Key Insights

•  InfluxData -  600+ customers serving from 1-person startup to Fortune-50 companies
• Support Organization -  non-tiered, roughly 12 people in US and UK. Very remote-friendly. Just started building customer success team. Case volume ~400 per month. How-to / Break-Fix case split 25 / 75
• Support Tech - Used to be Zendesk; migrating to Salesforce ServiceCloud. GitHub open source for docs. Discourse for Community Forums [InfluxData being an open source company promotes use of open source products]
• Case split across channels -  earlier >75% on email.  Now shifting towards web.
• Knowledge Base - Separate KB doesnt exist instead knowledge is added to product documentation. Though this will likely change in future
• Cool Idea -  Crowd sourcing for support answers worked very well at MobileIron (used a vendor called Directly)
• What's broken in Support Tech Stack-- Integration between CRM, Case Management and Bug database is broken. One has to throw tools, consulting dollars to fix it! 
• Focus for next 12 months - Build Customer Success Team
• Online Certification/Education - No standard tool that exists; a few open source tools though
• Metrics - Success : Customer Health (primary indicator product usage), Support: CSAT
• What would you do differently if you were to do this all over again -  Focus on Customer Success deeply from the beginning; that way you understand deeply about customer and product-market fit.
• Books:  

Survival to Thrival: Building the Enterprise Startup - Book 1 The Company Journey

Survival to Thrival: Building the Enterprise Startup - Book 2: Change or Be Changed

Startupland: How Three Guys Risked Everything to Turn an Idea into a Global Business

This is a special episode as John is a Distinguished Researcher at TSIA -- an organization that works with Support Leaders in B2B Technology world. Listen to it to understand what Support Leaders are thinking, what are they doing well, where are they struggling (where they are going to spend money) etc. Must-listen episode!

Key Insights 

• Fun fact: TSIA used to be called SSPA Services and Support Professional Association 
• TSIA maintains a list of Business Challenges Organizational structure: Swarming vs Tiered Structure,  Metrics: What metrics should I be tracking / Whats the average, Self-service / Knowledge-Management, Channel Mix: Introducing digital channels and moving away from phone/email & Customer Experience: How should we track it how to set compensation structure.
• For SaaS vendors, most of the support questions are around how-do-I-something (vs legacy hardware vendors where most questions were around break/fix)
• Chatbots:  17% of members say they have a chatbot in place but most of those are pilots and not launched.  30% of these customers have built themselves (John doesn't recommend doing it)
• Workforce scheduling/assigning cases automatically is still being done manually/spreadsheets in B2B.
• Only 5% of members have a mobile-app for Support. Making support website HTML-5 is not good enough most of support websites have rendering challenges on mobile phones/iPads. 
• 92% customers start with google search and if support portal content is not indexed, customers won't come to the support site. More than 50% of vendors are google-indexed though. Legacy hardware vendors tend not to do google indexing though
• Support issues are going up 20% YoY its hard for support organizations to grow in terms of people to match that demand so automation is a must
• Knowledge Management:  41% vendors responded that they could get 30% or more productivity gains if knowledge was being shared better.
• John is bullish about using RPA in Support. 
• Customer success is not a department but a culture! Also, using Professional Services team to do Customer Success.
• Next-gen Support: Persistent, Personalized, Intelligent
• Recommendation for CEOs: Don't treat Support as just a cost-center. They have the most visibility and influence with Customers & they can share great information and insight with engineering and product to make products better and control churn. Support also feeds into QA and engg teams.
• Recommended books: Bad Blood/Theranos and Applied Artificial Intelligence: A Handbook for Business Leaders

 Key Insights

• Tech Tool Stack: Zendesk (Ticketing, Helpdesk, AnswerBot), Sprout Social (Twitter, Facebook), Mode (Data Science)
• Ticket Volume: Few hundred tickets on a daily basis
• Ticket across channels: 80% Email, 10% (Twitter, Facebook, Instagram, Reddit), 10% (Google PlayStore, Apple App store
• Case deflection with Answer Bot is ~5%
• They have Tiered Support Tier-1/Tier-2 (outsourced), Tier-3 (internal).  8 support employees
• Metrics : First Response Time (FRT), Customer Satisfaction (CSAT), Full resolution time, Customer Request Volume
• They decided to delist from Galaxy Store due to overhead of software maintenance, need for support agents to have a separate testing unit and lack of RoI in terms of users.
• Chat "bots" didnt work for them in the past but they are considering introducing live chat
• Also, exploring customer communities after seeing good engagement on reddit

