Closing Panel Discussion: The Future of Content

 Panelists:

• Scott Abel
• Larry Swanson
• Rahel Anne Bilie 
• Dipo Ajose-Coker

What will be the next big thing in content production and delivery?

Dipo: Everyone excited about generative aspects of ChatGPT. Just starting to understand the generative isn't the only interesting part. 

Rahel: Intelligent content is content that is semantically categorized. Search engines will better be able to understand it. AI still favors structured content. 

Larry: We are the real intelligence behind AI. Guardedly optimistic. 

Scott: Precision matters. Many sad-ass excuses why companies are not precise with their content. Predict security issues. Will be a precision revolution. 

How will AI affect the ability of people to join our field?

Rahel: Every generation defines technology differently. The next generation of tech writers is going to be different, and they will "grow up" with that and will be more adaptable. 

Dipo: AI is just another tool. It'll be doing simple things. 

When you're hiring, what is your favorite interview question to see if they are a good fit?

Jack: I believe you can tell more about a candidate from the questions they ask. 

Dipo: Standard questions, depends on how they answer. 

Scott: Trying to find a good fit. Was asked once "What do you now want to do?"

When do you feel companies will start to look at ethical AI?

Rahel: We all know about accessibility. Going to be the same with AI. Now there is EU AI act. 

Larry: As a practitioner working with AI tools, be aware of what's happening upstream of where you are. 

What can we do across content ops to set us up for future success with AI?

Rahel: It's all interconnected. It takes a village to solve that kind of problem AI is just another tool in our toolkit. 

Scott: Need to be cheerleaders for the idea that everything needs to be standardized. Precision is what we need to work toward. If you can;'t tell me precisely what content you have, that means you don't have it. 

Larry: We can look to the data world for an example. There's this notion of FAIR (Findability, Accessibility, Interoperability, and Reuse) data. And there's the notion of trust, and AI does not have that trust. 

What are some other cautions that people should be aware of as they int3egrate AI into their business practices?

Scott: Need to figure out where is the value? 

Dipo: We know it consumes a lot of energy. We have also been creating and duplicating content, so why not use AI to reduce that? 

Rahel: Data sustainability is a big issue. 75% of data collected is never used, but energy for storage and search is huge.

Wednesday Closing Keynote

 Estimating for More Fun and Greater Profit

John Hedtke, Principal Consultant at Double Tall Consulting, said that knowing how to estimate is one of the most important skills we have as technical communicators. 

A good estimate contains a description of the project's deliverables, the time it will take to create the deliverables, the cost, and the confidence rate. 

A documentation plan states the goals, specifics the details, and describes the process. Most important to have a good outline. 

"Metric" is just a fancy way to measure something, and can be simple or complex. Nothing we do that we cannot measure in some way.

An estimate is a guess. A metric is a measure that is (hopefully) accurate. An estimate that uses metrics for its data may be a WAG or even a SWAG but it's still a guess. 

Personal metrics measure how long it takes you to do something. 

Rarely do we work on anything more than 15-020 minutes at a stretch. The act of logging keeps you honest. Logging also keeps you more honest. 

As you gather data, you can find out when you're most productive, when you're interrupted the most, times better for meetings/phones calls/administrivia, and how productive you are overall. 

If you're a contractor, 80-85% of your time is productive. Captives tend to work at 60-70%. Adding a day log tracks what you did and how you feel about it. 10-15% to your productivity can make a big difference. 

Day log tracks what you have done and how you feel about it. It is subjective. Identifies rhythms over the course of a project and helps predict cycles for future projects. 

Keep your day log private. Be complete, honest, clear. It's like a diary. 

Doc plan, time log, and day log gives you the infor to build your estimating spreadsheet.

Getting it WRONG: Lessons Learned from Building a Web Content Audit Tool

 Paulo Fernandes, Co-founder of Luscious Orange, said that they decided to build a content audit tool. Found that most tools focused on tech and metrics (automate it and add AI) and removed humans from the equation. Content audits exists in mashups of spreadsheets, tracking tools, and notes. 

Had to build a crawler first. Crawling is easy. Interpretation is hard. Computers don't have context to ignore what is not necessary to know about content. 

Most people don't know about sitemal.xml. CMSs generate it. Is a list of all files. A page on a website is not always the same as a page in a sitemap. A page does not need to be in your navigation to be visible to the whole world. 

