Set Top Box Lexicon
Request for Information
Set Top Box Measurement
Request for Information
Cross Platform Measurement

Request for Information - Set Top Box Measurement

download .doc

  • Instructions: Please fill out the following Request for Information and return a completed electronic copy of the RFI, in Word format, to the Coalition for Innovative Media Measurement by December 4th, 2009. Please send the document to CIMM c/o Alan Wurtzel, President, Research and Media Development, NBC Universal, at alan.wurtzel@nbcuni.com.

    Responses are due no later than 5pm Eastern Time on December 4, 2009.

    1. Obtaining set-top box usage data:
    A. Describe in detail the nature of the STB data to which you currently have access.
    i. List all of your current sources of STB data, including information on who owns that data.
    ii. Describe the size and commercial nature of the data (i.e. whether you can sell the data to interested subscribers)
    iii. State whether the STB data has actual return path data and from what portion of the sample/census?
    iv. Explain whether the STB data is reported for all boxes in a household, or on a one box per household basis
    B. Do you face any challenges in regards to obtaining STB data on a national basis? If so, please provide detail. Do you have any plans to overcome those hurdles? If so, please explain.
  • 2. Analyzing data across multiple DMAs
    A. Do you collect STB data across multiple DMAs? If so, how many DMAs do you currently collect data from?
    B. How do you account for different market demographics when analyzing data across multiple DMAs? If you are not currently doing this, how do you anticipate handling this challenge in the future?
    C. How do you reconcile differences in STB technologies when analyzing data across multiple DMAs? If you are not currently doing this, how do you anticipate handling this challenge in the future?
    3. Do you combine data from multiple data sources (e.g., multiple operators)? If so, please describe how you bring the data sources together.
    4. Projecting:
    A. Today it is widely assumed that STB samples are not projectable. Do you agree with this assumption? If so, why? If not, why not?
    B. Assuming STB data is not yet projectable, please describe your plan for reaching national and local projectability?
    C. What are known skews of your current sample(s)?
    D. Describe any weighting methodology that you use.
    E. Describe the benefits and weaknesses of your weighting approach.
  • 5. Are your daily viewing logs drawn from the same homes each day? Are those viewing logs drawn from a sample of the viewing universe you have access to or a census? If a sample is used, please explain the sampling methodologies.
    6. How do you deal with non-STB sets in a home?
    7. How do you deal with non-STB homes in your sample area?
    8. Do you have the capability to report on the viewing habits of individuals within the household? If so, please describe
    9. Measuring VOD, VCR, DVD and DVR viewing data:
    A. How do you measure VOD viewing data? Please provide detail on both a national and local market basis.
    B. How do you measure VCR viewing data? Please provide detail on both a national and local market basis.
    C. How do you measure DVD viewing data? Please provide detail on both a national and local market basis.
    D. How do you measure DVR viewing data? Please provide detail on both a national and local market basis. Are you able to report on live and timeshifted viewing? If so, what timeshifted timeframes do you report on?
  • E. How do you differentiate and account for viewing on standalone DVRs as opposed to DVRs built in to a STB?
    F. How do you differentiate and account for in-home single-unit DVRs as opposed to home-network DVRs?
    G. How do you differentiate and account for remote-server DVRs (i.e., Cablevision's proposed implementation)?
    10. How do you report HD networks as opposed to SD networks? If you don't do this today, please note this, and explain how you plan to report this in the future.
    11. How do you process syndicated programming? (E.g. Programs that don't run in a uniform time slot) If you don't do this today, please note this, and explain how you plan to process syndicated programming in the future.
    12. Out-of-Home Viewing:
    A. How do you measure out-of-home viewing?
    B. How do you distinguish between audiences in different viewing location (e.g., home, office, bars, hotels, airports)?
    C. If you don't report on Out-of-Home viewing today, please note this, and explain how you plan to report this in the future.
  • 13. Ingesting network logs:
    A. How do you ingest network logs?
    B. How do you identify content that is not coded, or for which logs are not provided?
    14. How do you determine what program or commercial is being played or watched at any given moment: Please address for:
    A. Live viewing
    B. Delayed viewing
