Blog

Log Data For Business Insights: How Marketing at Logentries Uses Logentries

Here at Logentries we’re dedicated to making log data simply accessible. So dedicated, in fact, that we eat our own dog food across every department. Our Co-Founder, Dr. Trevor Parsons, recently blogged about How To Track Business Metrics with Logs in 3 Simple Steps. But I wanted to show how less technical people use Logentries within the company. While I have years of experience in the tech industry at various start-ups as well as bigger companies, I’m by no means an engineer. However, Logentries has been designed with ease of use in mind and is accessible to those like me who are not command line or programming experts. So, how exactly do we use it from a marketing perspective?

Tags

We use tags to monitor events by both new and existing users. For new users some examples of tags we have set up are:

  1. “Sign Up” (trial)
  2. “Signup Paid”
  3. “Data Received”

For existing users some examples of tags we have set up are:

  1. “Switch to Paid”
  2. “Account Deletion”
  3. “Log in”

I won’t run through them all, but we’ve found setting up tags to be simple. For the “Sign Up” tag we just set the pattern to:

/event=’signup’/

Tagging_for_marketing

The other tags that I have created follow a similar process and we’ve been able to set up many tags that we monitor regularly to measure user growth, company health and more.

Alerts

Based off of these tags, we also have certain events that are high priority for us. We need to be alerted as soon as these events happen, so we utilize the “Alerts” functionality of Logentries. For example, every time an account is deleted, we want to know so that we can look into it (you can learn a lot from your losses!). All we had to do was tag the event, then trigger the alert from that. Now we see, in real time, every time an account is deleted…lucky for us those emails don’t come very frequently. When it does happen, though, we can immediately assess the lost account, triage the situation, and seek feedback or look for a win-back opportunity if depending on what happened.

Alerts_for_marketing

Graphs

While tags and alerts allow you to dig in at the individual user event level, the Logentries graphing functionality gives you a higher level aggregate view. In fact, you can quite easily graph whatever data you have already tagged to see how it trends over time (see some graphs taken from sample data we use for testing new features below):

Dashboard_for_marketing

We graph:

  • “Sign Up” (trial)
  • “Signup Paid”
  • “Data Received”
  • “Switch to Paid”
  • “Account Deletion”
  • “Log in”
  • And more

This allows us to see how these business KPIs are doing over time.

Setting up a graph is simple. All we have to do is select what type (pie chart, line graph, bar graph, etc…), the tags we want to chart, name it and we’re in business! Voila, a business dashboard that shows the core data I need…and because it comes straight from the application logs, I know the data is correct. For example, I can easily verify a sudden increase in ‘Sign Ups’ that appear in one of my graphs, by digging into the individual log entries to get a list of who exactly signed up and when they signed up precisely.

Search

Up and to this point, we’ve been talking about monitoring KPIs and seeing how these numbers trend over time. This is great, but can be accomplished in many ways. The next piece is a bit tougher to do. Because we’re working with log data straight from our application, we can see how users are interacting with Logentries. As a marketer, this is really powerful stuff and more data than I can get from something like Google Analytics or a CRM. I can see what users are doing on page, as they select different drop-downs, click buttons, etc…So how do we get this information?

It depends on how your application is set up, but we’re able to search our logs for the query “action=/.*/” over our desired time period and it will return all the user actions. One of our engineers configured our new JavaScript client-side logging library for me to be able to track these actions from the users browser.

One example of how this is being used internally is In our Quick Start where there are several libraries to choose from to set up your logs. One of the big things we monitor is which of these is the most popular. We’re always looking to improve Logentries for our users and this helps guide us as to where we can deliver the most value. We’ll also look at what features people are using and can sort it by paid, trial and free users. What features are popular? What features drive people to go from Trial to Paid (i.e. which features do people find valuable)? Are different features more popular on different platforms? How can we better educate users about the powerful features available to them and how to use Logentries to get the most insight into their application? With this data, I can confidently answer all of these questions and more and this has and will continue to help guide our marketing efforts as well as our engineering priorities. If we know how people are using Logentries, we can make it more powerful and easier to gain insight.

To Wrap it Up

The point of this post isn’t to tell people how to do marketing but, rather, to show how Logentries is used across all departments within the company and how even non-techies in Sales & Marketing are able to derive value from this application. While we work in a tech startup that develops software, we’re not engineers – yet, we can use Logentries to get some powerful insight into how people are using our application. Now, imagine what you could do as an engineer with system and/or application knowledge! Get cracking! Oh – and tell your sales and marketing team that there is value in them logs!

Logentries_Try_It_Free_Promo_W


Share This Post


Posted in Business Metrixs, Logentries

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong> <p>

Current day month ye@r *