A link and a file are not the same thing. Inventory is just a bunch of URLs. HTML has tags that can point to other resources. A file can exist on your server that is never linked. A link can exist on your site that points to nothing. 

Asset inventory is different from content inventory. 

Automation is great, but content audits are human work. AI can;t help with items that require deep context.

Audits generate tasks. Leverage all the great stuff that we already know about task management.

From Chaos to Clarity: How to Surface Strategic Insights from Content Inventories and Audits

 Vanessa Stuivenvolt Allen, Director of Content Strategy at Resolute Digital explained the different between a content inventory and an audit. Inventory is a living, breathing log of content on your site, and that includes metadatra. Key to a content inventory ias that it should be kept up-to-date. Amount of effort required depends on business needs. 

Basic content inventory can be just a spreadsheet. Should be collaborative. Can also use a database solutions. Can be helpful if there is more data you want to track. Can be your (C)CMS. 

Content inventories save time, are the basis for future audits, and has info to pull into reporting. 

If you don't know what's there, you're not managing it. 

Content audit is a qualitative/quantitative evaluation against a set of defined criteria. Allows measure content against business goals. It is a point-in-time artifact and ins not maintained. 

Audit tools are typically spreadsheets. Whiteboarding tools are also useful. 

Audits insights, gaps, and opportunities fo optimization.

Steps to create a content inventory.

Use technology. And teammates. Crawling tools, exports, and collaboration. 

Set up a site crawl. So much you can get, will extract a lot of data you need for an inventory. Also set up custom extractions, such as authors, publish and modify dates, and comment counts. 

Then you clean the export. You get a ton of information, not all of it needed. (Don't lose your original crawl file.) Clean out pages hat aren't required for content inventory, such as CSS, JavaScrpt, and redirect pages. Object it to capture content. 

Build the inventory. Non-crawl columns that could be needed include original and last-modified date, refresh date, template, author or content source, technical home. 

Some might need to be filled in manually. This is where teammates come in. But also let the technology help where possible. Use text to columns. Can help with URLs to understand where content lives. 

Why maintain an inventory? Know thy content. Avoid redundancy. 

Tools and software are not the solver bullet. It's aligning people and processes. 

Document a usable and scalable process for maintenance. Important to document who owns the fields. Make this part of your content team's onboarding. Educate them why inventory maintenance is so important. Delegate inventory owners and check in regularly. 

For content audits, why not audit everything? You likely don't have the time. Determine which content you care about to get the insights you need. Identify audit criteria and stick to it. Can always go back and audit later. Resist the urge to fix things along the way. 

High value actions audits can help identify inconsistencies, uncover patterns for highest conversions, and uncover areas of opportunity.

Tuesday Afternoon Keynote

 Phantoms of Content Strategy: The People Who Help and Hurt Content Projects and What You Can Do About It

 Michael Haggerty-Villa, Director of Content Strategy at Teradata started by telling a "true" and hopeful story, saying that one of the most effective content strategist he ever worked with wasn't a content strategist. It was a product manager. So much of content decisions come from product people. 

It is just "content strategy," no matter who is doing it. You don't have to be a content strategist to do content strategy.  

Stop talking about content strategy and just do it. Develop and use standards, systems, and processes that boost content. Guide others to ask the right question. Prepare to lose and be disappointed--and keep going. Celebrate the wins, big and small.

The DIY Revolution: Crafting a Content Experience to Drive Revenue and Save Support Costs

 Alisa Conboy, Director of CX at DocuSign, told us about how she would get so energized at past LavaCons and go back to the office ready to change the world--only to get hit by reality. 

Our north star, idealized goals, were personalized use assistance, in-product intelligence, automated content delivery, and analytic insights. 

Reality: Zero budget, limited skillset, and no leadership sponsorship. 

So we all suffer. Customers struggle, team morale suffers, and the company pays the price. Customers judge perceived value based on content. 

Our options are to wait, or to not wait and adopt a DIY attitude. Look into our toolset and figure out what to do. 

Projects selected included home page update, new "landing" page, content debt reduction, an enhanced feedback tool, and Google optimizations. 

$12M support cost savings, not from case deflection, but from just when support could point to existing content and close the ticket. 

The numbers got executive attention. It built credibility and trust, unlocked cross-team collaboration, and got invited to the "table."

