Loading ...
Sorry, an error occurred while loading the content.

Government and Using Ad Networks

Expand Messages
  • <john_estill@...>
    I work for state government in the web development team as the analytics person. (Among other duties) I have been asked in the past year by internal clients to
    Message 1 of 6 , Apr 24, 2014
    • 0 Attachment

      I work for state government in the web development team as the analytics person. (Among other duties)


      I have been asked in the past year by internal clients to place ad network tracking bugs on our site(s). So far I(we) have been able to resist the inclusion of tracking bugs because I can show my clients how our current analytics solution can track general statics for conversions.


      Reasons for no ad network bugs or scripting:

      1) Security policy does not allow 3rd party scripting on our site(s) because we cannot control the scripts.

      2) We do not want 3rd parties tracking citizen behaviors as they move through our site(s).


      I am seeking comments.      #1 isn't going to change anytime soon. #2... Is this a valid concern? I could also see it as being business intelligence data thing that we wouldn't want to share.


      What do you think?

      Cheers! -John


    • Patrick Mazzotta
      First, calling them bugs is a horrible practice. Beacons or tracking pixels is not only more appropriate, it s more accurate (bugs specifically refer to
      Message 2 of 6 , Apr 25, 2014
      • 0 Attachment
        First, calling them "bugs" is a horrible practice. Beacons or tracking pixels is not only more appropriate, it's more accurate (bugs specifically refer to errors or broken code in our space).

        You don't need to include third party scripts, and you can retain 100% control of everything. Feel free to contact me directly if you'd like a hand designing a fully controlled solution.

        Patrick Mazzotta
        Co-Founder & CEO
        Craedone Inc.
      • <john_estill@...>
        Patrick, I know that I am new to this discussion forum. That does not mean I am new to the world or our space. I am seeking a constructive discussion about
        Message 3 of 6 , Apr 28, 2014
        • 0 Attachment
          Patrick,
          I know that I am new to this discussion forum. That does not mean I am new to the world or 'our space.'

          I am seeking a constructive discussion about allowing third party tracking technology (web bugs) to run across a government agency property.

          Please do not not chastise me for using technical terminology that has been in usage for the past 20 years. Nor do I appreciate the sales pitch after being corrected.

          Thank you,
          John


          ---In webanalytics@yahoogroups.com, <pmazzotta@...> wrote :

          First, calling them "bugs" is a horrible practice. Beacons or tracking pixels is not only more appropriate, it's more accurate (bugs specifically refer to errors or broken code in our space).

          You don't need to include third party scripts, and you can retain 100% control of everything. Feel free to contact me directly if you'd like a hand designing a fully controlled solution.

          Patrick Mazzotta
          Co-Founder & CEO
          Craedone Inc.
        • Wandering Dave Rhee
          Hi, John, Before I reply, let me state clearly that I am speaking only for myself, and not as moderator of this forum, or representative of any of the
          Message 4 of 6 , Apr 28, 2014
          • 0 Attachment
            Hi, John,

            Before I reply, let me state clearly that I am speaking only for myself, and not as moderator of this forum, or representative of any of the organizations I work with in any capacity.

            I am of the opinion that government and other public-interest sites have a particular responsibility with regard to data collection and privacy of their site visitors, independent of what the local laws may be at the time.

            This means that, although I will suggest ways to overcome the two concerns you mentioned, ultimately I am against collecting information that does not provide measurable value (not merely hypothetical or potential value) to those who are being tracked.

            Your first reason, lack of control over the scripts being executed, is a valid one. To counter that, some tracking bugs (I don't mind the term myself, even as a trained programmer!) have own-site equivalent scripts that can be substituted. That is, the code that the tracking vendor wants you to execute is given to you, and you host it yourself. If they change it, they need to give you the code, which you then re-validate before modifying on your server.

            Your second point, not allowing third parties to track your visitors, might be addressed by allowing them only to place code on specific pages, usually the landing page for a given campaign, and perhaps the final conversion page (depending on the goal of that particular campaign). This could allow for limited third-party tracking, without allowing them to see any intermediate pages or details viewed, provided your site is set up in such a way that makes this possible (e.g., consideration given to Flash-based sites, or other tech where technically, only a single "page" is ever loaded). That is, fire off tracking code only at limited times, not on every single page. However, this requires a bit more implementation work, especially maintenance to take stale code off the appropriate page(s) in the future, and cannot be automated as easily as putting code in the universal header or footer.

            Also, you already said your existing solution provides that functionality, so this wouldn't add any new value, but only consume implementation and reporting resources, which presumably are already limited.

            A third point that I feel more strongly about would be limiting the collection of information about visitors that does not provide any value. I am quite against the idea of collecting information "just because we might need it someday, in case we think of a valid use for it." Exploratory analysis of existing information is one thing, but collecting information (personally-identifiable or not) without a clear plan in advance of exactly what it will be used for should not be done. There is a slippery slope argument that this is part of what defines a surveillance-state, and even if it's a spurious argument, I still believe there is enough truly useful analysis to do that it's a waste of time to devote limited resources to value that is merely hypothetical.

            One way to present this argument is to use a version of information theory: if the information does not (or cannot) change your decision, and therefore does not (or cannot) change your outcome, then you cannot possibly create any new value. Explained another way, unless you can see how you would make your decisions in a way that would yield more profit, you will only spend money (or resources) without any return, thus lowering your profit.

