The Fallacy of One Company, One Egg, One Basket

Single Egg in BasketI got to listen to 20 minutes of Google CEO Larry Page on stage at a 2014 TED conference today. One of the questions he was asked twisted along the way into some views as to why large companies fail. He said the common characteristic – which he’s doing everything to avoid at Google – is that of “missing the future”.

To that end, like a lot of people in the IT industry, I often hear of Microsoft being a case example of this. They generate tons of cash and are still hold stupendously large market shares in their core Windows and Office markets on PCs. They have also grown into a significant share of Enterprise IT based on what started as Windows NT and it’s associated layered products, and are cash generative with X-Box these days.

Bill Gates

That said, while I hold Bill Gates in high esteem (having met him twice and been singularly impressed both times – both in Microsoft’s early years), i’ve never been a Steve Ballmer fan. For me, he epitomised everything that got unseemly with the company. Gates knew his products well, could relate strengths and weaknesses against every main competitive product, and sold his own products based on their strengths. He had a darker, relentless and ruthless side, but that largely sat behind his main product focus.

Steve Ballmer

On the other hand, Ballmer was far more obsessed by financial performance and ratcheting up market share, primarily by tripping up competitors by all means possible. With that came a myriad of third party shills and supposed industry groups, campaigning hard with government bodies to impose sanctions on competitors, and to abuse the processes of standards bodies. They have strong armed vendors using Android into “licensing” demands with menaces, alleging use of their patents in the Linux code that is part of Android (but cannot relate in public which ones they are – and in test cases, have been shown to be possessing very weak and questionable claims). Then advertising campaigns under the banner of “Get the Facts” and more recently “don’t get Scroogled”, both with a reputation for poor accuracy and bringing Microsofts own brand reputation into disrepute, with no meaningful positive effect. Gone were all the pushes based on the strengths of Microsofts own offerings; instead just trying to trip everyone else up.

In the end, Ballmer got ousted by the board, albeit only after deciding to buy Nokia, electing to pursue a “Devices and Services” strategy, and to restructure the whole operation into one whole – removing the previous Business Unit, devolved P&L structure previously in place.

Strategy – Whoops!

A “Devices and Services” strategy is a particularly goody. In the industry, Apple follows a Devices model, where they run the entire value chain vertically – from basic hardware all the way up to services layered on top. On the other hand, Google run services that run horizontally across everyones devices, making profit from monetising customer purchase intents. So, a devices company try to keep all the added value around their own device from which they derive their money, whereas a services company needs their touch points to go across all devices that have market share in the industry. The two strategies are incompatible; i’ve yet to see the first example of a vertically integrated, horizontally integrated company exist.

With that “strategy” put in place, and all the previous warring Business Units now running under a centrally planned and executed structure, Ballmer did an exit stage left.

There is a case in recent history of a large high tech company that did an identically structured pivot attempt, and demonstrates one end game of such a “one company, one egg, one basket” strategy. I fear that this latest restructure puts Microsoft in grave danger of following the same example.

The Death of Digital Equipment Corporation (DEC)

I used to work for DEC, where 1957-1982 there were many customer industry focused product lines. Things like Engineering Systems Group, Graphic Arts (that’s Newspapers/Publishing), MDC (Manufacturing, Distribution & Control), Education, Commercial OEM, Technical OEM, Laboratory Data Products, etc. Some 130 of them, all pulling on the output of Central Engineering and Manufacturing, then applying their industry knowledge to ruthlessly deliver excellent value to customers in their target industries. Each product line paid for the sales presence they needed down to office level worldwide.

Sea Change

By 1982, there were some concerns in the Exec Committee that there was too much “who knows who” horse trading of capacity in manufacturing, so they elected to move to the same sort of model that Microsoft has just moved to. That of “one company, one strategy, one architecture” model, orientated around the strategic core asset – VAX (in the case of Microsoft this time around, it is Windows).

Aside from a financial bath in the very first quarter in 1983, where all the cracks in the processes that the Product line structure papered over got exposed, the company really mushroomed in sales performance. Unfortunately by 1987, some specific members of the Exec Committee started citing a date when they’d outgrow IBM in size (who were 7x DECs size at the time). DEC missed virtually every strategic product transition at the low end. Management hyped up attacking the high end with very expensive ECL chips, while the labs in Hudson had $300 CMOS VAX chips ready to go. The very impressive Prism/Mica chips (30x VAX 11/780) taped out. But an exec committee that kept on flip-flopping between 32 bit and 64 bit edicts into engineering, then losing patience with the delays they’d caused – and killing the very projects that were the foundation of the next generation CPU technology. Key staff left.

It took a further 5 years to get Prism/Mica out – adding a couple of instructions, fixing on a 64 bit address space, and calling it Alpha AXP. By then, the rot had set in. That Silicon was the fastest CPU in the industry by nearly 2x for getting on 10 years, but the company above it faltered. Ken Olsen (co-founder) got deposed, and Bob Palmer (new CEO) elected to move back to a different divisional model.

