Archive for July, 2013

Goodbye Sun Ray

23/07/2013

I’d welcome comments from anyone involved in the ‘Sun desktop world’, so read to the end!

The product set that is responsible for much of this branch of the Saul family’s fortunes was cancelled by Oracle on Friday 12 July.

Thank goodness I left well before this.

Because of my personal association with Sun Ray – I attended the first ever UK training course in 1999 and became the Sun Ray architect for Sun’s SEE region in 2004 – it’s easy for me to describe this as a tragedy. Other more choice phrases might be used by the sales and development teams who were abruptly fired with no notice after a quick conference call that Friday morning.

This is the way these mega corps do things – one minute you’re on a training course in Germany with the rest of the pre-sales team, learning about the new release, the next the powers that be have decided that you are surplus to requirements and you are out, along with the entire product set and to hell with the existing customer base, partner commitments and staff. No attempt to fit you into another role elsewhere, no warning. Even customers only found out via an obscure support statement – no direct contact, explanation or time to plan. Odd.

That’s not to say the writing wasn’t on the wall. But it needn’t have been, in my opinion – and this isn’t 20/20 hindsight, this is observation based on spending a long time in the frontline, actually selling to real customers. I and my then colleagues were all saying this stuff ages ago.

I learnt a lot from my time selling Sun Rays across the large region we covered in Sun. I say we, because it wasn’t all down to me, honest, though I can safely claim I played a role.

We were successful – my region was the number 2 region for Sun Rays after the US, with a fraction of the sales staff.

So where did things go wrong? There are some basics that were never addressed during the Sun years, some technical aspects that affected sales momentum and a major strategic sales error on Oracle’s part.

Let’s address these one by one.

The basics were often missing –

Sun never got around to creating some of the basic documentation and sales material that were fundamentally required when it came to selling this sort of stuff. The type of content about which an engineer familiar with the product would say ‘but this is basic stuff and not needed as it’s all obvious’.

The point is that it wasn’t obvious – not to customers and not to partners, even the more capable ones.

When you launch a product like this, you need basic sales material – a sizing guide, an FAQ, material that covers not just your solution but the third party stuff involved as well. The field created a lot of content, but this was often duplicated effort and was not shared properly. There was ‘traditional’ documentation, but that didn’t provide the answers people often needed.

During the early days, when people would ask for a sizing guide, the response from engineering would be ‘there is no sizing guide, you need to pilot this in your customer’s environment for a few months and then you’ll know what you need’.

This is plainly nonsense. When you walk into a car showroom and explain your requirements, the car salesman doesn’t make you rent five cars to drive around in for six months so you can work out what you like, only then to tell you what they cost.

From the first customer meeting you need to be able to stand confidently in front of the customer and give realistic estimates of what deploying your product will involve. If you can’t do that, the message you are giving your customer is simply this – ‘we don’t really know how our own solution works, but spend your money and time with us and let us know what you discover, then tell us what you’d like to buy’. This is not a sales strategy that works.

The principal thing that I learnt is that even in a developing market, it is possible to sell solutions that are, on paper at least, more expensive, complicated and confusing than similar solutions on the market. My mantra was always to ‘own’ the total solution – don’t go into a customer and tell them how your product functions but then say ‘you need to talk to Microsoft, VMware, for this bit, the storage guys for that part and work out the sizing on your own’. I was paid on Sun Rays that shipped, but made sure that I and the partners I worked with presented, explained and understood the complete solution. I wrote clear proposals that answered every question that would always get asked and I put together myself several of the documents that HQ never got around to writing – fundamental things such as sizing guides, answers to the usual things that come up, example RFPs, Statements of Work for PoCs, etc, etc, sometimes alone, sometimes with the help of key colleagues.

(To be fair this situation had got better by the time I left in 2011, but the product was released in 1999…)

In addition there was very little sharing of information amongst the team and very little team spirit in the later years.

But enough about me!