Learned that adaptability is king, less is more when pitching an idea, build credibility from small wins before asking for more, and partnerships add power. 

And it unlocked our future.

Taxonomies in the Age of AI: Are They Still Relevant?

 Rebecca Schneider, Executive Director of AvenueCX claimed that taxonomies are still relevant. And even moreso today. 

Taxonomies: A system for organizing content according to shared characteristics, or a way of describing things. A taxonomy describes and organizes stuff. 

Organizations have large repositories that can be used to create LLMs. AI can leverage content to understand relationships between terms. 

Unstructured content can create confusing output. Content is not always high quality. And even if metadata is applied, it can vary in quality and availability. 

Well, when no money for taxonomy development, can still use AI. Your mileage may vary--can require training and retraining (and retraining), which can cost more money. 

AI is well positioned to look at relationships between terms across multiple types of content and data. In essence, AI creates a taxonomy as it processes content. But not entirely satisfied with results. 

A study found that LLMs were good with widely known domains, such as shopping, but do not perform well in specialized taxonomies, such as computer science. 

What goes into creating a taxonomy? Start with a traditional approach: business goals and objectives, what are current pain points, what content needs description, who is the audience and what is their content needs.

Use AI in the best of both worlds. For example, use AI to identify patterns and relationships. Can be used as a jumping off point, but needs review. Feed results back into the system for further training. 

Ontologies are a formal naming and definition of types, properties, and interrelationships of entities in a particular domain. Take taxonomies and their associated facets and describing relationships, you can enrich your data, which can further inform your AI engine.

An ontology can be used for a retrieval-augmented generation (RAG) solution. (the RAG space is a fast-moving target.) 

What is success? A well-tested, vetted taxonomy/ontology that is implemented. Structured content that is business ready. Clean data. Active governance for data and content. Contributions from SMEs because they are the ones who can judge the accuracy of responses.

Navigating AI: Real Tales from Content Frontier (A Case Study)

 Amber Swope and Joe Gregory-DeBernardi of DITA Strategies and Lief Erickson of Intuitive Stack began to explain why content professionals are important for AI roles. 

AI project roles include a technical lead, an ontologist/taxonomist, and SME, a content pro, a project manager, a prompt engineer, and a business champion. Other team roles might vary. 

AI needs content pros who know the content set, understand content structure, know the audience, and can work with metadata. 

You are not a professional communicator and not curious. That kind of curiosity and research feedback is what's going to train your AI solution to be more useful.

What are being evaluating in AI projects? PDFs? PDFs don't have semantic data. Better to find the original source content. 

AI projects are more about content that technology. When you define success, you not only define the goal, but your tolerance for failure. AI won't replace your writing tram, but they may need to adapt. Generative AI requires high-quality data. Is your content ready? 

Writers' jobs are safe for a good long time.

The Horror of Modernizing Content

 Janet Zarecor, Director of Clinical Systems Education at the Mayo Clinic, and Alan Pringle of Scriptorium started by saying they are not there to tell you what tool to use. It's about finding the information necessary to pick the right to for your organization. 

You need an executive sponsor, one who is engaged with support and visibility. 

Create the most effective paths or communication to build awareness and get buy in and acceptance of staff. 

Most important messages include business reasons for change, why employees should want to participate, impact of the change, how the change is happening, and details of the change.

All of that is critical once you enter the discovery stage. 

First, do a content audit. Not just numbers, but quality. Also do a tool audit, and not just text. tools for audio, video, graphics, and delivery.

No, AI can't do this. You have to talk to real human beings to understand their pain points. 

Understand the lifecycle of content from birth to death. Map this out; it will inform what tool you will by. 

Governance explains the standards and practices that guide content. It explains who is responsible for each step. (Use the lifecycle to inform this.)

Who needs standardization? Everyone! From look-and-feel to naming conventions, language and writing style, inclusion and accessibility, and more. When people see your content, they will know what to expect. 

Goal for content operations project is to have a single source of information for everything, for all information types. 

We don't know what's coming in 3-5 years for content delivery. So you need systems that can adapt. But also a single source of truth that can slice and dice your content however you may need it. 

Smart, semantic content is going to be the basis of all your delivery types, especially if you are going to include AI.

The Ultimate Practical Guide to Measuring the Impact of your Technical Content

 Joe Gelb, CEO of Zoomin said that we understand that the content we create is useful and impactful. The question is, how many can measure that impact?

