Dialogue-A-Thon Focus

Hi all and thanks for joining the search dialogue-a-thon and the 2 week conversation! During this focused hour, I'll be consolidating some ideas that have been posted and suggesting some ways we can continue evolving those ideas into implementation plans. Please post new ideas, comment on existing ones, and let us know how you can help further!


Submitted by

Stage: Active

Feedback Score

0 votes
Voting Disabled

Idea Details

Vote Activity

  1. Upvoted
  2. Upvoted
  3. Downvoted
  4. Downvoted

Similar Ideas [ 1 ]


  1. The idea was posted


  1. Comment
    Vanessa Fox ( Idea Submitter )

    SEO best practices/guidelines - this has come up several times and I think is a key foundational element to .gov site managers being able to quickly make improvements and maintain them long term.

    The GSA hosts SEO best practices here:

    But it could use updating and improvement. Anyone want to volunteer to help with this? I'll gather all of the feedback into a cohesive set of recommendations and work with the GSA roll them out. In particular, all of the ideas we're about doing these two weeks (making infrastructure more crawlable, consolidating content based on user focus, merging sites, etc.) can use step-by-step checklists.

    SEO is really all about good site architecture, usability, and engagement so SEO best practices don't live in a silo; they're part of a larger integrated set of best practices, which is why below, I'm including all of the best practice threads.

    Some of the existing threads that talk about this topic are:

    And the opposite idea:

  2. Comment
    Vanessa Fox ( Idea Submitter )

    Content consolidation and user-focused content strategy: this is one of the most popular topics of this dialogue so far, and is so so important for usability, SEO, and overall site effectiveness.

    Below are some of the discussions going on regarding this topic:

  3. Comment
    Jeff Jones

    On the technical front, Google has started factoring in user experience and social media into its ranking algorithms, so keeping up with the latest search trends will require actually creating a welcoming user friendly website rather than obsessing over keyword density and link building. This is a Good Thing.

    I will still argue that content creators should not even be allowed to use the phrase "SEO", and focus on creating meaningful, helpful Plain Language content in Web Style, and not simply copy/paste from their press releases and divisional reports.

    Finally, understanding and applying the principles of the Semantic Web by applying meaningful metadata will allow both search engines and humans with alternate accessibility limitations or preferences to better access and understand content, as well as facilitate automated multi-channel publication of that content to current and future platforms.





  4. Comment
    Vanessa Fox ( Idea Submitter )

    Another topic has been around technology specifics. I linked to some of this in the first "seo best practices" post above, but there are some additional interesting points:

    QA testing (may need a checklist for this!):


    Move away from PDFs:

    HTML5 and semantic markup:

  5. Comment
    Vanessa Fox ( Idea Submitter )

    Jeff, I totally understand what you're saying about the term SEO. Unfortunately, it's come to have two fairly distinct meanings: one is around spam and manipulating search engines and the other is around making sure that you're solving people's problems (by using search data to find out what they're looking for) and building a site that search engine can crawl and index and that shows up for the right searches.

    We obviously are talking about the latter here. I've had this struggle with terminology for a long time. In my book, I talk about how I don't know if the term "SEO" can be reclaimed and we may need a new term to refer to connecting with our audiences through search.

  6. Comment
    Vanessa Fox ( Idea Submitter )
  7. Comment
    Jeff Jones

    I really am talking about "white hat" solutions-based SEO. Web managers and developers need to be keenly aware of these issues, but I've seen a push for *everyone* to worry about SEO, and I don't think it needs to leave the web-team's office.

    Using user-focused terms -- which is part of the Plain Language initiative --- and applying the appropriate metadata will achieve all of those good SEO goals, without trying to adhere to misremembered or outdated search engine tricks.

    Speaking in terms of findability sounds new-age, but it might get everyone focused on the problem, and leave the technical implementations to the web managers and developers.

    I'm not dismissing the need for good SEO, I'm just trying to stop people who don't know much about it from telling me how it's done. I have this problem with accessibility too, but that's another topic. :)

  8. Comment
    Lakshmi Grama

    While I agree that user-focused content strategy goes a great way in helping content be more findable, I also believe that utlizing a user-focused taxonomy to categorize content and integrating that with a search engine will help people navigate search results better while they are using site search. I don't see a lot of government sites using this approach - my site does not do that yet, even though we have a faceted search engine.

  9. Comment
    Vanessa Fox ( Idea Submitter )
  10. Comment
    Vanessa Fox ( Idea Submitter )

    Lakshmi - I think we posted at the same time! :) Yes, I think sites should work for those who prefer search as well as those who prefer a browse behavior. The latter helps search engines understand the site and better surface the content in search results as well so you get that extra benefit.

    Jeff, I think SEO best practices need to built into every part of the org, but there's no reason they need to even think of them as "SEO", because they are actually best practices beyond the silo of SEO, such as user-focused content strategy, accessibility, usability, etc.

  11. Comment
    Jeff Jones

    Vanessa: what you said. :)

  12. Comment
    Jason Birch

    Coming a bit late to this site, but something I think can't be said clearly enough is that government records need to be broken out of the prison of topic-specific search portals.

    Yes, it's still important to have advanced search, but the first priority should be to expose records as logically indexed (organized by topic, etc) linked and crawlable HTML pages.

    This is just common sense SEO, but I believe that by and large, government hasn't yet made the leap that private industry did years ago in optimizing sales catalogs for one-click access. As a citizen, I want to be able to type a trademark, patent, license or other query into Google and get some initial results (which I could then follow up or sideways to find other information if necessary). I don't want to have to search for the search portal where I have to enter my search.

    Vanessa, I know this is something you're passionate about, and the work of my group on our municipal web site was heavily informed by some of your earlier writing.

    By gradually moving our new application development and redevelopment to follow SEO best practices, we have been able to significantly increase access to our information. Now, when typing an address or business name into Google, you will often get a result from our property reports or business license database on the first page.

    We were also able to leverage this work when we recently transitioned to Google Site Search. Now, citizens can find what they are looking for across applications, and even drill down using refinements based on URL patterns.

    There is still a long way to go (we've got lots of data and limited time) but this strategy is in the top two most important things we're doing to improve user experience.

    In my opinion, this record level search optimization would be highest bang for the buck for the majority of existing government web apps.


    And yes, we're using sitemaps too in most cases, but they're not really necessary if the data is linked properly. For instance, one of our early projects is pretty well indexed without them: