I recently completed the first stage of overhauling the robotics webpage at ubuntu.com. As a product manager for Canonical, one of the things I am responsible for is the success of our robotics initiatives. Whatever ‘success’ may mean (more on that in a moment.) I intend for this to be the first in a number of blog posts to reflect on Product Management work that I do, the decisions I made, and why I made them. Hopefully in such a way that I can revisit this post when facing similar challenges and learn from my mistakes and the decisions I made. This is in no way necessarily the right or best way to do things, it is just the way I did it, for better or for worse. A black box of thoughts.

Setting expectations

First and foremost before starting a new project, I try to make a plan and set expectations. I.e what are we trying to achieve, why do this in particular, who is the intended audience, how does this benefit them, what would success look like? etc. Both the planning process and setting expectations could be another blog post, but here I’ll just speak to setting expectations as they pertain to writing the web pages.

Canonical is not a robotics company. The robotics team at Canonical works to support robots, specifically ROS and the community around ROS at large. So we are not trying to sell a robotics ‘Product’. So:

• What are we trying to achieve? — Increase awareness and adoption of Ubuntu robotics and ROS on Ubuntu.
• Why do this in particular? — The website is a hub of traffic where we will get the most exposure, most quickly. And it had become outdated, not accurately reflecting intentions.
• Who is the intended audience? — Organisations interested in using ROS on Ubuntu to drive adoption in industry and developers/new roboticists interested in getting started.
• How does this benefit them? — More information to help users make the best decision when thinking about using Linux for robotics.
• What does success look like? — Increased page views and clicks on links in the page.

Why the change?

Making perfect the enemy of good is a fine way to burn time. It’s bad time management, bad situational awareness and poor prioritisation learning to manage your inner perfectionist is important. Is there a real reason for doing the work? Why not stick with what you have and commit your time to something else. My reasoning for this work was as follows:

The work being done by the robotics team at Canonical was underexposed. When I joined Canonical, the robotics team was doing great work. But really only people in the ROS (Robotic operating system) community knew about. The obvious thing to do was to let people know the great work going on. In my view, this was reasoning enough for at least updating the website.

On the old page, it was hard to discern how Canonical was involved in robotics. It talked a lot about Ubuntu Core and security and a little about getting started, but it was very surface level. There were, and still are, links to the documentation and pieces of content that the robotics team created, but surprisingly little about what was actually being done. Another reason for at least a refresh.

Finally, looking at the old page, if I told you that the robotics team works predominantly on ROS, you would be surprised. The acronym was nowhere to be found. At Canonical, until recently the focus was on IoT. So despite everything going on in robotics and despite robotics being more than just IoT, the following was the case:

Which isn’t necessarily bad, but it’s part of the reason I was hired, more hands-on-deck. To bring robotics into the light and make sure the world knows the pretty phenomenal work those engineers do. The cumulation of these three things I think warranted an update, a refresh and more content.

What’s new

We know why the change, now I’ll describe what has changed and discuss why those changes were made afterwards.

Overview

First was the update of our old page. This was mostly a restructure to fit the new intention. Information was taken out and put onto the other new pages to make this one more concise and consumable. We added new links and more up-to-date content. The links throughout the pages will now send you to learn more about robotics on Ubuntu. And we refreshed the page with new logos under “Companies using ROS on Ubuntu.” These are some of our favourite robotics companies that we know for a fact are using ROS on Ubuntu. Paving the way.

What is ROS

This page came from a desire to show our support and admiration for ROS. The majority of work done in the Ubuntu robotics team is in support of ROS. The goal was to explain ROS to the uninitiated and speak to the benefits of using it. There are brief explanations of ROS and ROS 2 with links to all the right places. And we look at more companies using ROS on Ubuntu and the benefits of doing so. This page will slowly grow and accumulate more content over time. We will be attaching whitepapers and case studies that show exactly how and why companies succeed with ROS.

Community

The heart of ROS today is the community around it. The hardest thing about a community is how to join one, so we made the focus of this page, getting involved. Surrounding ROS and Ubuntu are thousands of developers discussing and contributing to these technologies. There are pages of documentation, and tutorials in both communities to help get involved, but new interested people don’t know where to start. This page is a hub for community content and links to get involved with ROS.

