June 25, 2010
Back to the future?
Beyond Micheal J Fox’s charming good looks, the Back to the future movies were built around the idea that its as easy to mess things up as make them better. Mark Kingdon’s been sent packing and Philip Rosedale has returned with the title of Interim CEO, Is this good? Bad? Indifferent? I expect it will be a mixed bag.
Second life is at a crossroads. Growth has stalled, several initiatives have stalled, and 1/3 of the lab staff got handed walking papers. Growing beyond a niche is hard, and several Virtual Worlds and their associated companies have gone bust in the past year. SecondLife has the advantage of basic profitability, and a user base which, has enthusiasm for the environment, when the lab isn’t busy making their lives difficult.
Philip has remarkable strengths. He has vision, he is charismatic, and he is wicked smart. Above all, he is passionate. . Philip is also, in part impractical, a dreamer, easily distracted, and undisciplined. These are, in many ways two sides of the same coin. Creativity and Vision are not enemies of disciplined slogging, but they are rarely close friends either. Whether Philip sorts things out depends a lot on how he parses the basic problems of SecondLife, and how he attacks them.
So.. given this, here are some thoughts on what he should do.
Six things to do:
- Start by listening to your customers, and listening hard. Feel free to call them residents, but keep in mind they pay your bills.
- Play to your personal strengths, and bring in people who complement them, and bolster your weaknesses
- When you look at things like Avatars United, Second Life Enterprise, and other projects which haven’t prospered, look at not the simple fact that they failed, but WHY.
- While you are getting the core business sorted out, get someone good asking hard questions about how to monetize things which aren’t land. The Lab has resources it doesn’t monetize, and figuring out how to leverage them is vital.
- Keep firmly in mind that Linden Lab is 200 employees and no longer a startup. If Second Life is to avoid being a niche player, it will have to grow again, and the management team needs to reflect this reality.
- You have a huge “technical debt” Start paying it down. Yes, its painful, but it prevents Second Life from scaling and evolving. Cleaning up code and architecture isn’t glamorous, but it unblocks progress.
Listening is hard, but vital. If you pay attention (LISTEN) to only one thing, start here.
Listen to what users have been saying for months. Fix search. Fix Group IM, Raise the Group Limit, Do Windlight estate settings, figure out how to allow megaprims in appropriate settings. Listen to your customers about viewer 2.x and tell them you’re supporting 1.23 while you sort it out.
Don’t run off to “fix” things without listening to the residents. When you think you know how to fix something, describe the fix, and then, listen again, to what people say. Run Serious Betas, and let people play with things early and often, and when they complain, again, Listen. Don’t run a two week beta, fix three bugs and say “We listened.” Start early, share code often and actually allow tester input to influence the end result.
When you listen, actually respond to feedback. Yes, it may take a little longer to complete the process, but Viewer 2.0, and Search have shown very clearly, how badly things go when you build a lot of code and don’t listen to the community as you build them.
Work on saying “yes” to your customers, not “no.” The lab has spent a lot of the past two years telling its customers no. Yes is a better answer. This may mean telling lawyers, technical people and maybe even Phillip Linden to figure out how to make yes possible. Second Life needs some yeses.
Look at what people do to make things happen in second life, and then listen to them about how to make it better. A lot of your customers have figured out how to run big events using voice, web streams, chat bridges and other tools. Listen to them about how they do these events, and why, and then talk to them about how Linden Lab can make this easier. Then go back to them when you think you have answers, and listen when they give you feedback.
Play to your strengths and understand your weaknesses
Operations needs a solid hand, make sure you keep one there. The grid has become massively more stable in the past two years, backslide on that, and everything else is at risk. Take your time when piloting and deploying code. The steady drone of pilot roll, rollback, pilot roll, rollback sends a bad, unnecessary message about diligence in testing.
Put in place a full up CTO. Someone with a solid track record in building distributed systems, and understands scale. Listen to them, and give them enough authority to put some focus on solving the long term technology issues which get in the way of growing out the grid and single region concurrency. Tell them you know there is a deep technical debt in the current code base and support them in paying some of it down.