So, he had PC, Storage, Components & Peripherals to play in the now horizontally integrated markets, and a few vertical industry ones (Financial Services, Consumer Process Manufacturing, Discrete Manufacturing & Defence, Healthcare, Telco+Media+Transport). However, he also got the salesforce commissioned (something against the traditions of the company), and introduced a “two quarters missed financials and you’re out” ethos in the divisional heads.

These final moves were Digitals death knell. You either had a lottery on who didn’t get fired, handing extra responsibility to those leaders who happened to string together two quarters of good sales (largely by seasonal industry trends rather than own performance). Or the panic stricken ones would stuff their distribution channels with stock, hoping that would delay the inevitable for at least one more quarter.

Following a few Bob Palmer legal stunts with Intel and with Microsoft, the company was sold to Compaq, having accumulated losses in Bob Palmers 5 year reign equivalent to the total profits of the company in 35 years under Olsen. Given the above, I think Microsoft need to be extra vigilant not to go the same way.

Microsoft now have that same type of structure

The new organisation structure will give Microsoft every opportunity to follow in DECs footsteps if they are not extra careful. The key litmus test will come when the company needs to move from Windows to whatever platform needs to come next, and I currently see no seeds planted of what that will be.

To that end, there are some early green shoots, and some dark clouds. Mr Scroogled has been promoted to “Head of Strategy”, which I can only imagine is to leverage his Democratic political connections for the future; I can imagine no other background to suggest that any company wide core strategy is safe in his hands. On the good side, they have allowed Nokia to produce a low end Android based phone handset, and look to have decided to finally release Office on the iPad formally. Outside of that, no real news at all. Many commentators wax lyrical on the lack of strategy for the future they are aiming at, or where they have anything that’s meaningful in the now mobile and tablet growth markets, and indeed in public cloud servers where the economics (and share) sit resolutely with Linux.

Capitalism vs Communism – Corporate Editions

Meanwhile, the other side of my brain says that centrally planned monoliths have an over dependency on the cleverness at the centre, while the chaos that is capitalism distributes the risk. Microsoft is such a monolith, and i’m not sure how deep that central cleverness is.

Down history, a chaotic network will always defeat a hierarchy. Just ask Gorbachev.

Ask not what your mobile phone can do for you, …

John F Kennedy Photograph (JFK)

Last nights Gillmor Gang felt like it arrived at a conclusion that the next big frontier for mobile platforms was the message bus that is notifications. From a consumer perspective, Google are good at this, albeit Google Now and Google Plus tread over each other occasionally, and Google Plus’s Circles quickly fall into disrepute. Apple’s notification system is mostly empty and unused. It was perceived that Microsoft didn’t have a strategy at all. Meanwhile, the messaging vendors running across multiple platforms are lined up for a battle royal to keep their respective user bases growing, and applicable in their niche use contexts (WhatsApp, Line, WeChat, Hangouts, Skype, Linc, Secret, Snapchat, Chatter, Twitter, LinkedIn, etc).

For me, interesting and pertinent comment tends to come from Feedly (mainly RSS feeds), my DoggCatcher Podcast consumption, a couple of mailing lists and the occasional post on Twitter, Facebook, Google+ and very rarely, in LinkedIn. For the most part, all of these social apps shift ungodly amounts of pollution in my stream, and are systematically getting worse. It really doesn’t surprise me that Twitter have had over 1 billion registrations, of whom only 1/4 are regular users now; the daily requests to add more suggested users does nothing for my feed quality – and in fact precisely the opposite.

My Nexus 5 with Google Now already flashes up a bus timetable and next bus eta when I walk past a bus stop. It has all the performance of the stocks I own already tabulated. It tells me the result of Aston Villa’s last match (1-0 against Chelsea – must be a bug there somewhere) and soon will tell me the next match due, along with the relevant league stats of both teams. And at the moment, it will throw in the name of someone from Google+ whose Birthday is today, although i’ve never heard of any of the folks listed in the 5 months since i’ve switched to Android. And if I have worked out how to integrate my calendar, it will tell me if I need to leave early for my next meeting in light of the current traffic conditions.

With that, most of my future use cases that help *me* are largely covered. Improving the efficiency of me recording my food intake and exercise routines may help; i’ve logged all my food intake and have it summarised as carbs, protein, fat, calories and exercise calories expended by day, every day since June 3rd 2002. My weekly weight readings go back that far too. My fitbit does a reasonable job counting my steps and I get a £5 book voucher to spend every 4 months or so for the privilege of admitting my exercise stats. So, an Apple iWatch with heart rate/pressure monitoring may add a bit more data meat for me to have graphed. So, what’s next to help… me?

