Revisiting Freshdesk as a Product Analyst
Freshdesk was my first B2B SaaS application. As a pre-sales engineer at Freshworks, I was trained to solve a business’ customer support problem/use case using the product. But over the period of my tenure at Freshworks, I became curious to understand how and why product decisions were made. In addition to the high-level business goals such as revenue, product-level decisions were made leveraging data to improve the overall customer experience.
Over the weekend, I completed my micro certification in Product Analytics from Product School using Mixpanel. My key takeaway was the importance define a question and then identify the data to be monitored that can answer the problem at hand. This learning intrigued me to test out Freshdesk once again. Not as a pre-sales or an end-user but from the lens of a product manager/analyst.
1. What is Freshdesk?
Freshdesk is a cloud-based online customer support tool that lets you integrate different channels of communication such as email, social media, web portal, widget, chat, and phone into one single platform. This unifies the customer support experience both for the end-user(customer) and the agent (the customer support staff).
With the recent change, Freshdesk is split into 4 different categories — Omnichannel suite, support desk, call center and customer success. For the remaining of this article, the focus is on the Freshdesk support desk, and the different metrics associated with it.
2. Customer Acquisition Metrics
From the lens of a marketer and product manager, the customer acquisition metrics help identify the channels using which customers/prospects get access to Freshdesk. The different channels of customer acquisition in addition to direct search include but are not limited to Google Ads, Referrals, Email, LinkedIn Ads, Social Media Ads etc. Analyzing the data from Similar Web indicates that the major channels of acquisition are direct search, referrals and ads.
Customer Acquistion Funnel
The customer acquisition funnel helps define the user journey from the point an awareness is created till a purchase is made.
My description for the different stages of acquisition funnel for Freshdesk as below-
When is a customer considered as acquired for Freshdesk?
I would define a customer to be successfully acquired if they sign-up for a free trail and complete the purchase (through the product or sales assisted) within a n-day period. Given that Freshdesk has a 21-day trial period, if the value of n is less than or equal to 21, I would categorize it to be a default trail conversion and any value greater than 21 would be categorized as an extended trail conversion.
Given the different marketing channels of Freshdesk, I would define the top 3 questions in the user acquisition phase as below-
- Which acquisition channel has the highest trail sign up percentage and conversion percentage?
Helps identify the most effective channel to capture the right audience for Freshdesk - Which country / region has the highest and lowest conversion percentage?
Helps strategize and budget marketing efforts across different regions - Which country / region has the highest extended trail conversion?
Helps design and run experiments for an extended trail period and monitor the resulting conversion percentage
3. User Behavioral Metrics
Freshdesk offers a 4 tired pricing — Free, Growth, Pro and Enterprise. The free trial starts with the Enterprise plan by default but allows user to downgrade or upgrade to any plan during the trial window. The features in each plan vary and this indicates that the product adoption metrics will also vary across plans.
A verified email during sign-up and setting up the support channels are vital to test the product. Hence, assuming a new trial sign-up in any plan, I will define top 3 questions that will help understand the user behavior with the product as below-
- How many users verify the email address and what is the time to verify?
Verified users indicate an interested prospect and the time to verify helps identify the known / unknown friction associated with email verification. A breakdown of the email inbox (Gmail, Outlook etc.) helps optimize the email verification process. - How many users set up at least one support channel and the resulting conversion percentage?
Given Freshdesk Support desk depends on the support channel integrated, this helps identify the most common support channel and the resulting conversion percentage for users with one support channel and multiple support channels. - How many unique tickets are created and how many responses are sent out for the tickets? What is the resulting conversion percentage for each bucket?
The number of unique tickets created indicates a successful set up of a communication channel and the number of replies sent out is a measure of product adoption. Though a ticket can have multiple replies, I would be interested to track the number of unique tickets and replies as a measure of adoption during the trial phase and the subsequent conversion percentage.
Suggestion: Despite being a B2B tool, Freshdesk does not capture the industry or vertical of an organization (SaaS, Travel etc.) during the onboarding process currently. Doing so will help curate a personalized onboarding experience.
4. Customer Retention Metrics
The most interesting aspect of product analytics is leveraging product data to prevent or reduce churn. Freshdesk offers two billing options and hence my definition of a retained customer differs based on the billing type.
Billed Monthly
In this scenario, customers are charged every month based on the plan and the number of agents having access to the tool.
A retained customer is easier to define in this case and is anyone with an active subscription status at the beginning of their billing cycle.
Billed Annually
In this scenario, customers enter an annual contract but are charged every month for the entire contract duration.
In addition to an active subscription status, I would consider the product activity as a measure to define an account as retained in the annual billing type. Though rare, it is possible that an account has an annual contract but no activity which might result in a churn in the next billing cycle.
What does an active account mean?
Defining an account to be active varies from product to product and also on the context. From the point of view of retention, I will consider a Freshdesk account to be active if it satisfies the below conditions
- A steady or growing ticket creation and response rate month on month.
- Agent login once in n-days and/or increase in agent count. The n-day period can be defined by analyzing historic retention data.
- Active subscription status with no billing failure during the contract period.
In addition to the frequency of usage, account retention definitely depends on the time to derive value from the product. I would define my questions for account retention as below-
- What features when set up and used result in higher account retention?
This helps identify the list of features that deliver maximum value and when bucketed into cohorts based on the plan, agent count, ticket count and business vertical will help curate the recommended feature list. - What is the time taken to set up recommended features and the frequency of use of these features?
A shorter duration to reach the Aha moment will result in better adoption of the product thus resulting in higher retention. - What is the activity step that users perform before canceling the subscription?
Identifying the pattern that accounts exhibit before cancellation, such as reducing the agent count or removing existing integrations, will help take a preventive call against churn.
Another important learning from the course was the necessity and importance of running experiments to continuously improve the product experience and user journey. Comparing the performance of the experiment group with the control group will help implement effective changes. Though the experiment results are derived based on the performance of dependent variables (successful email verification, faster WhatsApp integration etc.), it is important that we factor in the independent variables (region, business vertical, MRR) to reduce any known / unknown bias.
Conclusion
Though not an exhaustive list of questions, I believe the questions are valid for any B2B tool. There is a significant difference in my thought process when I look at Freshdesk as a product manager/analyst from when I was a pre-sales engineer. From thinking about how the product can solve a use case, wearing the hat of a product person helps me think of answers to what Freshdesk solves and why it solves that way.
If you are an analyst or a product manager or even otherwise, I am happy to discuss this and take your feedback to help phrase better questions :)