With all of this complete, another difference is this:

How things have changed

So, we know the why and the what, but not the how? How has it changed, what makes this new thing better? Well, let’s break it down.

SEO

To the uninitiated, I apologise. If you do not know what SEO means, I’m about to lift a veil and you’re unlikely to like what’s underneath. SEO stands for search engine optimisation. It is a way by which you can make intentional decisions to have Google pick up your website and list it on the first page of a relevant search. SEO is often about making small modifications to parts of your website. When viewed individually, these changes might seem like incremental improvements, but when combined with other optimisations, they could have a noticeable impact on your site’s user experience and performance in organic search results.

This optimisation really comes down to creating compelling and useful content. Users know good content and if it’s good enough they’ll want to share it and direct others to it. Building this reputation digitally and with word of mouth is how to optimise. But what qualifies as compelling and useful? In these pages, I try to do one thing well.

What the audience wants. Our target audience is people new to the idea of ROS on Ubuntu, at whatever commercial level they are at. My wager here is that if they are already interested they are looking for explainers and how-tos and if they have no idea what already exists, a summary would be useful. This is why the pages spend words explaining things and why the majority of links of the page go to things with titles that imply “how”.

Wordsmithing

To me, this is distinctly different from SEO. I treat SEO as a way to tune writing and content as a whole. A guide to optimising words. SEO can do some work to make up for poor writing, especially with an already well-established website. But ensuring the right words and the best messaging is just as important. In order to create compelling content that users want to read, they have to want to read it. So it has to be good. What that means can be subjective and frankly changes between mediums. Here are four things I try to do to make for better web page writing:

• Backup. If you can’t back up your writing with examples, with data or with actions then its speculation, the technical term, is bullshit. This can be difficult if you don’t have data and the temptation is to just say things. But pains should be taken to back things up as much as you can. A reader will usually be able to tell if you’re full of it.
• Specificity. Don’t beat around the bush. On a web page, you will only have a certain amount of words and only a certain amount of space. Getting to the point and getting to the point cleanly is important.
• Simplicity. Most web pages are not documentation, or textbooks, or research papers. Unless the audience is intentionally very technical, technical information has no business on the page. This can be hard to avoid when talking about technical things but simplicity is an ally to clicks.
• An active voice. When reading English readers prefer a subject, verb, object sentence structure. Which reads better: ‘People are bored by passive sentences.’ or ‘Passive sentences bore people.’

Structure

If you look at the pages I linked to above I encourage you to notice the order. Introduction, why you should care, and then what we’ve got. The page titled ‘overview’ is the first you will come to through any ‘traditional’ means of getting to the web page. Meaning we need to cater for the uninitiated. If you know what you want, you’re going to scroll down the page, if you’re just browsing we need to get your attention off the bat. An introduction needs to be a summary and you need to give people a reason to stay.

Then you need to give people a reason to stay. But the trick here is to talk to their reasoning, don’t start talking about how great you are, you’ll have time for that later. You have to address why they came to the page, talk to the problems they might have, be benefit led. If you make it focused on you, you lose. It's unlikely that they came to the page for you, they came to the page for themselves, for information.

Once you’ve explained what the page is about and told them you understand their problem and have a solution, you can tell them how. But again, you don’t necessarily want to talk about yourself. Only you care what you’re doing. You want to explain what you have in the context of why they came to the page. Go ahead and look at the pages of your favourite, successful product.

For example, I am a big fan of the Raspberry Pi, their product, their mission, their attitude, everything. If you go to their website (in a new tab) you see it’s done very well. It starts with an introduction, a nice picture and their mission statement, you’re in. Then they show you a bunch of projects and initiatives that answer your questions with actions. And finally, if you scroll down you see what they have, what you can do with them. But not in a “look at how great we are” way, in a “here is what we have for you” way. Its a big but apparently subtle difference.

Outro

Well, that was a lot, I could go on and dig deeper but this is turning into a ramble. With regards to the pages, there is more to come. I have recently been reassigned to look after other products so will have less of a hand in robotics. I intend to hand over the work to my new colleague to pick up. The broad strokes will be two more pages under the /robotics URL and additional content to add to the pages. When I get the chance to work on web pages again however I expect to update this post.