Key Insights

• When does the role of Chief Customer Officer makes sense in a company's journey?
• As CCO, following teams fall under Bernie Customer Success, Renewals, Alliances (Independent Software Vendors or ISVs like Workday etc, System Integrators or SIs like Accenture, Business Development). 
• Why customer success and renewals are separate for them? 
• Tool Stack: Centered around Salesforce, Khoros/Lithium (Community) Kite (Search), Gainsight (Customer Success), RingCentral (Contact Center), Pendo (NPS)
• Important Metrics Churn, Customer Sentiment, Support Coverage, Training/Revenue of System Integrators. 
• NPS Surveys Divide their customers in batches of 1/6th and run the survey every 2 months that covers all customers in an year. That way you get best of both worlds -- frequent customer pulse without sending too many surveys to everyone.
• Business Books:  3 Biographies -- Steve Jobs, Shoe Dog: A Memoir by the Creator of Nike, Elon Musk and books by Malcolm Gladwell specially The Tipping Point: How Little Things Can Make a Big Difference

Key Insights

• Unlike most B2B companies, Khoros doesn't have tiered support.  Philipe says a good way to think about their support is following --  Tier-0 is the customer, Tier-1 is the support community and Tier-2 is their own support. 
• Khoros provides support for primarily Fortune 5000 companies and their support team is structured accordingly -- 110 people globally. Philippe has 5 functions in his team -- Program Management (Surveys etc.), Information experience (document, in-app support, community), Support Engg (backend support), Enterprise Account Support / Technical Account Management and Technical support (biggest team..almost 70 people). Customer Success is done by a different team.
• Philippe believes that in charging for support separately from product subscription (as long as the support is something that customers value).
• Their technology stack --  No IVR, Salesforce case management, Community (Khoros), JIRA/confluence (Engg), Squelch and Pendo. 
•  Incoming ticket bandwidth: < 1% social , <5% phone (going down), 55% email, 45% community case portal. Philippe's talks about why email is so high for them.
• How to make communities successful:  1/ Provide a reason for customers to go to community ("whats in it for me?". 2/ Customers need to feel that they own part of the community (using gamification to encourage peer responses). 3/ Ubiquitous (insert community in blog posts, marketing newsletters etc.).
• Chatbots can be successful only to the extent that they are option-based or workflow-based.
• How best to organize documentation teams around the concept of Information Experiences (IX)
• Next-gen support will be smooth, low-effort, predictive.

Key Insights

• "Nobody wakes up in the morning thinking that they will call Support".
• Why email is not a great channel for support (a consistent message that I hear from all my guests).
• Okta is a unique B2B company in that they are actively looking at social media for understanding and resolving issues reported by customers.
• 70% tickets get closed by Level-1 support, 15% by Level-2 and rest by Level-3 (most sophisticated).
• Best way to make a case go away is to push information upstream (towards customer) so that the case would not need to ne filed. Requires combing and understanding patterns in data. This "personalization of support" is what will be the bedrock for next-gen customer support. 
• Related themes around case tagging/categorization and case routing (Rick mentioned that if a customer had a great experience with an agent before, it might make sense to direct the customer to the same agent even if the agent may not be the person to solve the case; at least they will take it to completion). 
• Encourage agents to document solutions through Knowledge Creation Program.
• Technology stack optimized around Salesforce.
• Frequency of customer engagements is so high that Support needs to be seen as  extension of the product team.
  • Favorite book -- Shackleton's Way by Margot Morrell and Stephanie Capparell.

Key Insights

o SentinelOne has ~2500 customers (globally) which includes both Fortune 100 customers and small businesses. SentinelOne is also part-SaaS, part-hardware. All this puts a unique challenge on customer support different language support, 24x7 phone service, customers coming through channel (distributors/MSPs) etc.

o Technology stack standardized on Zendesk. Call center is TalkDesk. Search is Zendesk but could be better. Salesforce is the system of truth (CRM) and there is a desire to standardize around that.

o A good part of our conversation was dedicated to challenges in getting the tools to work with each other specifically Zendesk to Salesforce and how customer success requires yet another bolt-on. Queries coming through channel has unique challenges as well (understanding full customer context, making sure that issue context is transferred from partner to vendor etc.)

 Key Insights

o Team follows the operation principle of PADRE - Pipeline, Acquire, Deploy, Run, Expand. Covered here in detail. 

o Technology Stack

Ticketing, Voice, Portal - Zendesk

Knowledge Management - Mindtouch

Online community - Lithium

Search - Solvvy, SearchUInify

Agent Training - Skilljar

o Marlene is currently focusing on following initiatives:

- optimizing Agent training and on-boarding

- understanding the persona of the person calling

- auto case routing (speed + accuracy = customer experience)

o Favorite business books:

Subscribed, Tien Tzuo

Nevertheless, She Persisted: True Stories of Women Leaders in Tech, Pratima Rao Gluckman

Key Insights

o David introduced terminology Rep-first and Rep-experience i.e., Focus on hiring and creating the best experience for agents in terms of giving them full knowledge about your product (especially the bad stuff).

o Dont mix commercial aspects (renewal, upsell) into customer success. Just let them focus on on-boarding, adoption and value add. 

o Tools Completely on Zendesk stack. For data insights, use Zendesk Insights and Google Data Studio.

o Chatbots not ready yet for primetime even for B2C.

o Metrics David said something very interesting here. I tell my agents that dont think about the queue, its my problem. You focus on quality and answering right and answering good. In summary, David measures himself on request-to-wait and his agents on CSAT, and one-touch resolution.  

o NPS hit or a miss because customer is not necessarily rating just the support. David uses NPS, CSAT and monthly interviews with customers to get a pulse of his customer base.