Technical content does selling when sellers are not in the room. 90% of B2B customers reference content before purchasing. 70% of B2B we traffic is for technical product content. 

Technical content powers every step of your customer journey. From awareness, learning, evaluating, and buying, to implementing, launching, useing, expanding, and renewing. 

What is content ROI methodology. 4 step framework.

Choose your KPI, capture data, quantify and monetize, and articulate value. 

Three pillars of KPIs: infrastructure, efficiency and cost, and revenue growth. 

Infrastructure KPIs revolve around traffic, search, engagement, satisfaction, SEO score.

Efficiency and cost reduction KPIs include cost of hosting, cases deflected, M&A acceleration, publishing and implementation time. 

Revenue and growth KPIs include awareness, time-to-market, new business from traffic, product adoption, self-service rate, churn. 

The ability for customers to get value initially correlates with renewal. 

How to choose the "right" KPIs?

Not everything that can be counted, counts. Reflect company values and goals. Realistically measurable. Tells a well-rounded story with data from each pillar. 

4 main sources of data: portal analytics, industry benchmarks, internal documentation team KPIs, and stakeholders' data. 

Track specific metrics over time. Decide on Day One and measure progress. 

The road to case resolution goes through case classification. For example, how to and installation/configuration cases could have been resolved by documentation. We can have a high impact on those types of cases. 

Tuesday Morning Keynotes

 Truth Collapse: The AI Meta-Crisis

 Noz Urbina, Principal at Urbina Consulting, started Tuesday by asking what's going on? He answered that we're not OK. All of us everywhere are being affected by artificial intelligence. 

In the content community, when it comes to AI, I'm getting a lot of harrumph harrumph harrumph. Al lacks empathy and can never reason.

 Who among us writes perfect the first time? who among us doesn't need practice and training?

Literacy reshapes how we think. Logic arrived in the world after writing. 

We live in pre-AI literate societies. 

An AI is a technology can learn. Google's search engine is driven by algorithms, but engineers could always change the algorithm.

AI isn't programmed. It is grown. Like other organisms, we don't fully understand it.

Shining a Light on Internal Documentation’s Extensive Impact

Michelle Irvine, Technical Writer at Google started her talk telling us about research that had been going on for 10 years. The research is on not only documentation, but part of it focused specifically on internal documentation. 

In open source documentation we can pull back the curtain and see a lot of it for ourselves. 

How does it behave in a larger system? We found that documentation quality underpins quality of everything we looked at.Quality documentation affects the quality of everything. It is a pattern that is persistent and consistent. 

Teams with quality documentation are 3.8x more likely to implement security practices, 3.5x more likely to implement site reliability practices, 2.5x more likely to leverage the cloud. 

Teams with quality documentation see increase in productivity and job satisfaction and decrease in burnout. 

Documentation is like sunshine. 

It's not just about more and more docs. It has to be relevant. 

Writers who have taken training create higher quality documentation. Teams need guidelines and style guides for higher documentation quality. 

Documentation is work and is expensive. Figure out what is critical and focus energy in maintaining that. Delete redundant content. Curate to stay fresh and relevant. Value doc work. Where this was happening, we saw an increase in quality. 

The Unseen Challenges and Hidden Costs of Adopting Markdown for Technical Documentation

Scott Abel, Content Strategy Evangelist at Heretto, spoke with about 2 dozen tech writers, managers, and consultants who use Markdown every day, as well as people who use other technologies as well. I held some panels about the pros and cons of Markdown.

Most are using Markdown in a startup environment. The challenges and costs were not the cost of Markdown as much as docs-as-code. So now, Docs-as-code or not?

A fit for purpose publishing pipeline meets a team's requirements for managing, storing, and retrieving information effectively, even if it lacks advanced features. 

Docs-as-code helpful on small team with small, simple, static doc needs. Not better option if you need dynamic, personalized, scalable content. Lot of writers work to maintain publishing pipeline--instead of writing.

First hidden challenge is scalability. Trying to scale docs-as-code is like trying to patch a small leak in your time with duct tape. 

Second hidden challenge is personalization. Docs-as-code has no metadata, so there's no opportunity for personalized content. Nothing is semantic. 

Third hidden challenge is multi-channel publishing. If you can accomplish at all, you have to add additional modules.