The industry is now off the starting blocks and into the calls of “Big Data”, “Internet of Things”, “Sensors everywhere”. My phone already knows the time, my location, who i’m calling, who’s calling me, how fast i’m travelling, where i’m headed (be it in my calendar or set as my navigation destination) and where I have notification of tracking data for an inbound package from Amazon. Some data based on data clues i’ve shared with Google (location, searches, Chromecast media consumption) and Amazon (purchases). I wonder if any Visa/Mastercard data makes it back. And now that the role of “information hub” has escaped from living room Games Consoles and into that Smartphone into my pocket, what value to I get back from it now?

A lot of the benefits are going to accrue higher up the food chain – in which case Steve Gillmor’s words may (as usual) be prescient.

One of my previous employers had over 10,000 staff, thousands of suppliers and a large number of B2B customers. One system there collected the metadata from email on who was conversing with who; anyone could go onto the system and see (in priority order) who was engaged with a specific supplier, or all the touch points into large enterprises they serviced. That speeded up the engagements (as it would do in any knowledge based business). That may also work for phone calls made or received on the company mobile in the future.

The same company also have high water marks in various business processes, so if an iceberg is heading your way that will break the customers SLAs, the management chain get the needed urgency and corrective actions instilled – before the customers notice. However, it is silo’d on specific tracking systems that managers have to dip into regularly.

For an Enterprise, one of the keys is to be able to link business processes and the exception handling flows so that the relevant people know whatever is important to them, when it is important to them. Some of my previous work was to graph important things simply to show, for example, what the flow of incoming cash was, it’s sources and any queries that may impale the chances of a customer paying their invoice(s) on time. Very much like the sort of card dished out by Google Now, but with some limited interactivity to dig down deeper into a prioritised list – to enable fast spotting of the root cause to address. It worked spectacularly well to help eradicate potential problems and to markedly improve DSO.

(For what it’s worth, once I could reach the database tables I needed, I prototyped the reporting needed to address the business issues very quickly in Tableau Desktop Professional. Then in line with corporate reporting platform decisions, self learnt then reimplemented the whole lot in Microsoft SQL Services Reporting Services (aka SSRS) – a very bitty, detailed and long process – where the reports still run to this day).

Some time ago, Facebook provided an alternative UI that made your friends the centre of your mobile experience. This largely fell into disrepute as many of the apps on a phone are gateways into simple process tasks, and the entry point wasn’t specific to a designated “friend”. John Borthwick wrote a piece on Medium about which Mobile apps appeared on a wide variety of home screens. Yahoo bought startup Aviate who provide a launcher that moves icons to the home screen – for immediate availability – based on the context of where you are and what you do regularly. I’m yet to see any analysis that segments which apps are used, when and how often; that would be a useful base to ask further questions.

In the meantime, linking apps into appropriate notifications from Enterprise systems may well be a useful thing for mobile applications. That historically has been the domain of Microsoft applications with custom extensions written in VBA (Visual Basic for Applications). It’s probably a sign of genius that you can do likewise with Google Apps now (Chrome Extensions were announced last week) – with add-on code written JavaScript – the most popular programming language in the world.

The main downside is that, for a business process, JavaScript (as indeed is true of VBA) is akin to writing stuff in very basic assembler. Mind bogglingly long winded and subject to excruciating minute detail. I think there’s probably a lot of mileage in being able to provide Google Now type cards with graphs and data you can drill into out of the box – all thrown into the notifications stream with an interface not unlike IFTTT (If This then That – one of John Borthwicks companies) to deliver the information to the correct people, at the right time, only.

I’m just waiting for the first signs that the Enterprise Software vendors will start putting the hooks in to enable Google to undertake the assault on this hitherto Microsoft stronghold using Chrome Extensions.

In the meantime, I also ask myself how folks like SAP and Oracle survive with their very clunky ERP software, all of which looks ripe for disruption with modern open source based software – but that’s another story about money, customisation and organisational inertia all by itself.


Did you know 2.8% of your customers are dead?

Gravestone saying "Rest in Peace"

Those are the exact words I mentioned to the CEO of a B2C company – where customers were paying monthly subscription fees. Unfortunately, his immediate question back was “Okay, but how many of those are on Direct Debit?”. I think my reaction to his interest in the number of people paying for his service, but not using it, was one where I thought he’d do damage to his brand than earn him extra profits.

It’s long been an accepted view that Marketing in IT circles is often considered a set of “hail Mary” throws to attract new potential customers, with little of the precision of folks who have that title in Fast Moving Consumer Goods (FMCG) companies. I’ve sat in meetings with a roomful of IT reseller “Marketing” folks to find I was the only company present doing systematic testing to find out what works, what didn’t and to use this learning to continuously improve. More a case of “getting the letter out” and losing the ability to learn anything; surely much better to send two different wordings out, to see which one pulled better – at the very least.