Get someone who is good at communications. Listen to them. Linden’s PR has been painfully bad and needs to get a lot better. Speak to your residents honestly, speak to them often and work to make that communication a two way street. A brown bag where people vent, but have little or no impact on behavior is not communication. When you hold in world events as part of communications, do a better job. The community has worked out how to hold large events with streams both in world and on the web. Linden should learn from best practices and not just “turn on voice and pray.”
Get someone who understands user experience and let them work closely with the resident community and figure out how to move beyond the current viewer 2 mess. Giving first time users a cleaner, simpler experience is important, but not at the cost of neutering the client in daily use. Defaulting some settings to permit a simple experience for new users, while allowing experienced users to personalize the experience is the way forward here.
Put someone in charge of the challenge of connecting users to exciting experiences on the grid. Make sure they understand why people thrive in SecondLife and how to make those experiences more accessible. Events and shared experiences are the hear of SecondLife. New users should have quick and easy ways of finding new experiences. SecondLife is not a game, but look at using game like rewards to get users to explore the grid, explore shopping, explore events. Add some low key rewards for people when they spend their first 100 lindens, when they attend their first event, when they make friends.
Learn from the failures
Look not just at what projects failed, but why they failed. A core appeal of second life is immersion were projects like Avaline and Avatars United helping, or hurting immersion. A core asset of Second Life is the rich content cloud and rich creator community. Did Second Enterprise tap into the asset cloud, or cut off enterprise users from convenient access to the content and economy on the main grid? A core value of any growing environment is the network effect of connecting users to users. Did Second Life enterprise have any ways of connecting enterprise users to the greater community, or did it hide them away from the core experience.
Retrenching to focus on the core experience is fine, but failing to understand why various project failed will prevent similar projects from going forward successfully in the future.
Second Life is an immersive, creative experience. Linden Lab needs to look at projects which enhance the core strengths of the experience, and recognize the unique nature of a virtual world. Turning the lab’s back on the rest of the web, and on customers beyond the current user base is just as much a recipe for failure as ignoring the current users.
Monetizing beyond land (and business models)
Linden lab has several core assets. The grid itself, and the actual regions are the heart of the service. Revenue from land is both the Lab’s most secure revenue source and biggest challenge. At the end of the day, land rental is value added server space rental and servers get cheaper every year, and margins get thinner every year. Second Life has important assets beyond the grid which are not monetized.
The content cloud, and access to the economy of second life is a huge potential source of revenue. Second Life has some of the richest most innovative content in the internet. Competitors to second life are emerging, and will continue to emerge. Helping make Second Life an internet wide hub of content creation and selling is a huge opportunity. This requires getting beyond the current model of a single monolithic grid and into spaces where Linden Lab is a hub not just for land but other services.
Identity and users are another important asset the lab has consistently ignored. Look at how to allow users to take their digital selves beyond the current SecondLife grid and how the lab can be a hub for identity in other spaces. Don’t couple these opportunities with requiring people to expose a real life identity. Second Life is as much about coming *to* the virtual world as it is dragging the real world into the virtual spaces, but by the same token, people’s virtual identities are deeply important to them, and they want to “be who they are” in many parts of the web, not just the current grid.
Make “premium” membership meaningful, and make “free” membership less valuable. Free, ready access to the grid is important, but, there is no reason to not put significant limits on what you get for free which encourage people to become premium. Free accounts are important, because they let people quickly enter and explore the world, but look at limiting them so that people are encouraged to become premium memberships. Separate out avatar name from account, and allow people to have alts share premium status. Be careful about privacy when looking at these choices, but recognize how and why people create identities within the grid, and work to provide a model which both helps users do what they want to do, and gives the lab better insight into what customers are doing, and how provide them with real value.
As time goes on, Linden is going to face an every more diverse ecosystem of competitors in its niche. The lab needs to position second life to be part of a growing ecosystem, and to profit from the success of the whole ecosystem, not just the current user base. Sorting out how to monetize all of Linden’s opportunities, not just land sales is a vital part of thriving in a larger web of virtual spaces.