Fourth hidden challenge is formatting and design constraints. Particularly difficult in docs-as-code because you don't have as much control over the content. 

Fifth hidden challenge is maintaining a fragmented toolchain. Spending significant amount of time when not done right. 

Sixth hidden challenge are build and deployment issues. As a tech writer, don't want to hear about "build and deploy" unless I'm documenting it. If you buy a software product, it's their job to make sure it works right instead of yours. 

Seventh hidden challenge is accessibility limitations. 

Eighth hidden challenge is SEO obstacles. 

Ninth hidden challenge is cybersecurity. When devs can't find the information they. need, they search elsewhere. Criminals put malicious code there for devs to copy and use. 

Tenth hidden challenge is change resistance. People who have only worked in docs-as-code environment see it as an easy way to get stuff done. They have no incentive to change. And they have never worked in other environments, systems with management and governance. 

No semantic metadata means Markdown is nothing more than a new version of WordStar. 

Why, then?

The developers like it, it's perceived as easy, and it's perceived as free.

Monday Afternoon Keynotes

 The Business Case for Content Operations

Sarah O'Keefe, CEO of Scriptorium defined content ops: content operations.

We're all content nerds. But nobody cares about content (other than the people here). Focus on business drivers. How to make content do something for the mission or the organization. 

Business drivers include compliance, cost avoidance, revenue growth, competitive advantage, and branding. 

Tell the story. Content ops can avoid costs, can get to market faster, and scale, and provides a better experience. 

The people who have the money want to head dumb stupid jargon terms, so you have to use dumb stupid jargon terms. 

Writing for a World that Doesn’t Read

Eric Kuhnen, President of GlobalLink CCMS, said that the National Endowment for the Arts issues a study every 5 years: "Reading at Risk." Literary reading in America is declining in all groups. It parallels a larger retreat from civic and cultural life.

What if they are wrong? What if they completely misunderstood the trend? What if they misapplied the data?

The Story of Cathy Content

Stefan Gentz, Global TechComm Evangelist at Adobe, noted the rise of AI engines in the past couple of years, including Adobe Firefly, as well as ChatGPT, Google Gemini, Microsoft Copilot, and others. 

AI is changing a lot of things, including how we consume content and how we search for content. No one wants hundreds of pages of search results. 

Connecting systems with our content tools is often a problem. But we can solve that. 

Managing too much content is also a challenge. 

A CCMS allows you to create small, reusable content blocks and then assemble them for publication. A CCMS is a single source of truth for content.

Content is THE Most Strategic Asset for Unparalleled Customer and Business Success

 Bernard Aschwanden, of CCMS Kickstart, began by defining corporate assets as resources that hold value. Historically, content was seen as a cost, necessary for business operations, not something that built revenue. In many cases, content was intangible. Content, specifically, structured content, connected company, brand, and people. 

Centralized content streamlines collaboration across organizations. Content informs prospects. Content plays a pivotal role in the financial success of a business. 

Strategies include centralization, a single source of truth for content. Centralization streamlines operations--and not just for TechComm. 

Consistency includes using the same format, style, and content patterns. Consistency matters; it builds trust and ensures compliance with industry standards. 

Automation helps with repetitive tasks. It reduces human error and saves time, enabling teams to scale. 

Content reuse matters. It supports content consistency. Reusing content eliminates the need to create the same content over and over. Modular content saves time and ensures compliance. 

A goal without a plan is just a wish. Link workflows to strategic goals. Regularly review where you are. 

Taxonomy and metadata matter. Supports content as a strategic asset. 

Information Architecture (IA) supports findability, usability, and efficiency. A well-structured IA ensures a seamless user experience. 

There's solid ROI numbers for structured content.

Achieving Customer Success Through Self-Service Support

 Patrick Bosek, from Heretto and Barbara Green, UX Content Lead at ACS Technologies explained that there was a lot of change happening in the product line with slow content production in wikis and PDFs. There was limited reuse and not scalable. 

Started developing cloud-based software. Spent long hours as the product line expanded. 

The objective was a holistic content operations ecosystem. Content team needed more control of our publishing environment. Getting no love from outside the organization. 

The content had to serve multiple audiences with 4 user types, 3 faith profiles, and many versions of software. Then acquired 5 companies in a 2-year period, which included all their content. 