I’m reminded of one person I met who worked in the field for a Chocolate vendor, and like all his industry colleagues, could relate to 1/10 of one percent shifts in his market share in the retail outlets he supplied through. He was expected to have an action plan in place if anything slipped a little, or to do more of anything that slightly increased his share. One day, fed up selling Chocolate bars, he decided to move to a company selling software to large IBM computer installations.

He walked in to see his new boss, and made the fundamental mistake of asking what his software products market share was in IBM Mainframe installations working in the Finance Industry and in the counties of Avon and Somerset. His new boss looked at him as if he’d arrived from the Planet Zog, and told him just to get on the road and sell something. He ended up thinking this was dumb, and set up his own company to fix the gap.

He elected to start sending out questionnaires to all the large IBM customer sites in the UK (there were, at the time, some 1,000-1,500 of them), getting a telesales team to help profile each site, and to reflect the use of hardware and software products in each. Then sent out a quarterly summary, segmented by industry, of what everyones peers were using – so the survey participants saw value in knowing what their peers in like organisations were doing. He subsequently extended  the scope to cover other vendors, and gradually picked up a thorough profile of some 30,000 installations, covering over 80% of Enterprise IT spend in the country.

At that point, he had an ever-evolving database of all mix of hardware and software in each, coupled with all the senior decision makers details, and even the names of IT projects both planned and underway in each. The last time I had a meeting with him, he could aim me into the best 5-10 prospects for my IT products and services that aligned with what my ideal customer would look like (in terms of associated prerequisite products they were already running) – with a single rifle shot – allowing sales focus and without spreading sales effort over many unproductive lead follow-ups. Marketing (and Sales) Gold. Expensive to use, by worth every penny.

He subsequently sold his company and the database to Ziff Davis, then to Harte Hanks – the same folks who compile the list of dead people (from published UK Death Certificates) that was part of the profiling exercise I undertook and that I mentioned above. They then sold the same Database assets and it’s regular surveys to the company where it resides today.

Apart from that data, there are a number of other useful sources you can draw on. I once managed to persuade MySQL (before Sun, long before Oracle, ownership) to get their customers profiled, just by relating postcodes we deduced from sampling contact addresses and/or the same from location information on their web sites. It turned out that 26% of their base existed in System Integrators, Web Development and Software companies, while the remaining 74% was flat as a pancake over 300 other SIC codes. Very difficult to target as a whole unless you had the full list of customers – which only they did! There are also various mailing lists, MeetUps, forums and resources like GitHub where you can get a view of where specific developer skills are active.

All very basic compared to Consumer Marketing, where armed with a name, a date of birth and/or a postcode, you can deduce a pretty compelling picture of what your B2C customer looks like, family make-up, what they read and their relative wealth. When I was at Demon Internet (first UK Internet Service Provider), we could even spot one segment of very heavy users that, back in 1997, turned out to be 16-19 year olds living in crowded accommodation, playing online games and with no parental supervision of the associated phone costs. We also had the benefit of one external consultant who was adept at summarising 550 pages of BMRB Internet Survey number tables, producing an actionable and succinct 3-5 pages of A4 trends to ride on.

Today, with even the most expensive mobile smartphones starting to commoditise – and vendors looking to emphasize even the smallest differentiation now – I wasn’t too surprised that Samsung in the USA have today landed an ex-VP of Proctor & Gamble to head their Marketing Efforts going forward.

With that, the IT industry has now come full circle – and FMCG class Marketing skills will start to become ever more important in our midst.


Crossing the Chasm – on one page of A4

Crossing the Chasm Thumbnail

Among my many Business books, I thought the insights in Geoffrey Moores Book “Crossing the Chasm” were brilliant – and useful for helping grow some of the product businesses i’ve run. The only gotcha is that I found myself keeping on cross referencing different parts of the book when trying to build a go-to-market plan for DEC Alpha AXP Servers (my first use of his work) back in the mid-1990’s – the time I worked for one of Digital’s Distributors.

So, suitably bored when my wife was watching J.R.Ewing being mischievous in the first UK run of “Dallas” on TV, I sat on the living room floor and penned this one page summary of the books major points. We grew the DEC business from £12m to £52m in 2 years after using this work, so a lot of the lessons appeared to be good ones. Just click it to download the PDF with my compliments. Or watch him describe the model in under 14 minutes at the recent O’Reilly Strata Conference here. Or alternatively, go buy the latest edition of his book:

I did use this as part of a project to increase sales of Trafficmaster SmartNav jam-avoiding in-car navigation systems, which I conducted with the assistance of two professors from Cranfield Business School. We picked three target use cases to take us over the Chasm, with varied success.

