INCURABLE DATA GEEK

I just recently – finally – finished reading Zappo’s CEO Tony Hsieh’s Delivering Happiness book, which I liked a lot. With my first company I didn’t think much of employee happiness; if something couldn’t be solved with spreadsheets, it didn’t matter.  And something as fuzzy as happiness certainly couldn’t fit into that framework.

The approach that we’ve taken at my current company Trulia since the 2005 beginning is quite different and we had the hypothesis that if we can even try to measure Employee happiness, it will always be front and center in our minds, and possibly we can even improve it as a result.

I thought to share a few insights about our very own Happiness Project.

First, on a quarterly basis we send out a simple anonymous survey that has multiple choice (1 to 5 points) questions around three areas:

1. How happy employees are?
2. How well are we living up to our stated culture as a company?
3. How well we are aligned as an organization (i.e., does everyone know what/why/how we are trying to achieve something)?

The basic idea is that if we truly are what we say (as an organization) and if we are all working towards a well stated goal (that everyone knows and buys into) and if we enjoy the process ,we ought to become very successful! Here are some of the questions we use to try to get into the heart of those issues:

On Organizational alignment:

• I understand company’s top business priorities?
• I believe the company is heading in the right direction?
• My manager keeps me informed about relevant information that helps me do my job?

On Happiness:

• I would recommend Trulia as a place to work?
• Trulia provides the flexibility needed to balance work and personal responsibilities?

I’m sure we’ve got a long way to go to make our system world class, but here (chart) is a good example how we’ve step by step improved our metrics, from 2010 to 2011 while doubling the size of the organization.

Here are a few lessons that I’ve learned along the way:

• It’s easy to create survey fatigue unless the loop is closed with consistent reporting of results back (to all employees) and action plan to improve things. In other words, there’s little incentive to offer feedback, unless participants get something tangible in return.

• Our thoughts about happiness seem to be influenced very heavily by the very recent memories. A raise is the 3 days leading into the survey, will surely spike the results! To get consistent results, it is important to measure results at the same time, e.g., just before quarterly reviews.

• Averages are useful, but I’ve typically found the open ended feedback and business unit (or by function) results most actionable. If, for example, engineers are loving the company every moment, but sales is having tough time,  they both average out the results and you obviously can’t draw many conclusions.

• I’m (still) a huge believer in measuring employee happiness and drawing insights and actionable improvement ideas from the results.

I’d love to hear more ideas and resources how to improve this. Feel free to shoot me an email  sami at trulia (dot) com.

(If you’ve never heard of Trulia where we do all this magic, check out http://www.trulia.com/about for reference)

Through my early experience at McKinsey&Company, I – unfortunately - learned that time is almost an unlimited resource in getting things done: after all, there are 7 nights a week that exist as a flexible resource you can always tap into. But we’ve always been focused on Impact at my company Trulia, not just getting things done or pure output. After having a growing feeling that maybe I’m not maximizing my impact despite lots of to-do’s (and what I thought was a high output), I decided to do what one of my friends, Ron, encouraged to do: take an inventory of time use.

I decided to track my time use by 15minute intervals for three weeks. I used a simple and quick approach:

1. Define three things: (i) activity (e.g., desk work, meeting), (ii) area of impact (e.g., HR, strategy, sales), (iii) specific work (e.g., interview candidate) and (iv) perceived impact on business (1 to 5).
2. Use an always-on spreadsheet to track all activities several times a day to keep an accurate track. This only took <5minutes per day to complete and worked universally across all devices, which was better than some of the time tracking applications I found.

The results were quite interesting and different from what I expected. Most shockingly, of my 70hr and 6-day work week, about 3.5 hours each day went into low-quality meetings and unnecessary emails. So much about being super effective!

Hours worked per day

Turns out my weekly input at work was about 67 hours, typically 6 days a week with Saturday off and a “short” day of about 6-8 hours on Sundays. While this is far from the 100hr weeks, I felt a bit overworked at the end of the period with limited time for “white space” and more creative thinking.

Type of work

The first eye-opener came from the high-level split between type of work: 84% or more than 9 hours of my average 11hr work day went into email and meetings, which seemed like an obvious area of focus to have a bigger impact.

Most shockingly, a deeper look revealed that a total of a quarter (26%) of my total time went into uncategorized email aka “Inbox work”. That’s 17 hours a week cranking email mostly for the sake of email.

I was glad to see that I was spending a good amount of time on other areas that were actually real focus areas for the quarter:

• - Clients, partners, sales, other external relationships: 25% of my time, which was important given my focus on growing revenue top line and cultivating client relationships.
• - Recruiting top talent: (only) 15% of my time. I’ve always considered recruiting as one of the most important roles of a founder (and any senior leader in a company). I thought I was making a huge effort to recruit top talent, but I should spend way more than 15% of it; Clearly an area for improvement for future.
• - Strategy work: 12% of time. This was longer term planning, things beyond current quarter. Seems too low for a founder. Staring at your feet when running fast is the easiest way to stumble, so another area of improvement.
• - Logistics/prep includes travel, driving to meetings and other “hanging out at office kitchen” type of time use, which was only 4%.

It is clear to me that if I want bigger impact, I need to focus on eliminating or getting more out of meetings and email, rather than adding more hours to a 70hr work week.

Re-engineering meetings

When I looked at the actual use of meeting time,  the good news was almost 45% of it was in client/external meetings, close to 20% in recruiting and the average Perceived Impact Measure (1=low, 5=high) was about 4.1, so I wasn’t totally wasting the time. However, 1.1hrs/day was wasted in low impact (<4.0) meetings, or almost half a work day per week.

I’ve decided to implement the following improvements to my meetings in the future:

• - Stop participating in “nice to know” –meetings.
• - Define my personal meeting goals and contribution before a meeting. If impossible, skip.
• - Stop scheduling 1 hour meetings, instead 15, 30, 45minute meetings.
• - Try to do 5 minute “meetings” at the water cooler without scheduling them

Re-engineering email

Email was the biggest time sunk with almost 3.5hrs per day, of which most was uncategorized “stuff”. I’m now focused on getting rid of that by doing the following:

• - Outlook/gmail always offline, only sync it a few times a day
• - Batch process email: flag anything that seems to take more than 15seconds and reply/delete the rest immediately.  Flagged emails I batch process at least once a day (end of day). Keep inbox always empty.
• - More effective use of BCC and stop (or limit) sending unnecessary “ok” , “thanks”
• - Encourage everyone to write short emails with a clear summary and next steps in the beginning.
• - Fast delegation of tasks and responsibilities to one (or more people) to stop huge group emails and chains in the beginning
• - Sort long chains by subject at the end of day: either delete, make a decision or delegate after reading only the last email in the chain

Final thoughts

This exercise confirmed again that the best gains for bigger impact come from more effective use of time rather than adding more hours to work week.

In addition to re-engineering my meetings an email use, I am trying to implement more white space and proactive and long-term planning/thinking to minimize reactive work (aka firedrills). One way of achieving that is to have mini-breaks between meetings and focused work periods. These guys have some good principles that I’ve been using for some times: Energy proj. Of course, sleeping helps too.

Another change I’ve already implemented is better weekly planning: I try to allocate time for the really important things each Sunday, before the start of each week. For example, to be more effective in recruiting, I define specific tasks for each day or week that drive towards that goal.

I plan to repeat the time inventory exercise in the next 6-12 months to keep myself accountable and track progress…

To do very well in – or even just to complete - an Ironman distance triathlon (2.4mile swim, 112mile bike, 26mile run), you would need to do many bike+run brick workouts, complete weekly 2-3 hour long runs, pedal 6+ hour rides and spend at least 20 hours swimining-biking-running  per week? Yes or No?  

I did none of those and had the best triathlon results of my life in 2011, including a sub 9hr finish at the Hawaii Ironman World Championships. An accident, luck or a secret sauce for endurance success? Read below and let me know what you think.

I met a number of ordinary, yet super-fit, individuals during my recent race trip to the Hawaii Ironman World Championships. With a limited sampling, it seems that many of the even first time Ironman triathletes who qualified to the event train 20hrs or more per week and some consistently 25hrs a week. I was not surprised to hear that there was one common denominator between my and most others’ preparation and training: I had done almost the opposite from everyone else. Since I heard the disbelief and question “..and you did what?” so many times, I thought this topic is worth a brief post for others’ benefit.

