Dear All,

Please see below the action items and notes captured by staff from the Data Sub Team call held on 18 July 2018 (17:00-18:30 UTC).  Staff have posted to the wiki space the action items and notes.  Please note that these will be high-level notes and are not meant as a substitute for the recording. The recording, AC chat, and attendance records are posted on the wiki at: https://community.icann.org/x/ioxHBQ.

Best Regards,

Ariel

Ariel Xinyue Liang

GNSO Policy Support Specialist

Internet Corporation for Assigned Names and Numbers (ICANN)

==

ACTION ITEMS

• Sub Team to review the trademark owners survey and suggest which questions remove/truncate to reduce the length; Sub Team to consider whether splitting the TM survey into shorter surveys based on sections (e.g., Sunrise, Claims, Reserved Names) would be more effective in reducing lengths.
• AG to consider the suggestion allowing respondents preview all the questions to set the right expectation (e.g., via a PDF).  
• AG to provide input on TM survey questions that respondents could not really answer.
• AG to consider breaking the TM survey into three short surveys (i.e., Sunrise, Claims, Reserved Names); respondents get links to the three surveys and they can take as many as they'd like; or respondents get randomly selected to take one of the three short surveys based on certain qualification/selection criteria.
• AG to consider combining/consolidating Q4g, Q5, and Q6 in the Registrar Survey.
• AG to consider combining/consolidating the questions with related tables to reduce the Registry Survey length to 20 questions
• AG to clarify in Registry Survey Q2 that we are seeking responses from the actual registry operators, not the backend registry operators.
• AG to direct respondents from Q7 to Q10 and Q11, and then ask Q8 and Q9 as add-on questions in Registry Survey.
• Staff to schedule a Sub Team meeting on Friday, 20 July at 16:00-17:30 UTC (DONE).

NOTES

General Comments Regarding Survey Length

• Set respondents' expectation and note that they need approximately 20 minutes to complete each survey (instead of 15-20 min).  
• Create means (e.g., worksheet) for respondents to return to the survey and answer the questions that require research and data gathering.
• If specific contact information of individuals is provided, they can return the survey with a password and complete.
• Registry survey has 40 questions, registrar 36 questions, and TM survey has 56 questions. The current length of these surveys would cause a lot of drop off.
• ACTION ITEM: AG to consider the suggestion to allowing respondents preview all the questions (e.g., via a PDF)
• Break the long survey into smaller surveys? Prioritize and reorder the questions so the important questions go first?
• If there are no follow-up questions to solicit more elaborate responses, it would be challenging to get WG's approval of the survey results.
• If the WG’s interest to get certain data has negative impact on the Sub Team’s ability to obtain data in general, the Sub Team have to challenge the WG. Some data that the WG wants to obtain is not obtainable. TM owners are not keeping record the way that ICANN asks for, for example.
• Expedia’s user experience surveys create two paths: 1) just do the rating (stars), 2) provide more extended responses. Consider adopting this approach to let respondents choose whether to complete a short or full survey.
• Registries/Registrars have incentive to answer the surveys.
• At the beginning TM survey, perhaps spell out the type of legal counsel should take the survey and the type of data/fact that respondents should have it handy to facilitate their process answering the survey.
• The lack of clear indication of the time needed to complete the survey would be off putting.
• The questions in the TM survey are more general and doable type of questions, and less difficult and intimidating than the CCT-RT survey. We should let survey takers understand the specific information we are seeking instead of asking them to analyze it.
• Ideas for lowering the drop-off rate for the TM survey:
• Present a short survey option and a full survey option (e.g., click here to answer Qs on Claims; here to answer on Sunrise?)
• After presenting the short survey, ask respondents whether they want to take a subsequent full survey,
• Inform respondents that they would need 40 min to complete the survey, but they will be provided with a password to save their responses and return to the survey to complete; AG could look into setting this as a default option – two-step process: Want to take the survey? Email us, and AG will provide an individual password,
• Calculate the time needed to complete each section of the survey.
• ACTION ITEM: AG to provide input on questions that respondents could not really answer.
• The short/full survey approach sounds reasonable, but AG would leave to Data Sub Team to determine whether this approach would get the information that is ultimately useful.
• Would the short survey approach skew the results? Not sure whether skewed responses matter as the goal is to get the data. It is better to get the data we can get rather than "all or nothing." 
• If we split the long survey into short surveys, prefer randomization instead of inviting respondents to pick and choose which survey(s) they wish to take (may need to qualify respondents differently). This would generate less bias. If there is no randomization, you could get responses from people who have strong opinions about certain issue, hence more tendency to bias.
• ACTION ITEM: AG to consider breaking the TM survey into three short surveys (i.e., Sunrise, Claims, Reserved Names); respondents get links to the three surveys and they can take as many as they'd like; or respondents get randomly selected to take one of the three short surveys based on certain qualification/selection criteria.

**

Registrar Survey

Q1

• Make the question more direct: Which registrar are you employed by?

Q2

• Ask where the headquarters of the registrar is.

Q4g, Q5, Q6

• ACTION ITEM: AG to consider combining/consolidating Q4g, Q5, and Q6.

Q19

• Reword the question to "At what point in the registration process do you typically display the Claims Notice..."
• Options for drop down menu may include the relevant steps of a domain name registration process:

• WHOIS Lookup
• When the domain name is selected
• When the domain name is added to the cart  
• When the consumer log into the registrar account or create an account
• When the consumer is present with the terms & clickthrough agreements
• When the consumer is present with the renewal information
• When the consumer proceeds to check out

• Most people may get the Claims Notice just before the check-out step.

**

Registry Survey

General Comments

• ACTION ITEM: AG to consider combining/consolidating the questions with related tables to reduce the survey length to 20 questions

Q2

• ACTION ITEM: AG to clarify that we are seeking responses from the actual registry operators, not the backend registry operators.

Q7-12

• ACTION ITEM: AG to direct respondents from Q7 to Q10 and Q11, and then ask Q8 and Q9 as add-on questions

Q9

• Respondents who already offer such TLDs would be able to understand the vagueness of the wording. Leave it as is.

Q10a

• "Publishing reserved names would violate local laws" would apply to a small number of jurisdictions.
• Drop-down menu options:

• Such publication would violate local laws
• Trade secret
• Reluctance to post competitive data
• OPEN field

Q16

• This is a concluding question, put at the end.

Q17

• Old wording is confusing too, e.g., 'special rules', 'certain groups'.
• Do we want to offer special group access based on their restricted TLDs and eligibility criteria? There may be special rules for government affiliated groups to register restricted TLDs before brand owners (e.g., police.city). Q17 should be worded broad enough but at the same to put some parameters around it.
• Take out 'brand owners' in the brackets  
• Remove “Sunrise"; just say "…first access to their target audience..."

Q21

• Did you face any challenges with regard to both of your LRP/APLP and mandatory sunrise period?

**

Actual / Potential Registrant Survey

Q5a

• Concerned about the "more than once". Respondents may not remember.
• Ask a follow up question -- For subsequent time(s) that you get a claims notice, did you do anything differently? Open Ended follow-up question could provide the opportunity for respondents to elaborate without making it onerous.
• We really want to hear back from registrants that received Claims Notice more than once.