A run down of how Therefore Interactive, a web-based software strategy, design and technology firm in Toronto implements content strategy in it's process. Many of the touch points are related to Drupal, and generally CMS practices.
3. Therefore Interactive is a digital agency that specializes in
building superior enterprise communication systems through
open source web technologies.
4. What is Drupal?
Drupal is an open source CMS used by
governments, businesses and nonpro鍖ts all
over the world.
Different from Wordpress Drupal allows for
Site Builders to use the Drupal admin UI to
control almost all facets of site con鍖guration
without writing any code
Drupal allows us to apply content strategy
and information architecture methodologies
in a highly practical way.
5. Content Strategy at
Therefore
A new project generally begins with a speci鍖cation document outlining
in broad strokes what the client is looking for.
Lets go through some of the planning and strategy we as developers
and designers at Therefore use to go from speci鍖cation to development.
6. From speci鍖cations to
user behaviour
Where possible the speci鍖cation document is
outlined to de鍖ne the business needs and
behaviours that de鍖ne the goals of the project.
We do so using a standard BDD (Behaviour driven
development) Story/Scenario format
7. Story: Site visitor browsers events
In order to browse current events
As a site visitor
I want a calendar which allows me to current events
Scenario: A visitor lands on the event page
Given a site visitor is looking for current events
When he visits the event page
Then he see events which are in the current month
And events which are less than a week away
8. Roles and Personas
It is at this point that we may also begin to re鍖ne and
establish personas and the fundamental user hierarchy
of a site (membership or editorial).
Personas allow us to validate the various roles we are
creating in a system.
In Drupal understanding our roles early on will allow us
to understand what content management functions
should be exposed to each user type, and correlate our
user stories directly to agents in the Content
Management System.
9. Robert: The CEO with a startup plan
Persona
Robert is the CEO of a successful franchise based publishing platform called LocalLife. LocalLife
prints locally based monthly newsletters which are funded by local advertisement. The content is
focused around local events and classi鍖eds, and the franchises do the leg work of sourcing the
material. While Robert is highly successful, the web business is not something he has experience
with - but he's excited, and he's looking to take risks. Robert is 52 and mostly prefers to work from
his cottage in Muskoka when he can. In his downtime he enjoys golf with his old college pals and
boating with his family.
Goals
Robert wants to expand his business online. Robert sees an opportunity for premium hyper-local
advertisements on the web and is looking to 鍖nd ways to target users with ads based on
neighbourhood. He also wants to build a social network where users can generate and contribute
content themselves (cutting out the franchise model completely). Robert has advertisers already
looking to buy in, and so he's eager to move quick to capitalize on these opportunities. Robert is
also looking for ways that automation and web infrastructure can minimize overhead. Robert is also
excited to be able to oversee things more remotely and spend more time up at his cottage.
Scenarios
In order to chose a partner in building my big idea
As a CEO
I need a digital agency that respects my vision, money and time
10. User feedback and Industry
Research
Once we have created stories which fully outline
the initial spec, we may seek to validate our
assertions with user feedback.
Here we might also do some sectoral research to
determine how similar organizations model their
content and what information types in are common
to their sector (common content types, taxonomy,
etc)
13. Content Auditing
As a best practice we seek to import existing
content as early as possible.
Often existing content or data is opinionated, and is
coming from an existing database with its own
schema or structure.
Scrapers may be used to crawl an existing site for
auditing via a spreadsheet or for better technical
interface using an API (for this we might use a tool
like import.io).
16. Entity
Diagramming
Here we outline any new content
types we will need to create to
build the site.
We also begin to map out if there
are any relationships between
different content and how those
should be managed.
Once our content and data
structure is understood using
entity relationship diagramming we
can then begin outlining how we
will build these content types.
Often at this point a spreadsheet is
created with all content types, and
their related 鍖elds.
18. Merlin
At Therefore we have
created a Drupal contrib
module called Merlin which
allows us to to scaffold all
of our content types and
鍖elds by type simply by
importing a spreadsheet
into Drupal.
https://www.drupal.org/
sandbox/churel/
2160815
20. Modelling user 鍖ow
Referring back to our user Stories we begin to outline how users
will engage content.
Here we would establish our sitemap and clarify where the various
types of views, templates and content detail items will exist.
21. Where possible, it is at this
point that we may also refer
back to our roles and
personas to try and model
the overarching structure of
a site to best match the
expectations of real users.
In cases where budget
allows we may even
establish automated point
and click testing following
our Stories and Scenarios (to
minimize regressions and to
con鍖rm Stories are
complete).
22. Design and prototyping
Here we will outline and
begin and mocking up the
various UI components (lists,
displays, landing pages)
which are important to our
personas and roles.
We establish our core
navigation patterns and lean
on established best practices
to model our Information
Architecture for best usability
and accessibility.
24. Once we have established the
templates, views and
components which de鍖ne the
user experience of the website
(how users retrieve information/
content) we can begin the
process of design and
prototyping.
Therefore often uses a Mobile 鍖rst
approach allowing us to establish
simple linear hierarchies to how of
content laid out on a page.
we may also begin to prototype
certain components to determine
early through User Acceptance
testing if our assertions about
content hierarchy and navigation
patterns are viable or if they need
to be iterated on.
25. Development:
documentation continued
Once core development has begun, it is critical for us to
loop back over what we have done to ensure our content
strategy and information architecture documentation stays
in step with what we are building.
User Acceptance Testing allows us to pass features as
they are development back to the client (and where
possible the actual users of the site)
As changes are requested to the content/data model and
its front end implementation content strategy and
information architecture documentation is preferably
maintained.
26. Thanks!
For welcoming Alex and I as your new co-organizers.
Please reach me for questions or feedback at
sean@therefore.ca