Key content ops capabilities: personalization, content reuse, version control, push-button multi-channel, API delivery, and findability. Interestingly, personalization is one side of the coin where the ogther side is content reuse.

Just 10-15% reuse saved 225 hours per month. 

The first step is moving to structure(d content). Scared to move to DITA, but knew we had to. Understood the value and knew it made sense. Initially limited to our newest (cloud) product. Along the way, revamping strategies, creating standards, right down to how we were going to name files and other assets. Used a conversion company for this step. Positives and negatives; it's messy. 

Storing content at the time in a Git repo created friction with authors.

Next step was to bring in content management technology. Main concern about DITA was that it would be hard to hire qualified people. Needed a user-friendly tool. The tool needed to appeal to us as writers. 

Then was launching the new website for content. 

But help site v2 was a bit of a mistake. The design wasn't scalable as the company acquired more products. 

A v3 design that was scalable didn't require re-authoring content because the content was already there and structured. It is now the single source of truth for 27 products. The documentation website is now 56% of the company's total web engagement.

Companies that don't build great help experiences can be leaving half of their customer interaction on the table. Companies invest millions in marketing site and a tenth of that on documentation sites. The Technical Publications department has this impact and should be getting the funding.  

In a digital-first world, help sites are the first place most software customers look for issue resolution and source-of-truth.

The latest step was ti implement in-product help. One click takes users to information, including information, release notes, community, and chat.

The Unlikely (But Necessary) Marriage of Content and Engineering

 Joe Gollner, Managing Director at Gnostyx Research Inc., said that when you start a story, you need to follow it to its conclusion. 

We don't have a firm definition of the word "content." From Latin: "that which is contained." It is "potential information." Can spawn multiple information transactions. We perform information transactions. The content is what's contains, and it may be delivered in many different ones. Technology is always involved; something happens to turn a content assent into an information transaction. 

Content is therefore an "asset." Something of potential future value. 

Content is also an organizational expression. It is collaboratively developed by teams. It is always hard work. It is basically what an organization knows. 

Content is a complex composite artifact. It's grounded in data resources and connected to knowledge precedents. 

Engineering is surprisingly dependent on content assets and processes, the documented understanding of what a system is intended to do. 

Content engineering: The application of engineering discipline to the design, acquisition, delivery, managements, and use of content and to the technologies deployed to support the full content lifecycle. 

AI consumes content, feeds off it. The worst case is that AI runs amok. It consumes information indiscriminately and without management guidance or guardrails. This form of AI is as popular as it is dangerous. 

A rigorously engineered approach to content creates the conditions for effective AI utilization. The goal is an ecosystem where content and AI work together. AI offers an opportunity to see content recognbized as the strategic asset that it is.

How We Saved $57K in Deflected Support Cases in Five Months

 Pam Noreault, Principal Information Architect at Ellucian said that most organizations think that content costs money. The goal is to show the value of content. 

It takes a village, including doc vendors, search vendors, UX, information development leadership, support and product management. The UX team did research for us as the project was going.

Customers couldn't find information quickly, did not understand what site to use, found older content that was irrelevant, and found things that were broken. 

Identified 5 things to improve:

• Improve the user experience
• Increase self-support and case deflection
• Consolidate content to one site
• Increase site use
• Show iterative improvement. 

While there was $57K savings in case deflection, an estimated $13.4M in implicit case deflection. These are cases where someone goes to the docs, does a search, and doesn't open a ticket. We also saw that topics views went up 91%. 

In 2024, added continued to simplify, added AI to the site to provide code explanations and topic summaries. Found $181K in explicit case deflection savings and $25M in implicit case deflection savings. 1917% increased topic views.  

Find what you can measure. 

It takes time for perceptions to change. Not all changes are well-received. Overcommunicating still misses users.

Monday Morning Keynotes

 The 2024 LavaCon conference opened with three 20-minute keynotes. Here are some highlights.

Building Influence: A Roadmap for Content Expertise and Leadership

Melinda Belcher, Executive Director at JPMorgan Chase talked about tactical strategies to establish yourself. Good advice for anyone in the workplace. 

What is influence? It is not authority. Influence is about inspiration. It's about bringing people along with you. It is not something that will ever be handed to you. You have to earn and build it. 

Often you will have influence on people and not realize it. 