Technical aspects –

The Sun Ray was way ahead of its time but, towards the end of its life, had been overtaken by other devices on the market. Sun were particularly late in the game when it came to functioning multimedia features. It still remained a good solution for many customers.

Sun should have moved more quickly to making the Sun Ray a way of accessing a Windows desktop. During the early years, the feedback from the field was always that customers were not interested in moving to a Solaris desktop with Sun Rays – the only way we’d make this a success was if we could deliver Windows desktops in a supported fashion. When this move did occur, things got a lot better a lot more quickly, but it’s a shame this didn’t happen earlier.

Later down the line, the VDI architecture became unnecessarily complicated. I don’t know if this changed in the last two years, but during my time, to have a redundant solution three servers were required. The engineering logic behind this was that we were aiming for larger enterprise customer with larger installations and those large installations would easily need three servers, probably more. This was fair enough, but ignored the sales logic that even largest customers started off with a smaller number of users and the investment required to purchase three servers that would mostly sit idle whilst running a smaller number of users was prohibitively high an entry point. It also ignored the average size of the majority of the deals that I was aware of. Make it easy!

The Sun Ray architecture was always, at first glance, ‘more complicated’ than competing solutions, but in my experience, when pitched properly, it was easy to overcome the objections – providing you had the right supporting info to hand, namely the sizing info, answers to common questions, etc.

Sun Ray didn’t have the full complement of surrounding products that a company like Citrix had, even back in the mid-2000s, but again, this is something that could easily be explained away to new customers. You wouldn’t kick Citrix out of an account, but if you were first through the door you could easily position Citrix as being overkill – there is truth in both sides of the story, as in every sales situation! Sadly, after the acquisition of Tarantella, the Citrix relationship cooled considerably, so we were always competing. Had co-operation continued with an up to date Citrix client for Solaris Sparc and x86, Citrix could have been a key ally and enabler.

The major error –

We were affected by some of the same malaise that affected Sun as a whole.

It’s easy to hide behind technical failings when a product is canned, but in my limited experience, blaming the product is very often the case of a workman blaming his tools. The industry is full of mediocre products that become big successes. Microsoft are an obvious example. It’s how you sell them.

For example, Sun was never very sophisticated at incentivising their channel with rebates and other tricks of the trade (to the best of my knowledge, at least. I am happy to be corrected). This sort of thing was a real eye opener during my brief tenure at HP, who incentivise their channel in all sorts of ways.

There was a culture at Sun that seemed to say to partners ‘this is obviously the best technical solution for you and your customer and if you sell it you will make money’. Perhaps having a mediocre solution encourages people to be more inventive? ‘This is obviously not the best solution on the market, but the price to customer is similar and if you sell it I will pay you a rebate so your margins are better than with anyone else’s offering’. As a partner, which solution would you push?

The Sun Ray focused channel I had available were typically very good, however – usually smaller, capable partners looking to differentiate themselves from everyone else out there, with good skills in house and the right attitude to using me and my colleagues’ time and effort. I never flew to Tunisia or Beirut and had a wasted minute, for example. When we were in competitive situations with HP and Citrix, our partner was usually streets ahead of their competition, who were simply forwarding quotes rather than putting together a total solution.

The channel could always have been better, but this isn’t where the main blame lies. The blame lies with a lack of sales management during the final days of my time at Oracle and, I am assuming, during the past two years after I left, coupled with one major mistake.

The most successful period of Sun Ray sales was when we all reported to the software sales team – there were software sales reps, ‘desktop’ product sales like me and pre-sales, all in one team, working alongside the ‘normal’ Sun account managers or regional sales managers who got paid on anything that was sold in that account or territory. There was also some kind of desktop overlay overlay team, whose role I never really understood. If a Sun Ray was sold, multiple people got paid commission. Too many people? Definitely, but the whole point of Sun Ray was that it was a growth product intended to drive new revenue in new accounts or expand business in account who had bought pretty much everything else Sun had to sell. There may have been too much overlay, but there was no fundamental problem with having a product focused sales teams to assist everyone else. The software sales team had clear targets and regular forecast calls. We were also incentivised to overachieve – doing more than 100% was well worth your while.