I did manage to get 300 of them installed in the complete Virgin Atlantic “Upper Class” Chauffeur Car Fleet, which avoided the normal driver practice of having to wake the passenger up with an abrupt piece of braking as they neared the customers destination. Also saved a phone around to find out who was closest to a customer who just called in (found that the McDonalds outlets around Airports was a favourite haunt of their drivers).

By hook and by crook, we also managed to get test units in the cars of board members of 7 of the top 10 System Integrators in the UK at the time (given they had hundreds of expensive staff on the road every day; thought keeping them out of traffic jams would help their bottom line). Incorrect thought as it turned out; staff competitions to win a one of ten SmartNavs in each company (to promote them down the line, hoping for bottom up demand) invariably ended up with a glut of them on eBay.

Trying to engineer a bridge over a chasm proved difficult when the product is already designed and the B2B use cases wide ranging, especially when sales appeared to be over a myriad of different small segments. The company subsequently went B2C, and I suspect directly into the headlights of the steam roller that TomTom became.

My PA redrew my hand-drawn sheet of A4 into the Microsoft Publisher document that output the one page PDF. I may have the source file somewhere, so if you want a copy of the source file, please let me know – drop a request to: [email protected].

“Big Data” is really (not so big) Data-based story telling

Aircraft Cockpit

I’m me. My key skill is splicing together data from disparate sources into a compelling, graphical and actionable story that prioritises the way(s) to improve a business. When can I start? Eh, Hello, is anyone there??

One characteristic of the IT industry is its penchants for picking snappy sounding themes, usually illustrative of a future perceived need that their customers may wish to aspire to. And to keep buying stuff toward that destination. Two of these terms de rigueur at the moment are “Big Data” and “Analytics”. There are attached to many (vendor) job adverts and (vendor) materials, though many searching for the first green shoots of demand for most commercial organisations. Or at least a leap of faith that their technology will smooth the path to a future quantifiable outcome.

I’m sure there will be applications aplenty in the future. There are plenty of use cases where sensors will start dribbling out what becomes a tidal wave of raw information, be it on you personally, in your mobile handset, in lower energy bluetooth beacons, and indeed plugged into the “On Board Diagnostics Bus” in your car. And aggregated up from there. Or in the rare case that the company has enough data locked down in one place to get some useful insights already, and has the IT hardware to crack the nut.

I often see desired needs for “Hadoop”, but know of few companies who have the hardware to run it, let alone the Java software smarts to MapReduce anything effectively on a business problem with it. If you do press a vendor, you often end up with a use case for “Twitter sentiment analysis” (which, for most B2B and B2C companies, is a small single digit percentage of their customers), or of consolidating and analysing machine generated log files (which is what Splunk does, out of the box).

Historically, the real problem is data sitting in silos and an inability (for a largely non-IT literate user) to do efficient cross tabulations to eek a useful story out. Where they can, the normal result is locking in on a small number of priorities to make a fundamental difference to a business. Fortunately for me, that’s a thread that runs through a lot of the work i’ve done down the years. Usually in an environment where all hell is breaking loose, where everyone is working long hours, and high priority CEO or Customer initiated “fire drill” interruptions are legion. Excel, Text, SQLserver, MySQL or MongoDB resident data – no problem here. A few samples, mostly done using Tableau Desktop Professional:

  1. Mixing a years worth of Complex Quotes data with a Customer Sales database. Finding that one Sales Region was consuming 60% of the teams Cisco Configuration resources, while at the same time selling 10% of the associated products. Digging deeper, finding that one customer was routinely asking our experts to configure their needs, but their purchasing department buying all the products elsewhere. The Account Manager duly equipped to have a discussion and initiate corrective actions. Whichever way that went, we made more money and/or better efficiency.
  2. Joining data from Sales Transactions and from Accounts Receivable Query logs, producing daily updated graphs on Daily Sales Outstanding (DSO) debt for each sales region, by customer, by vendor product, and by invoices in priority order. The target was to reduce DSO from over 60 days to 30; each Internal Sales Manager had the data at their fingertips to prioritise their daily actions for maximum reduction – and to know when key potential icebergs were floating towards key due dates. Along the way, we also identified one customer who had instituted a policy of querying every single invoice, raising our cost to serve and extending DSO artificially. Again, Account Manager equipped to address this.
  3. I was given the Microsoft Business to manage at Metrologie, where we were transacting £1 million per month, not growing, but with 60% of the business through one retail customer, and overall margins of 1%. There are two key things you do in a price war (as learnt when i’d done John Winkler Pricing Strategy Training back in 1992), which need a quick run around customer and per product analyses. Having instituted staff licensing training, we made the appropriate adjustments to our go-to-market based on the Winkler work. Within four months, we were trading at £5 million/month and at the same time, doubled gross margins, without any growth from that largest customer.
  4. In several instances that demonstrated 7/8-figure Software revenue and profit growth, using a model to identify what the key challenges (or reasons for exceptional performance) were in the business. Every product and subscription business has four key components that, mapped over time, expose what is working and what is an area where corrections are needed. You then have the tools to ask the right questions, assign the right priorities and to ensure that the business delivers its objectives. This has worked from my time in DECdirect (0-$100m in 18 months), in Computacenter’s Software Business Units growth from £80-£250m in 3 years, and when asked to manage a team of 4, working with products from 1,072 different vendors (and delivering our profit goals consistently every quarter). In the latter case, our market share in our largest vendor of the 1,072 went from 7% UK share to 21% in 2 years, winning their Worldwide Solution Provider of the Year Award.
  5. Correlating Subscription Data at Demon against the list of people we’d sent Internet trial CDs to, per advertisement. Having found that the inbound phone people were randomly picking the first “this is where I saw the advert” choice on their logging system, we started using different 0800 numbers for each advert placement, and took the readings off the switch instead. Given that, we could track customer acquisition cost per publication, and spot trends; one was that ads in “The Sun” gave nominal low acquisition costs per customer up front, but were very high churn within 3 months. By regularly looking at this data – and feeding results to our external media buyers weekly to help their price negotiations – we managed to keep per retained customer landing costs at £30 each, versus £180 for our main competitor at the time.

