Optimizing your content like an engineer, at STC Summit
This week I’m attending STC Summit 2019, the annual conference of the Society for Technical Communication (STC). I’m blogging my notes from the sessions that I attend. Thanks and all credit go to the speakers. Any mistakes are my own.
Sara Feldman‘s session was titled, Optimize Your Content Like an Engineer. She talked about incorporating engineering principles and quantitative measures into continuous content optimization, and how this would bring better cross-departmental collaboration and return on investment (ROI).
An engineer’s approach is to change your strategy and direction, rather than just change your approach.
Inverting your perspective to outside-in
Just because we may be better at putting the customer first than other team members such as some engineers, that doesn’t mean we have no room for improvement.
Things are changing all the time, especially in terms of technology. Our method needs to evolve accordingly. We should challenge our assumptions and the ways we do things, to improve customer experience.
Sara mentioned trends such as artificial intelligence (AI), and self-service as a starting point for support calls.
Strategies to address the above trends:
- Automate engagement with AI.
- Distribute consistent knowledge across your support channels.
Content is like water
This comparison helps us separate the message we want to convey, from the format of that content.
Content strategy is what we should be doing, and content engineering figures out how we should do it. A content engineer bridges the gap between content strategy and development.
Content engineering makes it possible for writers to deliver good value.
Measures and methods
Sara touched on the following measurement techniques:
- Leading and lagging indicators – Leading indicators are activity oriented. They represent the work you do, and are directly actionable. Lagging indicators are outcome-oriented, and they help you gain insight into the future. They’re harder to influence because you can’t change them directly (for example, you can’t directly change customer satisfaction (CSAT) measures). They confirm a pattern.
- Vanity metrics – Page views, time on page, bounce rate, number of active/unique/new users. These things are easy to measure, but they don’t tell you much about outcomes. Vanity metrics have an important place, to take the pulse, look at trends, get started. To get benefit from them, you need to correlate them to outcomes that you’re looking for. Reference: Sara recommended an article called Measuring Page Velocity with Google Analytics.
Here are some methods drawn from engineering principles:
- Eliminate unnecessary work – This technique is sometimes called maximising work NOT done.
- Refactoring – This means cleaning up your code. The goal is to simplify the design of existing code without changing its behaviour. The principles apply to content too: fix unhealthy dependencies and other sources of confusion and clutter.
- Data-driven design – Use data to determine what your customers need and want. The data should guide your decisions on a continuous basis. Use the data to define your minimum viable product (MVP). In other words, trim down your scope to get the content to your users sooner.
- Retrospectives – Debriefings to determine what went well and what didn’t, and use the results to improve your content and processes.
Future of content
Sara said this topic is kind of fun and scary at the same time. She mentioned Information 4.0, and the Information 4.0 Consortium. A group of very smart people who’ve got together to decide how information needs to behave in order to support Industry 4.0 (the trend towards automation and data exchange in manufacturing technologies).
Some of the recommendations for Information 4.0 are that content should be:
- Molecular – separate from format.
- Spontaneous and triggered by contexts.
- Ubiquitous – online, findable, and searchable. Gated content (content behind a signin) doesn’t qualify.
- Personalised based on customer profiles.
- Dynamic and continuously updated.
- Accessible to other systems, e.g. via APIs. This is also known as content as a service.
- Independent of channel, thus usable in multiple contexts.
So, content must be ready for anything? That’s hard to achieve. Here are some litmus tests from Sara:
- Chatbots (conversational UIs)
- Voice assistants (smart speakers)
To work with the above interfaces, content must be independent of format, and must come in small chunks.
A mature content experience
These are the requirements Sara mentioned for a mature content experience:
- Prioritise user experience and intent over authors. Intent means modifying the type of results you’re giving based on what you think the user needs in their current context.
- Blended content delivery. Our content is still too often presented based on our internal organisation structure. For example, help content, user manuals, training, all in separate places.
- Every snippet is snippet one. (This concept is based on the well-known concept of every page is page one.) Snippets must be able to stand alone. We need to be conscious of the language we use and how we reference other things that may not be available.
- Feedback loop to business objectives and products & services. Your customers are giving you lots of useful information as a result of your content analysis – make sure this feeds back to the business.
Thank you Sara for an interesting and engaging look at content optimisation.