Ok, so that’s where the Borat-ism’s end.

Like half of the western business world it seems, I have just read The Lean Startup by Eric Ries. As I read, I make notes (it’s the only way I retain stuff), I thought others might find it useful to see the key points I took from this fantastic book.

I have posted the full list of 100 on my Tumblr. All of them good, but if I had to select 20 they would be these:

1. All processes should be geared to accelerate the feedback loop
2. Find a synthesis between the vision and what customers want – if none exists, change your vision and/or your customers.
3. A start up is a catalyst that transforms ideas into products
4. Steve Blank – ‘get out of the building and start learning’
5. Genchi gembutsu > go and see for yourself
6. When in doubt simplify (MVP)
7. Scale something that works rather than invent something that might work in the future
8. Customers only care if it serves their needs
9. Remove anything that doesn’t contribute to the learning you seek
10. The only way to win is to learn faster than anyone else (stealth is false security)
11. A good design is one that changes customer behavior for the better
12. A solid process lays the ground for a healthy culture, where ideas are evaluated by merit and not by job title
13. Productivity measured by validated learning not in # of new features
14. Only 5% is the big idea, 95% is the gritty work…
15. Pivot > a structured course correction designed to test a new fundamental hypothesis
16. A start-ups runway is the number of pivots it can still make
17. The ability to learn faster is the essential competitive advantage that Start-Ups must possess – inbuilt unfair advantage vs large organisations
18. Startups drown not starve – too many ideas, too many opportunities, not enough focus, not ruthless enough
19. Be tolerant of mistakes first time but never allow the same mistake twice
20. The goal of innovation is to learn what is currently unknown

Related articles

• Eric Ries – ‘pivoting’ immortalised in a New Yorker cartoon (lunatractor.com)
• Steve Blank vs. Steve Jobs (empwaynek.wordpress.com)
• The Lean Startup Frenzy (infoq.com)
• Lean Startup Creator Eric Ries Packs The House at London Event (thenextwomen.com)
• Eric Ries talks about his new book, The Lean Startup (eu.techcrunch.com)

If you’re like most organisations, It’s pretty likely that your data is subject to the well established approach of ‘Extract, Transform, Load’ or ‘ETL’. ETL goes hand in hand with relational databasing (yes, as a verb, like spelunking) and data warehousing.

To cut a long story short (and you can look up the long story yourself) ETL is really heavy on design, because it concerns itself with the content of the data. Known sources, with known content are accessed in known ways. The known content is transformed to a known schema and stored. The data is then used to form known reports. Anywhere where something must be known, you require design.

By our count, there are 5 design points in ETL.

We don’t think this is very agile, or reactive to change. You’ll be fine if your world stands still. But whose world does?

We think that life gets a lot easier and more agile when you ignore content the content itself (bear with me…) and concentrate on the nature of the data you are storing.

What does nature mean? Well, for a start: “Performance data” is values that change over time. In other words: Time-Series. Your sales data is a time-series, your web analytics of Facebook data is a time-series, there are many more.

Time-series is a nature.

Another nature might be 3D special data, or textural content. At Media Science we specialise in time-series. We make sure to maintain granularity of content (by ignoring it – clever huh), and support aggregation, filtering and comparison. We do this with only 2design points. Once at the beginning (we have to find where data exists) and once at the end, where we need to know how users want to ‘access’ their data. In the middle, storing by nature means there is no redesign for any new data. Got a new source you want to throw into the mix? No problem, we’ll hook it up and out the other end it comes.

Related articles

• ETL (Wiki English) (acordocoletivo.org)
• Is your BI development process standardized? (platonblog.net)

We have three great companies speaking:

Simon Campbell will reveal how his company enables clients to generate a return on social media activity through the use of artificial intelligence.

Chris Downs will talk about how companies data is being shared throughout the social space and how they need to get in control of it, and fast.

Giles Comeau will share how they are delivering insights from social media to some of the world’s biggest companies via their proprietary market intelligence engine MeaningMine.

Event timings as follows:

18.30 Event opens (drinks & nibbles provided)

19.00 Speakers (incl Q&A)

20.00 Informal networking & drinks

21.00 Official Social Media Week closing party. Get your tickets here

Londata is brought to you by:

Thanks to the generous support of our sponsor:

About Londata:

We created Londata to provide a space where people intresetd in the commercial opportunities presented by data to come together, listen to a great speaker or two and share war stories. Our aim is to make our attendees and their companies even more successful.

Related articles

• Say hello to Londata… (media-science.com)
• Measuring The ROI Of Social Media (viralblog.com)
• Still Asking ‘Why Social Media?’ ? (mervecam.wordpress.com)
• The DNA of a company: How social media is impacting business models (socialmediaportal.com)
• Social Media Week: Social Media Action Camp #SMWsmac [Event] (digitalbrandmarketing.com)

1. Market validation is all that matters – don’t spend too long predicting what the market _might_ want, get out there and find out what they _actually_ want. It was only once we launched and talked about real rather than planned product that we got real opinion and desire to purchase (or not) expressed.