I have many other examples. Mostly simple, and not in the same league as Hans Rosling or Edward Tufte examples i’ve seen. That said, the analysis and graphing was largely done out of hours during days filled with more customer focussed and internal management actions – to ensure our customer experience was as simple/consistent as possible, that the personal aspirations of the team members are fulfilled, and that we deliver all our revenue and profit objectives. I’m good at that stuff, too (ask any previous employer or employee).

With that, i’m off writing some Python code to extract some data ready ahead of my Google “Making Sense of Data” course next week. That to extend my 5 years of Tableau Desktop experience with use of some excellent looking Google hosted tools. And to agonise how to get to someone who’ll employ me to help them, without HR dissing my chances of interview airtime for my lack of practical Hadoop or MapR experience.

The related Business and People Management Smarts don’t appear to get onto most “Requirements” sheet. Yet. A savvy Manager is all I need air time with…

a16z brilliance vs the Leaking Bucket

Digital Ocean Logo

When I worked for DEC, I used to have a brass plaque on the wall in front of me that reminded us in the Software Services Division of our three priorities. It said, in order of importance:

  1. Warranty Customers
  2. Presales
  3. Consultancy

Paraphrased, this says: look after your customers before you go anywhere near trying to get new ones. Next, support the rest of the company selling the whole toolbox to solve customer needs (software was typically only 10% of a project sale). Finally, if we’d done those first, only then did we try to make profit for our own unit alone.

The other euphemism we knew was an old American Football one, which describes a now illegal play; that of “smacking the helmet”. That’s the crash helmet of a 30 stone athlete running at you, as anywhere his head ends up heading, the full weight of the body will follow. So, a well aimed deflection early in a move causes a disportionate effect when the rest of what’s behind it follows. And in the IT market, that front end constituency is the software development community – aka “Developers”. Hit that effectively, and you’re in great shape.

In theory, it’s a great time to be a software developer. Hardware, Storage and Network capacity is fairly inexpensive. Tools to build everything from Mobile to Enterprise applications are predominantly open source and available to all. So, a lot of the early decision making for where to site your applications is where you find a cost effective on-ramp – and more often than not, you’ll stick to where you first deploy as your business scales.

When you are a developer, you get to hear about Amazon Web Services (AWS) and their fantastic growth. This a result of their CEO Jeff Bezos telling his staff that they would deploy all their businesses as APIs, and allow other companies to use their spare compute/storage capacity. Spikes in demand necessitate massive over “just in case” provisioning, even though those spikes are few every year and very seasonal. That said, the amount of options on there is now wide and complex, and hence a learning curve before you can price your development hosting cost out. Examples here, but for my needs, it would be circa £80/month.

You also get to hear about Google Compute Engine, which open up Google’s capacity to developers who can write to their own specific APIs; that said, they appear to favour apps that can take advantage of their own unique database and auto scaling features. If you want a price, then there is a web site where you can enter a number of parameters, and it will articulate a dollar cost – which in my case, were not inexpensive. Or you can have a crack at this.

Likewise for Rackspace, who do a lot of work among the startup community, but again who have pricing suited to paying for their excellent support services. Most of which developers don’t actually need while starting to build their systems. Examples here.

