacting with the document object model. Originally Javascript was used for pop up windows – to instruct the browser to show another window automatically – or infamously, depending on the content.

But new uses developed, requiring more interaction with the DOM in the browser. The DOM is an operational code within the browser that is accesses to tell the browser how to render a page or another media that appears in a browser.

The business value is the capability for developing browser features that the user can see and interact while cutting calls to a server and reducing time to render a site or app.

A dashboard with visual elements, telling the story of trends that lead to business decisions, is a great example. The data can be called without multiple calls to a server, permitting visualization for tablet devices.

You can read how JavaScript is influencing technology in the following posts:

CMSWire – How To Use JavaScript for App and IoT Development

IOT Solutions – The IoT Ushers In the Ascendance of JavaScript

Also check out this Zimana JavaScript post on coding basics that can help you understand JavaScript better for your website and app development plans.