Key Insights

o Hardware vs SaaS support Risk is shifted to the vendor in SaaS world. This is both good and bad. Good because support is now being seen as part of the overall product offering and consequently support organizations now have a stronger voice internally. Bad thing is that if you lose a customer, its on you :)

o Future of Support Exception Center and not a Contact Center.

o No email support they are 2x/3x more inefficient in terms of solving the case as compared to web/chat-based cases. Why inefficient? 1/ Emails usually do not contain full context to solve the case. 2/ Email dont offer real-time communication (like chat)

o Roughly 20% cases come over phone and rest over chat and web. 

o First B2B organization (that at least I heard of) that is moving from inbound calls to outbound call i.e., customers request online to be called back at a certain time. Nir brought up that the biggest advantage in this approach is that when they make the call they already know about who the customer is and their entire context which helps them solve problems faster.

o Tool stack Consolidating around Salesforce (CRM, KnowledgeBase,Search). RingCentral for Contact Center. They also plan to launch chatbots by end of this year. 

o Favorite Business Book The 7 Habits of Highly Effective People / The 7 Habits for Mangers By Stephen R. Covey

Key Insights

o First, no relation to our previous guest :)

o Their attrition is super low ~2% (TSIA says average is about 15%). They have built a collaborative culture in support e.g., Level 3 people do not own cases, they just help other folks close theirs.

o Panzura NPS is 82! Average for tech industry is around 21. One of th

o One interesting metric that Caeli does NOT keep track of is time-to-close. She explains why in the podcast. Caeli also keeps track of this unique metric (which I havent heard before but it actually makes lot of sense) number of tickets escalated to engg.

o Support Tools Salesforce for everything. 

o Caeli mentioned that search continues to be broken. Google search appliance (now EoL) was cheap and easy to setup and it worked. Current solutions are opposite.

o They build lot of tools/scripts internally around product health specially data warehouse that stores all the logs pertaining to a given case.

o Favorite business book In search of Excellence by Tom Peters, and, Grit by Angela Duckworth. 

Key Insights

o Greg introduced the concept of "People-first Revenue Generation"  i.e., put People above Process and Technology.

o 5 Metrics that Support leaders should keep track of:

CSAT - Customer Satisfaction

FRT - First Reply Time

Requestor wait time 

Agent Turnover 

Agent satisfaction (Not EPS Employee Promoter Score).

o Option Bots vs Answer Bots

o Support Challenge --   Companies continue to look at support as a cog which doesn't make sense in a recurring revenue/empowered consumer market. 

o Focus on empowering both customers and agents vs ticket deflection different intent and will likely have different results. 

o Tiered support  will tend to create culture of competition. You need to watch out for that. Instead create culture of collaboration serve the person in front of you i.e., Level 3 coaches Level 2 and so on.

o Customer Experience vs Support Experience 

o Favorite Business Book The Service Culture Handbook by Jeff Toister.

Key Insights

o Hiring tips from James that I found to be really unique:

1. Good balance between senior/junior engineers (preferred ratio 2:1). Junior engineers usually turn out better than experienced folks.2

2. Dont just look for EE/CS majors. James has successfully hired even bio/physics/maths majors who are also interested in technology 

3. Focus on hiring folks from outside the area (such folks will invest heavily in building a culture of assimilation)

o One thing that got my attention is that their attrition is 3%. TSIA says attrition in support is usually around 15% . James shared that they have taken following steps (specifically in their India support team) to control attrition: 1/ Invest in growing support engineers technically 2/ Hiring strong managers 3/ Offer slightly better than market rate.

o Split Customer Success team based on Customer ACV and their ability/interest in being advocates (this seemed a unique segmentation choice). Top bracket customers enjoy 1:10 or 1:20 (CSM/Customer) ratio. For other brackets, ratio is higher. Earlier they started with 1:50 ratio for everyone and that didn't scale.

o Tool stack: Zendesk (CRM/Knowledge Management), Ring Central (Contact Center)

o James says that NPS is the most important customer metric you can measure

o Favorite business book for James: Delivering Happiness by Tony Hsieh

Key Insights:

o Nutanix NPS is ~90. For B2B Technology companies, average hovers around 30. Clearly, Nutanix is doing something unique.

o The team that provides technology tools for Support sits within Engineering rather than in IT/Support. I thought this was super clever -- that way head count of the team will come from Engg budget rather than IT budget (Engg budget > IT budget). Also, that way support tools team is sitting closely with engineering allowing them to better align with the product.

o They have built on top of existing support tools (e.g, Salesforce ServiceCloud) to provide unique experiences for their customers e.g., with more than 50% of their machines sending data back to them, they can generate "cross-customer" insights  to do proactive support.

o Parag's favorite book: Measure What Matters, John Doerr,