Early in my own work, I saw a report from Mike Prettejohn‘s company (Netcraft) about a small New York company called Digital Ocean who were growing like topsy. From 137 machines in Dec 2012 to (at the time of writing this) 54,142 in Feb 2014:

Digital Ocean Server Growth

The main appeal to me (like a lot of developers) is that you can provision a server instance with one of a range of prebuilt Linux configs within 5 minutes. And once it’s up, it’s $10/month for a virtual server instance with 30GB of Flash Storage and 3TB of Network Bandwidth per month. Add a further $2/month to get weekly backups and the ability to take as many snapshots of your system(s) as you feel comfortable. Very simple, predictable and does the job. The words you’re reading here are being served off a Ubuntu Linux Server in Digital Ocean Amsterdam, using a WordPress network I built using one of their available images. DIY, not for everyone, but if you know what you’re doing and you can secure your site, it’s about as cost effective as you can get.

Besides seeing the volume of site growth, I look at the Netcraft Hosting Provider Switching Analysis, which gives an indication of how each provider was both growing or churning its customer base – and if there was churn, where it was going. The thing that struck me were the number of sites that were relocating from AWS and in particular Rackspace over to Digital Ocean. At a time when Rackspace have been giving profit warnings, the numbers were over 4 figures of customer sites per month – some 31,279 sites in a year.

Mention Digital Ocean to Rackspace staff (I know two of them), and the best positioning I have from them is that they have many competitors that keep them up at night. That said, I shake my head and wonder if they’re spending all their time looking at new customer acquisition (bath taps at full bore) while leaving that very large plug out of their bath.

With that, Andreessen Horowitz yesterday put some major VC funding into Digital Ocean. Given they are gobbling market share – and that my gut says it’s heavily developer focussed – I think they are a fantastic bet. I wonder when AWS, Azure and Google will have a comparable offer, and until then, i’m sure progress will continue on the current relentless path. Digital Ocean have been a revelation to me so far.

Well done, a16z. You’ve picked a great team. Again.

Bill Gates, Compaq Plus and some new thing called Windows

Apple Lisa MouseMicrosoft MouseVisiCorp VisiOn Mouse

Back in 1983, I worked in Digital’s UK PC Dealer Team, where I was the sole presales technical guy helping to grow Rainbow PC sales through the PC Dealer Channel. Around 120 independent dealers, many of whom had a background in selling Commodore, Apple and miscellaneous CP/M based machines to consumers and businesses. As the second largest computer manufacturer in the world, everyone (including ourselves) expected the PC market to become an IBM vs DEC battle ground.

We had seen the launch of the Apple Lisa, a machine that scared everybody. While most of the vendors saw the windowing system and thought Apple would eat us alive, consumers got equally scared of the $10,000 price tag. However, it set in train an arms race to provide an equivalent for other PC vendors.

The authors of Visicalc (the first and most popular spreadsheet) started engineering a Windowing system called “Visi-On”, another called Quarterdeck a system called “DesQ” that could work out of the box with existing applications, and there was a rumoured response on the way from Microsoft.

In May, we had a visit from Phil Sutcliffe and his US CEO, Bill Gates, who carried in a Compaq Plus (IBM compatible as large and heavy as a portable sewing machine) and set it up to give a demo to around 20 of us around a conference room table. Two of the DEC VIPs were stuck in a board meeting upstairs, so we all sat around the table like lemons, waiting for their arrival, Gates included. I couldn’t help myself, so I turned to him and said: “I notice you have two buttons on your Mouse there. The Apple Lisa has one, and the Visi-On mouse has three. I’m curious, why did Microsoft pick two?”.

With that, he spent a good ten minutes relating a thorough drains up of his thought process, which included many examples of areas that really sucked when using both the Lisa and in VisiOn. Extremely thorough, well thought through, and left an impression of “Wow”. The sort you walk away with if you ever meet someone who turns into a walking encyclopedia.

When the VIPs arrived, the first thing he said as he shook their hand was “When are you going to drop CP/M and move to DOS?”. A reference to Digitals then preferred OS, given it already ran thousands of applications and was an on-ramp to Concurrent CP/M, which allowed you to hop and skip between 5 running full-screen applications. That done, he then gave a demo of a new product called “Windows” that Microsoft were at that point building.

He left with some degree of frustration at not persuading the senior folks to switch immediately. Phil told me afterwards that when he got back to David Fraser, then Microsoft UK General Manager, he told him “There was only one guy in that room who knew what he was talking about – hire him”. At the time, there were few UK employees – about 10 or so if I recall. I was duly invited for interview, spoke to David Fraser and International VP Scott Oki, but ended up declining the move.

The one thing that’s always struck me ever since is how asking a good question often has a much bigger impact than knowing the answers. It’s usually a sign of good management if staff are aimed at audacious goals, and questioned about detail rather than having it prescribed to them.

About 4 weeks in after I was first handed DECdirect Software to start, I got invited to a chat with Peter Herke – the General Manager of the DECdirect Catalogue operation at that point. He had a reputation of chewing out senior people who didn’t know every detail of the business they are running (in fact, for junior people, he just asks a few pointed questions and asked them to come back with the answers when they were to hand – people learnt to keep the finger on the pulse very quickly).