            According to that theory, using the information as secondary validation for a decision you have already made is not valuable. Although this additional information would give you more confidence to confirm your decision, if it can never change your decision (for example, from a no to a yes), then it can never actually generate a difference, and thus never pay for itself.

            Besides the many costs of information gathering (validating and testing the code, implementing it, maintaining it, all the coordination meetings and emails to manage the process with all the stakeholders, generating reports, analyzing them, comparing them with other information sources, meeting about why there are discrepancies, deciding what to do about them, etc.), there are also the ethical burdens of properly storing the information (which you cannot simply assume the third party trackers will do), preventing it from being combined with other information (which could de-anonymize it), destroying it after an appropriate time (which other parties typically don't do on a periodic or auditable basis, besides which you wouldn't have the resources to conduct such an audit), and a whole host of other, similar issues.

            The other item you mention, potential competitive threat, is also valid. Many tracking companies may, as a matter of routine course, share your data in some form (aggregated, anonymized, etc.) with other parties, either as part of partner agreements, or as a paid service (reporting on industry trends). While you could theoretically rewrite the standard contracts to remove that, in practice, it's unlikely they would remove only your data from their standard collection set, as it would incur significant expense to alter their routines for only one customer. Perhaps, if they do this routinely, and you are not the exception, it might be feasible, but I would expect that to be highly unlikely.

            Forgive my long-windedness (let's call it a personal style, shall we? Sounds better to my ears. ;-) ) -- but obviously I have strong feelings about this topic, and I feel as though I do not do enough to help our industry behave in the most professional and responsible manner it can when it comes to setting data collection, protection, and retention policies.

            I hope you find the above helpful in formulating a clear and defensible policy, regardless of whether or not it's one that I would personally agree with. Also, welcome to the forum, and please let us know if you can think of anything that would help make it more valuable to all our members.

            WDave


            On Thu, Apr 24, 2014 at 5:09 PM, <john_estill@...> wrote:
             

            I work for state government in the web development team as the analytics person. (Among other duties)


            I have been asked in the past year by internal clients to place ad network tracking bugs on our site(s). So far I(we) have been able to resist the inclusion of tracking bugs because I can show my clients how our current analytics solution can track general statics for conversions.


            Reasons for no ad network bugs or scripting:

            1) Security policy does not allow 3rd party scripting on our site(s) because we cannot control the scripts.

            2) We do not want 3rd parties tracking citizen behaviors as they move through our site(s).


            I am seeking comments.      #1 isn't going to change anytime soon. #2... Is this a valid concern? I could also see it as being business intelligence data thing that we wouldn't want to share.


            What do you think?

            Cheers! -John



          • <john_estill@...>
            WDave - Thank you. You present many ideas and concepts that I have not really been able to verbalize to my coworkers. Long-winded or not... I enjoy reading
            Message 5 of 6 , Apr 29, 2014
            • 0 Attachment
              WDave -
              Thank you. You present many ideas and concepts that I have not really been able to verbalize to my coworkers. Long-winded or not... I enjoy reading well presented material. :)

              The part about information theory is interesting. I do not have experience or education in that field. I'll have to do some research.

              The point about only gathering enough data to make decisions and then not storing it for too long is valid. I keep about two years of detailed data, then roll it off. I then store high level 'executive' data so that I can produce those dreaded squirrel reports, "How much has our traffic increased since last year/5 years?"

              Again, thank you for the great reply and warm welcome. I should have joined a couple of years ago when I was originally given the assignment.
              -John


              ---In webanalytics@yahoogroups.com, <wdaveonline@...> wrote :

              Hi, John,

              Before I reply, let me state clearly that I am speaking only for myself, and not as moderator of this forum, or representative of any of the organizations I work with in any capacity.

              ...

              <The rest of the discussion has been truncated for brevity in the reply>

            • Wandering Dave Rhee
              Hi, John, Glad you found something useful in the post! And we re happy to have you (and everyone else) here and participating! :-) WDave
              Message 6 of 6 , Apr 29, 2014
              • 0 Attachment
                Hi, John,

                Glad you found something useful in the post!

                And we're happy to have you (and everyone else) here and participating! :-)

                WDave


                On Tue, Apr 29, 2014 at 2:35 PM, <john_estill@...> wrote:
                 

                WDave -

                Thank you. You present many ideas and concepts that I have not really been able to verbalize to my coworkers. Long-winded or not... I enjoy reading well presented material. :)

                The part about information theory is interesting. I do not have experience or education in that field. I'll have to do some research.

                The point about only gathering enough data to make decisions and then not storing it for too long is valid. I keep about two years of detailed data, then roll it off. I then store high level 'executive' data so that I can produce those dreaded squirrel reports, "How much has our traffic increased since last year/5 years?"

                Again, thank you for the great reply and warm welcome. I should have joined a couple of years ago when I was originally given the assignment.
                -John


                ---In webanalytics@yahoogroups.com, <wdaveonline@...> wrote :

                Hi, John,

                Before I reply, let me state clearly that I am speaking only for myself, and not as moderator of this forum, or representative of any of the organizations I work with in any capacity.

                ...

                <The rest of the discussion has been truncated for brevity in the reply>


              Your message has been successfully submitted and would be delivered to recipients shortly.