I won’t have a scientifically or statistically meaningful sample of individuals to draw conclusions from, but I hope that I can make a point by using my own experiences as a data point to extrapolate from. For context, this year 2011 has been so far (and by far) the most successful year for me in the triathlon adventures. For example, I have become the 2011..

• Overall amateur champion at Wildflower Triathlon Long Course
•  Overall amateur champion at Hawaii 70.3. Ironman
• Age group world champion at Ironman 70.3. distance
• Age group world champion runner up at Ironman World Championships (Hawaii) with a sub 9hr finish time

In addition, all objective metrics (such as power  measured by “watts” on the bike, running pace) as well as relative metrics (how I’ve performed against my other competitors) are significantly up from the previous two years. So something is working quite well, while many other things have remained constant over the past three years: my overall health, work load, sleep, nutrition/diet and race body weight.

With a 300+ employee company to run and just more than an hour per day to dedicate to workouts, my training has always been “little, but with great quality” thanks to the amazing principles by Matt Dixon of PurplePatch Fitness.

Here’s what 9 out of 10 triathletes and training tips in most magazines tell me about triathlon training (the conventional wisdom), and especially, how to become a superb athlete at the Half- and Ironman distance.

1. It takes 20+ hours a week to qualify to Hawaii Ironman, and certainly that if you aim for a “top” age group performance. 
The unconvential wisdom: No – it is possible with about 12hrs/week

Since the beginning of the year, I’ve had 2 weeks with more than 15 hours of training, and one of those weeks included three days (3-4hrs/day) of casual bike touring in Finland during a mini-vacation. An average week is around 12hrs, which also includes warm-ups, cool-downs and some commute bike riding.

   Weekly training hours in 2011:

In comparison to my own  previous years, this volume is about the same or 1-2 hrs per week LESS than earlier. 

2. You need a massive, fatigue accumulating 2-4 month build-up and then taper 3 weeks to be really fit for a big long distance triathlon race. 
The unconventional wisdom: No – it is possible to be race ready almost every week without the need for a long taper, by focusing on intense efforts over a day or two and then recover well during each week without training oneself into the ground. 

Even more so than limited training hours, weekly focus on recovery has been the most radical change for my 2011 “season”. Most endurance athletes (from recreational to more serious) seem to purposefully dig a really really deep hole (in terms of fatigue) over several months to barely survive until their taper period, and then start a 2-3 week race-taper and hope for the best to get out of the hole and be fitter and fresher on the race day than when they started the massive build. I’ve seen this ranging from first time marathon runners to experienced triathletes.

In weight lifting this approach would be unheard of:  If you don’t get stronger (lift more weight or more repeats on a given weight) after each workout, you’d immediately change something or have a significant recovery period before next lifting session. It is obvious that if you don’t become stronger, your body is doing the opposite: breaking down due to too much or too frequent weight workouts.
Many endurance athletes tend to think that grinding through workouts day in, day out, no matter how tired, makes them stronger eventually at some point.

The approach we applied to my training this year was that of the weightlifters’. I never trained more than 2-3 days before a good rest day. And if my numbers (pace, speed, watts) weren’t up in the next workout, I took another rest day or two. The principle was simple: I should get stronger and faster every week (or day), not just hoping to get there after a 3-4 months of hard work and a 3 week tapering period.

I started using Restwise (which I’d highly recommend to every athlete) to better quantify my recovery (chart from last 3 months below). If my recovery score was too low, I took a rest/easy day until the scores were up (80-100%).

Similarly, I used quick 10-15min mini-tests in most workouts to see if I was improving; and sometimes stronger efforts like 20min max effort on the bike with maximum power. If the numbers weren’t up from previous time – time to rest.

I was race ready and well rested with 2-4 days of easier workouts throughout the previous 8 months. My final Hawaii Ironman taper was 4 days, after a 50minute all out (395Watts average; which, again, was a personal all-time record) bike time trial race just 5 days before the Ironman race. Similarly, I decided to race a Leadville 100Mile mountainbike race at 11,000feet altitude and prepare with a 48hour “taper”. These were only possible because I didn’t have 3 months of accumulated fatigue to shake off, but fresh and progressively fitter legs throughout.