I spent the whole evening before thoroughly remembering every statistic and every detail of the emergent business, from helicopter view to the smallest thing. When I got to his office, I sat down, he closed the door and just said “Are you enjoying it so far?”. That completely threw me. I recall saying “Well, it feels like i’m sitting in an aircraft cockpit, and I can see all the dials moving. I’m at the stage of watching them all, trying to work out which are the important ones”. I then took him through what I was doing, what the challenges were and how I was addressing them. At the end, he asked when I was expecting to launch to the outside world, and told me that it was my call, and not to do it until it felt right. To me. He just wanted me to have $35 million revenue in the bag 11 months later, and apart from that, it was my ship to pilot. And to ask him for help and advice if I needed it along the way.

Big team effort (8 telesellers I shared, 2 tech support, 1 logistics person and me), but we launched 3 months later, flew past the $35 million target well ahead of fiscal end of year, and in fact hit $100m within 18 months – at over 89% gross margin.

Just goes to show what people can achieve if given the latitude to grow, and just having good questions asked of them to help them steer themselves along the way. I’ve treated every employee i’ve had since like that – and have been proud of the results every time.

Enterprise IT meets the Hall of Marbles


Every time I hear a vendor utter words such as “private cloud” or “hybrid cloud”, I see visions of brakes being applied. Usually by vendors or IT departments, who sell (or buy) the concept that they can evolve gradually into the future as prices plummet.

I had my first taste of a likely future state a few years back. The London Organising Committee of the Olympic Games needed to provision 500,000 email accounts for a 2 year period, lasting 12 months before and 12 months after the 2012 games. I recall that several prospective suppliers configured branded hardware, software and management using their own hosting plus services, and came out in pricing around £7-10/user/year. Using their existing scale, Google came in at 14p/user/year.

While a prestigious win, I could find no-one that thought Google were selling at a loss. So, what happens when those sort of scale advantages hit Enterprise IT? Do brand vendors not see the tidal wave of low cost, immediately available computing and storage coming their way, a tidal wave due to hit this side of 2017? I think Simon Wardley describes the reality very eloquently in one of his blog posts, relating the story of “The Hall of Marbles”:

It’s the exponential growth part that catches most past suppliers out and that’s due to this expectation of gradual change due to the previous competitive stage (i.e. product vs product). To explain this, I’ll use an analogy from a good friend of mine, Tony Fish.

Consider a big hall that can contain a million marbles. If we start with one marble and double the number of marbles each second, then the entire hall will be filled in 20 seconds. At 19 seconds, the hall will be half full. At 15 seconds only 3% of the hall, a small corner will be full. Despite 15 seconds having passed, only a small corner of the hall is full and we could be forgiven for thinking we have plenty more time to go, certainly vastly more than the fifteen seconds it has taken to fill the small corner. We haven’t. We’ve got five seconds.

Hence for a hardware manufacturer who has sold computer products and experienced gradual change for thirty years, it is understandable how they might consider this change to utility services will also happen slowly. They will have huge inertia to the change because of past success, they may view it as just an economic blip due to a recession and their customers will often try to reinforce the past asking for more “enterprise” like services. Worst of all, they will believe they have time to transition, to help customers gradually change, to spend the years building and planning new services and to migrate the organization over to the new models.

Alas, Amazon alone is estimated at $2bn in cloud revenue for 2012 and predicted for almost $4bn in 2013. If that growth rate continues then by 2016 they will be in excess of $30 billion in revenue. They also have rapidly growing competitors such as Google.

The cold hard reality that many existing suppliers probably don’t comprehend is that the battle will be over in three to four years and for many the time to act has already passed. Like the rapid change in climate temperature in Greenland, our past experience of change does not necessarily represent the future.

In industry, we have a long history of such rapid cycles of change and inertia is key to this. These cycles we call “revolutions” as in industrial, mechanical and the revolution of electricity. During these times, change is rapid not gradual and disruption is widespread.

I’m typing this post into a WordPress network site I provisioned on a Linux instance in Digital Ocean Amsterdam. It is costing me $10/month for my server instance, 30GB storage and 3TB/month network bandwidth, plus another $2/month for regular backups and and ability to store as many snapshots i’m comfortable with. Digital Ocean, like AWS, are gobbling up capacity at a rate of several thousand new server instances every month, and i’m sure Google and Microsoft Azure aren’t standing still either. Just a relentless tide of predominantly Linux servers that are simple to provision and build applications upon.

We are in a time when mobile, the cloud, a deluge of data and collaboration tools – based on open source software – are brewing a perfect storm. Google, Amazon, MongoDB and GitHub will, I’m sure, be there when the dust settles. Who else?


Simon delves deeper into the subject of vendor inertia in his excellent “Bits and Pieces” blog at