Establish credibility. It's what influence is built upon. It's about meeting or exceeding expectations. Unfortunately, it means you have to be your own cheerleader. It is a necessary evil. Start by taking a hands-on approach and partnering with your peers. Know your industry. Share with not only team members, but cross-functional folks. Identify and showcase quick wins.

Invest in yourself. A strategy of continuous learning is not only good for your career, it's good for your life.  Set up a personal learning system. Stay accountable to yourself; find an accountability partner. Part of that is to write it down. You're much more likely to do something if you write it down. Sahre what you learn with others. 

Cultivate trust. Set the pace, protect your space. Check in purposefully with people, make sure they have the information they need. Document and explain yourself. When you take the time to be mindful about that with them, they will be mindful about that with you. Cultivate a "short toes" culture by putting outcomes over ownership, encouraging cross-functional collaboration, and separating personal identity from professional output, and that way no ones toes get stepped on. 

Finally, spark joy in yourself and others. Relationships are the ones you chose. Work is situationships. You're together not because you want to be but because you need to accomplish something together. Make it as positive as you can. Cultivate a joyful mindset around work. Choose inquiry over advocacy. 

Metamorphosis: Empowering Our Craft's Evolution in the Dawn of GenAI's Era

Fawn Damitio, Sr. Manager of AI Infrastructure at Meta, and Peggy Sanchez, Sr. Manager at Hewlett Packard Enterprise said that as content developers, we have to learn from the past. Technical writers were some of the first usability testers. 

The Internet exploded in the 90s. Don Norman and Jakob Nielsen helped found the discipline of user experience. 

We're already experts at researching and finding information. We have to learn to recognize AI use cases and to develop prompt expertise. These are critical for career development. 

Have to get into the AI race. Find a way to use it and get better at it. 

Can't do this is a silo. Need to share. Thrive through the power of community. 

The tool is not going to take your job. Someone who can use the tool better than you might take your job. So learn how AI can make you more productive. Ask AI to explain something to you. AI can help you do things much like an intern. It can hell you problem solve. There are ways that AI can make you more productive. That's not your job, but it can help you get your job done faster. 

We're in an immature field, so always, always, always have to review and edit the results. 

Are Your Canaries Still Singing? An Optimist's Guide to Designing for Failure

 Relly Annett-Baker, Head of UX Content Strategy at Google said that the idiom "good vibes only" sounds like it comes across as a threat. "You're bringing problems not solutions" makes it sound like failure is not acceptable. But you have to plan for failure. 

The tech-specific version of "good vibes only" is "move fast and break things." People who move fast and break things: toddlers, speeds runners, world record karate choppers. But what do you do with the mess?

Failures aren't inherently opportunities. It's a neutral situation. It's what you do with that situation. 

But they aren't inherently bad either. Everything is data. And system with many moving parts, something will likely go wrong. And if something will go wrong, like to prepare for it. 

Siloed success is the number one cause of systemic failure.

 

Optimizing Content Operations: Calculating the ROI of your Production Pipeline

 Rahel Bailie, Content Solutions Director at Technically Write IT, practically shouted from the rooftops during her afternoon workshop that you raise the value of content by reducing content debt.

You have to accept content as a value stream, not a cost center. 

As with many things, you have to know where you are before you can know where you want to go. The first step of calculating ROI is to get a baseline of your current operating model. For example, you might have a workflow via email and attachments to communicate. Or your audit trail might be in spreadsheets. 

Next, standardize your content structures. Decide what structures will work for your users. 

There are more steps to review, and once you have evaluated each one and figured out how you can streamline, compare the cost of each in both time and money. 

Look for the 8 types of waste in your processes:

• Transportation: moving files from folder to folder
• Inventory: Outdated or duplicate content
• Motion: Recording actions in spreadsheets
• Waiting: Manual review processes
• Overproduction: Multiple copies of the same document
• Overprocessing: Creating variants in separate files
• Defects: Hard coding content into code
• Skills utilization: People doing rote tasks easily handled by a computer

Building and Maintaining a Prompt Library as a Business Asset

 Jason Kaufman, Principal AI Content Strategist at Irrevo suggested during his afternoon workshop to challenge AI's own work, saying that once you get a response, you can then ask where it got its information. 

Over time, models change. In your prompt library, the behavior of the prompts will change, known as prompt drift. This is definitely a reason for curating your prompts. 