3. You need to do big brick workouts (bike+run) to be able to run fast off the bike. You also have to train to run on tired legs by doing massive bike rides a day before a long run. 
The unconventional wisdom: No – I’ve had my all-time fastest runs and overall times in triathlon events in 2011 with literally ZERO brick workouts in the last 8 months (except 4 races).

I couldn’t find a single real brick workout from my training log in 2011, except a couple 5 or so minute shake ups and one 10 minutes jog after a bike ride. It certainly takes a few events (or workouts) to get used to the feeling of running after a long and/or hard bike ride, but that feelings will never go away. I still feel crappy for the first 5-15minutes of running off the bike, similar to feeling in my first triathlon 5+ years ago.

Secondly, I’ve avoided doing any major run workouts with tired legs. Running with tired legs and/or bad form is the easiest way to injure oneself. I haven’t found scientific  research that would explain why training with tired legs (=lower power) and bad form (=injury risk) would actually make a better runner even if you have to do that after a bike ride in a race.

Instead, I’ve had all my runs in an almost fully recovered state, which has allowed me to run much faster and stronger each time.

4. To be able to run a strong marathon at the end of an ironman or half-ironman, long runs of 2-3hrs are must, maybe even more. 
The unconventional wisdom: No – most amazingly, I very rarely ran more than 80minutes and only did one 2h run during entire 2011.

As you can see from the chart below, I rarely hit the trails for more than 80 minutes. I did one 2 hour, mainly for self-confidence as I couldn’t believe that I had to run a marathon in 3 weeks but had not done a single run more than 15 miles.
At the same time, I’ve recorded my fastest ever runs on both half-ironman (13.1mile run) and Ironman (26.2 mile run) distances this year.

Instead of logging miles and spending hours running, risking injury and compromising other workouts 3-4 days following a massive run, I’ve focused on a lot of race pace (below/at/over) running. A typical “marathon” workout could be 3-4 times 15minutes where 5 minute sections are below, at, or slightly above expected race pace. That’s a 75-80minute workout and I’m able to recover in 24-48hrs vs. 3-4 days after a massive 20mile+ run.

More often than not, after people fade at the last half of the run in a triathlon, they say they need more and longer long runs. I would guess that the most common reason for fading at the end is just bad pacing in the beginning (of the bike or run), bad nutrition/hydration or simply not enough race pace running in training – and not that the long run wasn’t long enough in training

5. You need to do double sessions, maybe even triples with lunch hour training.
The unconventional wisdom: No – Except one single week in March, I never did more than one workout per day (always morning) in 2011.

I’ve found that is is completely possible to get strong performance gains with a single workout per day and 10-12hrs or so per week. Typically a big week for me looks like this:

• Monday:          Rest (or 30min easy swim)
• Tuesday:         Bike intervals on trainer (60-90min)
• Wednesday:     Run intervals on trails (60-70mins)
• Thursday:        Bike intervals on trainer (60-90mins)
• Friday:            Rest / Swim (40-60mins) / easy run (legs recover)
• Saturday:        Bike “long” (4-5h with no intervals, social time with wife/friends)
• Sunday:          Run “long”  (80-90minutes with intervals and Swim if time)

Accident, luck or a secret sauce?
I realize that I am extrapolating from a single data point, but before you stop spending time with friends and family, reduce nightly sleep to 4hrs and spend all your time logging miles in a quest for improved endurance performance, read this post one more time. It might  help you reinvent the conventional wisdom.

For the business readers: Many of these same principles apply to improving performance at the office. For example, unfortunately hours spent at the desk, lots of “hard work” and conventional wisdom are often associated with great performance, when in fact efficiency, true business impact and unconventional wisdom are the things that actually propel individuals, teams and companies to a greater performance these days. And too often the person who gets the promotion is the one who spends the most hours at the office rather than the one who comes up with a real break-through idea and executes it efficiently. Not very different from the mile counting triathlete who still follows the conventional wisdom and finishes at the bottom of the race results despite most hours in her training log?