Customer Success - CMS Design Training

Screen Shot 2020-03-02 at 12.34.23 PM.png
Looker Report pre vs post training

TLDR: Updating support reps design training to be tailored to HubSpot’s CMS improves customers experience when filing design tickets.

PROJECT BACKGROUND: Within HubSpot’s Customer Support teams, representatives can specialize in one area of the product. As a specialist, you answer other support representatives questions about that part of the product in a slack room everyday and in twice weekly office hours. I had taken a software and coding certificate program before starting at HubSpot, so I had enough knowledge of front end coding languages to become a Design Specialist. When customers have questions regarding their website, template or email designs, these are counted as design tickets. Design tickets often involve HTML/CSS, JS and a deep knowledge of how to use the HubSpot CMS. Since any knowledge of coding is not required in order for a Support Rep to be hired by HubSpot, there is a large knowledge gap for these questions. 

RESEARCH GOALS: In order to alleviate some of the pain points of design tickets, I wanted to find exactly what these pain points were. During office hours, I had taken notes on what types of questions Support Reps were asking. To help confirm my notes, I also sent out this survey to Support Reps asking general/broad questions they have about design tickets:

https://docs.google.com/spreadsheets/d/1cM83i7_al9udlRMaSSmrI9dwslwkoxvc2ktyk2uuKdk/edit?usp=sharing 



HYPOTHESES: Once I had collected these questions and based on my notes from office hours trends, I was able to extract these themes:

  • Questions people brought to office hours meant they used the various HubSpot resources and could not find an answer, so these questions were often high pain points and led to customer frustration. 

  • These questions usually took many days to resolve, since Reps had to wait for office hours, and if the question was not answered at Office hours, then had to be Jira-ed, taking more time

  • Many Support Reps did not feel comfortable making any changes, since the changes would affect live content

  • Support Reps were unsure if external sources would resolve issues for inside HubSpot

  • Could not identify between bugs in the software or coding mistakes resulting in unwanted visuals

  • Support Reps would find possible answers, but did not have confidence to employ answers since they had no knowledge of repercussions 

  • Unsure if changes should be done on page level or template level


METHODOLOGIES: After collecting and reading through the survey answers, I met with the Learning and Development team as well as HubSpot’s UI UX team for various resources on how to gather qualitative results from these answers. Through brainstorming and reading, I was able to gather that these questions stem from not understanding any fundamentals of coding, in a combination of lack of confidence to employ changes. I gathered these two core issues, from analyzing the question/notes and noticed a pattern of unsure if a requested change is possible code-wise (not understanding fundamentals), not knowing where/which coding language a change would take place (not understanding fundamentals) while observing Support Reps being scared to employ any changes or troubleshooting themselves since the changes would be reflected on live sites/templates (confidence issue).


To address these two main core issues, I created a design training. I researched how to teach for confidence issues, which I found live troubleshooting with encouragement/direction can help this - to show the user they can resolve cases, and give encouraging steps. For fundamental questions, I broke down basic differences between HTML vs CSS, definitions and where changes should be done (template vs page editor).

I created google slides for this lesson and gave the training for an hour each, for two Support teams (around 25 people each). In the slides, I also touched on the commonly asked questions I see during Office Hours, how to troubleshoot for these, and numerous helpful resources they should bookmark for future cases. 

Links to slides: https://docs.google.com/presentation/d/1Lg_Bt0lkvCDptwPKRJzJR1qdqg4XhD2XDvzJrkq6eUY/edit?usp=sharing 

RESULTS: Post trainings, I asked the participants how helpful they thought the trainings were, any post training questions they had and if they thought their design training question resolution times have decreased. Some participants answered, all with generally positive results. 

This is the link to the post training survey: https://docs.google.com/spreadsheets/d/1qc5vgBbxPanwKu1KZBrpXOYDdB32aYV-xdST7gWaA9U/edit?usp=sharing 

In order to prove if the resolution time for design cases worked, I used Looker. I collaborated with a colleague (Noam Kahn) to utilize Looker to look at design labeled tickets NPS and days to resolution pretraining vs post training, for the two teams I gave the training to. 

The looker report can be viewed here: https://looker.hubspotcentral.net/looks/44493?toggle=dat,vis&qid=LKmylZOvPvRNyBH7Mm0SNA 

For Michael Boyles team, design cases NPS post training increased by 6 points and days to resolve decreased from 3.1 to 2.9 days. This means, customers had a better experience when filing a design ticket because their question was answered quicker! For Lindsay Hannons team, design ticket NPS increased by 12 points and days to resolve increased from 2.0 to 2.9 days. I believe these results mean, that while resolution time may have increased, the correct answers are being found for the customers, because NPS scores have increased significantly. 

For the teams that I have not given the training to yet, I created an infographic of helpful steps and resources Support Reps can take and posted it in the #Support-Design room. 

Link to PDF: https://drive.google.com/file/d/1_xyc2Oz2rgqhE9yyn6nKNI4YYlsEgK1L/view?usp=sharing 

STAKEHOLDERS: Throughout the project, I identified the stakeholders to be: Customers, Support Reps, PEs and Support Managers. These groups are stakeholders because they’re looking for quicker resolution times, ability to find answers, improved customer satisfaction rates and more information/troubleshooting done on Jiras.

So, in conclusion I researched the pain points of Support Reps and customers, created a training tailored to help alleviate these pain points and analyzed the results in order to #SolveForTheCustomer 


Resources: 

https://www.interaction-design.org/literature/article/ux-report-writing

https://www.interaction-design.org/literature/article/map-the-stakeholders

https://wiki.hubspotcentral.net/display/UXTeam/Email+Enablement+Customer+Feedback+-+Email+Health+January+2020+-+UX+Research+Results 

https://wiki.hubspotcentral.net/display/UXTeam/Teamwork+Makes+the+Marketer%27s+Dream+Work%3A+A+Deep+Dive+Into+Collaboration

https://uxdesign.cc/triangulation-perspective-how-to-design-with-confidence-5044c8e21189

http://cdn2.hubspot.net/hubfs/439788/Content_Offers/UX_Design/User_Experience_Audit_-_Sample.pdf?t=1484066350369

https://usabilitygeek.com/write-ux-proposal-how-to-guide/

Kelsey Segaloff (HubSpot UX Researcher) gave me these example UX Research reports for me to take a look at in order to create this report:

UX research recent results.

On how marketers work together

On ads users and targeting