The lab is not a startup
You said it yourself this week. You have been working on parts of Second Life for a decade. Even after laying off 100 people, Linden Lab is larger than most startups. The lab needs to be more focused, and more agile. This does not mean the lab can shoot from the hip. SecondLie Scribe used a wonderful phrase in this panel discussion “Manager of managers.” Its a description of something that is very much not what startups are about, but is what bigger companies need. Look forward for how the lab needs to function in its next decade, not back. Take the title of “interim CEO” seriously, and find people who can focus on running the company well, so you can focus on Vision and Strategy, and listening to your users.
Most companies don’t get second chances to grow beyond their initial success. Linden Lab may not either. Second Chances are rare. The lab has a passionate but nervous user base. This is a double edged sword. Passionate advocates can be your best source of ideas, inspiration and new users. Passionate advocates thwarted and made bitter can be your loudest critics and take their ideas and passion and go elsewhere with them. Second Life needs to execute less like a startup and more like a nimble successful company. There are people who have watched what Linden Lab did well, and what Linden Lab did poorly. Unlike the field 10 years ago, there is a lot of technology, there is a large community of developers, users and potential competitors who will be looking to see if they can succeed where Linden Lab has not.
Second Life does not have the excuse of immaturity, or the luxury of obscurity. Customer Service, development, and user outreach need to be professional and serious. If Linden Lab does not learn from its mistakes and move forward decisively others will, and the challenges to Second Life will grow.
Paying the lab’s technical debt
This is a painfully technical point, but not a trivial one. Technical Debt is very real and does huge damage to companies and projects which ignore it. Linden Lab has a decade of technical debt. Some of it has been paid down. The changes which have made going from Havok 4 to Havok 7 far less challenging than the move to Havok 4 are a good example. On the whole tho, the phrase “We can’t do that, our code isn’t in a position to do that” is a phrase heard far too often in Linden Lab.
Technical debt has gotten in the way of scaling regions, adding caching to regions, offloading work from regions and breaking the grid into more manageable chunks. Technical debt gets in the way of supporting new business opportunities for the lab, and slows every type of development. Part of moving beyond a startup mentality is paying down technical debt on a regular basis.
Linden Lab needs to make understanding its technical debt, and planning to pay it down a part of the technical culture. This is part of moving beyond being a startup and becoming a company that can execute efficiently. Someone at the lab needs to be an advocate for investments which will improve long term scale, long term flexibility and long term stability. The lab needs to position itself to adapt to changes in the world such as mobile devices, and cloud computing. Technical debt gets in the way of business and technical flexibility. Arcane, rigid code leads to arcane, hard to fix, change and use systems. Users don’t want to know why its so hard for the lab to permit “undelete” of objects, or rollback of a parcel’s state, they want the grid to what they want, in ways which delight them. Paying down technical debt is a vital part of delighting users.
Second Life is at a very real crossroads. Linden Lab needs to take serious steps to sort out very real challenger. M was chosen by Phillip, and the board at Linden Lab to lead Second Life beyond where it was in 2007. Bringing back a founder to a startup is fraught with peril. There were reasons Phillip and the board wanted to bring in a new team, and those reasons haven’t gone away because the team they brought in failed to produce the desired results.
Linden Lab is fortunate to have cash, to have a fairly sustainable business model in the short term. The lab needs to fix serious problems and fix them soon. Second Chances in business are rare, and third chances vanishingly so. Linden Lab needs to figure out how to listen to its customers, make a product they want to keep using and figure out how to grow again.
Second Life still has huge potential, but the longer growth stalls and alternatives mature, the harder the road ahead will become. Linden Lab needs to focus on delighting its current customers and positioning itself to grow beyond the current user base. This is a huge challenge. Philip Rosedale has met a huge challenge in getting Second Life to where it is today. I believe that Philip and Linden Lab can meet the current challenges, but only by being smart, disciplined and willing to Listen and Learn.