    C. VOD viewing
    15. Are you reporting on all commercial and program activity that appears on air? If not, how do you plan to do so in the future?
    16. How do you analyze and report on commercial vs. program data? Please include detail about pod, pod position, etc.
    17. Can you currently provide second-by-second data? If so, is this on a production basis or a special-analysis basis?
  • 18. Please describe the edit rules you use for STB data analysis, and how such edit rules should be implemented. Please provide details on your rules, including the basis for them, as pertaining to:
    A. Box on / set off
    B. Faulting
    C. Dwell time
    D. Signal latency
    E. Any other key rules
    19. Which, if any, of your edit rules are dynamically derived (i.e., vary based on the variations in viewing behavior)?
    20. Each STB data company usually develops internal processing guidelines and edit rules surrounding issues such as faulting, dwell time and signal latency. Would it be beneficial to establish common definitions and develop an industry lexicon for these sorts of issues? If so, which elements would benefit most from being defined by the STB data industry? Additionally, how do you think these guidelines ought to be developed?
    21. Integrating and overlaying advertising schedules:
    A. Do you enable users to overlay advertising schedules on your data? If so, how? If not, why not? Do you anticipate offering this option in the future?
  • 22. Categorizing viewers by media behavior:
    A. What is your methodology for categorizing ad-skipping viewers by media behavior?
    B. What is your methodology for categorizing non ad-skipping viewers?
    23. Do you currently overlay data elements such as purchase data, lifestyle data, psychographics and/or engagement measures to current commonly-used media buying and selling metrics? If so, please describe which of these elements you overlay and what is involved in this process.
    24. Do you employ a single-source sample, or fusion of different samples?
    Please describe the benefits of your approach, and, if the latter, how this fusion is or will be implemented?
    25. What other "partners" or suppliers would be appropriate to include in your STB measurement solution (e.g., demographic data suppliers, sales data suppliers)?
    A. How do you plan to collaborate with these partners?
    B. What issues do you anticipate in making this collaboration a success?
  • 26. Collecting additional data from STB homes:
    Do you draw, or plan to draw, any other data from homes that provide you with STB data? Please address for the following data types. If you do not currently collect this data, but plan to in the future, please describe how you plan to implement this capability.
    A. Household demographics
    B. Shopping card data
    C. Behavior/attitude-shift data
    D. Any other sales or behavior-based data that links to non-packaged goods categories
    27. How do you demonstrate media ROI? If you are not currently doing this, do you have plans to demonstrate media ROI in the future? If so, how?
    28. Targeting:
    A. Can your data be used for real-time ad targeting? If so, how?
    B. What level of targeting granularity may be possible in the future (e.g., zipcode, household)? How could you implement this level of targeting?
    29. What is your estimate of subscriber cost increases as the sample expands? How did you arrive at this figure?
    30. How do you anonymize your data and protect consumer privacy?
  • 31. List the metrics you currently report.
    32. Do you capture interactive (non-tuning) data from boxes that support it (e.g. ad click-throughs, eRFI, EPGs)? If so, please describe.
    33. What methodological research have you conducted to validate your current data, sample methodology and/or results?
    34. What methodological issues do you acknowledge with your data or analysis? How do you plan to overcome these issues?
    35. What is the time lag between telecast and availability of program data? And what is the lag for commercial-level reporting?
    36. Where does your data get processed?
    37. Please describe your current data delivery system and capabilities.
    What are your plans for upgrades?
    38. Can you interface with established third party processors such as Donovan and MediaBank? Please list all and describe how you collect / share data with these processors.
    39. What are your plans, if any, regarding engagement with the Media Ratings Council?
    40. Is there anything else you think we should know about your company?

  • IV. Additional Information

    More details on the process for submitting proposals will be forthcoming shortly.

    In the interim, please submit any questions or comments to:

    Alan Wurtzel
    President, Research and Media Development
    NBC Universal
    alan.wurtzel@nbcuni.com
    212-664-7565

CIMM's Privacy Policy
For more information, please contact Marzena.DelGaudio@cimm-us.org