Every time a prompt is created, get a sample output. Ensure the result is what's desired. Over time, you can compare future outputs to your original sample, and tune as necessary.

Building Taxonomies to Leverage Content

 Heather Hedden, Taxonomy Consultant at Hedden Information Management, began her afternoon workshop explaining why taxonomies are important. Explaining that a taxonomy is a controlled vocabulary organized into a hierarchical structure, she added that taxonomies help control and organize content. 

The catchphrase is "things, not strings." An example of this is results of a search can be extracted from concepts, not just the words. 

Taxonomies provide consistent concepts for tagging and retrieval. They bring together synonyms for the same content. They organize concepts into hierarchies, which guide users to specific topics, and into facets for filtering and refining searches. 

An ontology is a model of a knowledge domain. it includes relationships, they are customized semantic relationships.  

Knowledge organization systems: term list, name authority, taxonomy, thesaurus, ontology, from simplest to most complex and expressive. Ontology is the semantic model, can connect to taxonomies, term lists, etc.

User Research For Content Teams: Hands-On Testing Techniques

 In her Sunday workshop, Bridget O'Donnell, Senior Content Strategist at Alibaba Group, led participants through some techniques for doing user research. 

Screener questions are used to recruit users for the research. Such questions should be designed to not be leading, but more vague. 

For usability testing, a maximum of 8-10 tasks is optimal, fewer if the tasks are complex. Surveys ideally would also be 10 questions or fewer, even though online surveys are are longer. Limits help encourage people to participate. 

Avoid biased questions; biases can include confirmation, selection, social desirability, interviewer, and primacy/recency. Others include question order, observer expectancy, and non-response biases. 

To avoid bias, ensure a diverse and representative sample, create neutral and unbiased questions, and randomize the questions. 

Often asking a question directly is the worst way to get information. Develop open-ended questions. A yes/no answer is an indicator of potential bias.

Design AI Prompts and Workflows to Improve User Journeys

Noz Urbina, of Urbina Consulting began his workshop by surveying attendees about their familiarity and use of AI. 

An LLM is a large language model, not a large database model. It can't do calculations. 

Anything in a conversation is unique to that conversation. If you chat to the same persona in a different conversation, that persona will know nothing from the first conversation. 

Content Value Design is an approach for analyzing audience journeys and stakeholder ecosystems. Start with building persona trait profiles. 

Journeys are questions over time. What are people trying to understand, learn, do. What are the questions that drive interactions. These are what drive journey maps.

Two Weeks to Go!

 Here we are, Sunday, October 13, 2024, and it's just 2 weeks to LavaCon. And I don't know about you, but I'm raring to go already. I understand that we're expecting hundreds of content professionals mixing and milling about in the conference rooms and hallways of the Portland Hilton, and the anticipation is already building.

While the conference technically starts on Monday, the 28th, Sunday feature pre-conference workshops that take deep dives into several subjects. There may still be seats for last minute registrants, but note that the schedule-at-a-glance on the LavaCon website has already been updated with 3 workshop cancellations, 2 in the morning and 1 in the afternoon. 

Morning workshops include AI prompts and user research, while afternoon workshops feature content operations and taxonomies. 

The (very) long range forecast suggests typical autumn weather for the Pacific Northwest: daytime temperatures in the 50s, nighttime temperatures in the 40s, with lots of clouds and good chances of showers and rain. I mention this in part to remind folks that there are conference-related activities that are not in the host hotel. Monday night is storytelling night and Tuesday night is a dragon parade followed by karaoke. 

But also, it's Portland. Among the places to save time to visit are the original Voodoo Doughnuts and Powell's Books. 

This blog will be a living document of the conference. I'll be live blogging the sessions I attend, sharing brief notes about the speakers' subjects. I'll also be taking photos throughout the conference, and hope to share some of the best ones here. (Some will also be shared on the conference's social media pages.) 

There also will be a daily newsletter with conference news and highlights, information about sponsor/vendor conference-related events, and most importantly, information about any last-minute changes to the conference schedule. Look for copies every morning. 

I love both attending the conference, learning all that I can from my peers and interacting with all of those peers over the 3+ days, and doing all the work of the blog, the newsletter, and the photography. It is exhilarating and it recharges my interest and my energy in my work. It is also a bit exhausting. I wouldn't have it any other way, and I look forward to seeing all of you in 2 weeks.