2. You’re only as good as your next pivot – we though the market wanted self-serve capability (because that’s what they _said_ they wanted). Turns out, when it comes to it, clients wanted delivery of end result, and are prepared to pay the premium that this requires. We started self-serve, pivoted into a managed service. Almost immediately this turned into revenue and now break-even is in sight (the second key milestone). Never stop listening, learning and being prepared to flex when the market speaks (within the context of your overall vision…)

3. Lean rocks – it really is about MVP > release >measure/feedback > iterate. we worship at the altar of Mr Ries. Every time we simplified and released early we became better.

4. Don’t be the sledgehammer to crack the nut – our early meetings saw us landing our ‘change the world vision’ and listing all the clever stuff we do. Cue cross-eyed reactions from potential clients. We were guilty of telling clients everything we could do and hoping something was of interest. The world changed for us when we started asking better questions about their current status and desired solution. We were shocked how _little_ they actually wanted. Make sure you sell what they want (ok, maybe a little bit more) and don’t scare them off with what your capable of. The old sales cliche of acting in proportion with your mouth and ears holds as true as ever (listen twice as much as you talk)

5. Best not design a brand that looks _just_ like an existing major player – sadly we didn’t know this until 24 hours before launch, doh. And we only found out because we put it in front of a bunch of people. Unfortunately we did this in a ‘look at us, aren’t we cool’ way rather than as an early attempt to get valuable input. If only we’d opened up things earlier we would have saved a bunch of time and money. Oh well, new design was ready to go in 3 days and we love it. But still…

6. The world wants your data solution – the journey from unconscious incompetence to conscious incompetence is well underway. Clients need and want help. What they really want is simplicity and for the headache to go away. The distress sell is alive and well…

7. Microsoft really is the start-up’s friend – From day one we were a cloud business and we chose Windows Azure as our environment almost immediately. A decision we have not regretted. As a PaaS (Platform as a Service) offer, Azure is much more than ‘boxes in the cloud’ and hosts _everything_ that we do. It just works and has helped us keep costs tied to the scale of our business. We have had amazing support from both the product and commercial teams in both Redmond and London/Reading, which has directly lead to engineering and commercial benefit. You would be mad not to see how they could help. Bindi Karia is a great first port of call. She’s not hard to find.

8. It’s easy to make your blog low priority…

Keep learning, keep striving to be better. They tell me it’s fun

Richard

Related articles

• It’s official: Windows Azure supports Node.js (gigaom.com)
• Recent Trends in Cloud Services Adoption (clean-clouds.com)
• A CTO’s take on cloud (gigaom.com)
• Cloud 101: What the heck do IaaS, PaaS and SaaS companies do? (venturebeat.com)
• Minimum Viable Products (MVP) (adamdawood.wordpress.com)

Ok release 30 is out!

This week has been all about getting ready for our new app so although users can’t see any changes to SmartView this release’s work has gone to support the launch of our super secret second app!

Till then have fun with SmartView and see you next week.

Each of us is now able to connect remotely from our personal laptop onto our work PC. It doesn’t sound amazing (and is hardly new, granted) but this is my first experience of remote working and, for me, it really is an eye-opener because it not only allows us to stay in to sleep a bit longer and/or not spread our sickly winter viruses, but it also assures us the quality and capacity of our professional computers while comfortably at home. It means that, as long as our computer in the office is switched on, I can connect on my session from anywhere in the world and code the awesome idea I just had while visiting the Acropolis or cooking dinner. The freedom given is massive.

So, of course, you’re going to think ‘What about internal communication?’ and you are right. We thought a lot about it before jumping in. We have our daily scrums, part of the ‘agile’ process, that take place every morning in front of the whiteboard. This is critical in our iteration because it tells everyone who is doing what and who might need whom to help. So how to participate to the scrums from home? After an unsuccessful attempt with a webcam, we gave the iPad a go. It works perfectly well! It really was like being all together (except that I was wearing my penguin socks and had not brushed my hair – now remove that image…).

PP’ing (Pair Programming: two of us sitting through a piece of code to nail and understand it, the collaboration making the work more effective) is also possible with RDP. We both have the same code open in Visual Studio at the same time, so it is just a matter of talking to each other. It being typing on Skype or shouting on the phone, the result is the same in the end.

Having the liberty of occasionally working from home removes non-productive commuting time and, more than that, can be beneficial for those coding tasks where you need to bend your head and jump into it. The quiet and comfort of home can sometimes be a better place to get it done.

So, I will even dare say that RDP is an iFeature: before having it, I didn’t realise I needed it; now I couldn’t go without it.

Now if only we could find a virtual scrum board, we would really fly…

Marie (Penguin socks wearer and passionate RDPist)

Last week we were all over the data and graphing areas to make them look real nice in time for Christmas.

The top developments that we’re excited about this week:

• Arrange your data in folders that you define  (for those of you who like to keep your worksheets neat and tidy).
• Wide graphs!
• Numbers over 10k are nicely displayed in chart axes (10k, 100k, 1m, etc).
• Nice hover action on above-graph icons.

1. We pulled the google analytics plug-in for ux review, we thought it was a bit messy (see picture above).
2. We’ve been working on a Sysomos plug-in which has made it’s way into the back end but not quite to the front end.

Keep an eye out for our next release blog for more info on this!