Sales were good.

Towards the end of Sun the software sales division was changed and Desktop split out into its own management structure – people like me now reported into the overlay overlay team mentioned above. Again, there was nothing wrong with this in principle, as the primary route to market for us – the hardware account reps on the ground – still existed.

The problem was that this overlay team either lacked direction and appeared hamstrung by internal politics as opposed to playing the role of a sales driven organisation. I will admit, it’s easy for an infantryman like me to blame the generals – I truly don’t know what was going on upstairs and I can’t attribute blame to those in charge at the time, as I don’t know what they were going through with their management. However, the fact remains that despite a dedicated field sales team, all of whom had long experience with the product set, we still didn’t see the changes the field was asking for and there was little sales direction and drive coming from above. No one seemed to own the number and drive that number. Where people did nominally own the number, it was all a bit waffly.

During my time with this team I never once had a forecast call – not once, in several months. A call where my manager grilled me on very opportunity in my pipeline, asked for my commit for the quarter and held me to it. I still can’t quite believe this state of affairs. If I missed a forecast call or report two weeks in a row at Citrix, I would be fired.

So, no one was driving the number in a ruthless, sales focused way. Even the most disciplined field sales rep needs a second set of eyes and ‘encouragement’ to ensure they are doing things right and are focusing properly. Plus, during this time, what numbers were being reported to the new Oracle overlords and how were those figures and forecasts arrived at?

Secondly, the amount of admin was ludicrous.

To manage my pipeline I had to use Siebel. Every Sun Ray opportunity I added to Siebel required 15 fields to be filled in using a very old version of Internet Explorer. Changing a field led to a screen refresh. It took ages.

To add insult to injury, about 14 of these 15 fields always had the same info as every other opportunity – region, sub-region, product set, etc. I once asked if it were possible to have these sections pre-filled in with the default entries every time I created a new opportunity. This was apparently not possible.

So you had sales guys, who are not the most disciplined people when it comes to admin, being forced through hours of tedium to add an opportunity, with no forecast calls to review the sales situation. Presumably you then had people looking at Siebel and making business decisions from numbers that were pretty shaky.

But enough reliving the agony of Siebel, it’s time to summarise the key sales mistake that I believe eventually led to Oracle canning the product. The issues described above are present in every organisation to some extent. Sun Ray failed because of something very specific.

After the Oracle takeover, Oracle decided that ‘no overlay sales teams’ would be allowed. No ‘double bubble’ commission. What this meant for us was that the account reps across my huge region with whom I worked to sell Sun Rays were no longer goaled on Sun Rays. This was the major error that stopped sales growing as they could have.

For example, a deal with $400K of Sun Rays and $80K of servers had previously represented $480K of revenue for an Oracle hardware sales rep. Well worth pursuing with Chris’s help – longish sales cycle, complex sell, but rewarding at the end.

Now, overnight, that deal was only worth $80K of revenue to that account rep.

‘Chris, you’re a nice guy, we’ve worked well together in the past, I’ll hand over my contacts at the Ugandan Water Authority to you and you’ll have to take it from there…’

All of us ‘Desktop Sales’ people were in the same boat. I had a target that was larger than the previous year’s, but overnight my entire salesforce – the Oracle hardware sales team – had gone. It was now me, a pre-sales engineer and most of the Middle East and Africa to cover, with a channel still in disarray following the Oracle takeover.

There was some good news – I now got revenue recognition for keyboards and monitors…

With this situation it was clear I would never come close to doing my numbers, even if the other problems had all been solved. This is when I realised there was no future for me and I had to leave. Either I stayed and bumped along with no commission coming in for as long as I could, or I got out. As I couldn’t see the pain of the Oracle transition fading for the foreseeable future, I chose to leave the company. At that time I also learnt a lesson, which was not to be too loyal to a company or product – it’s arguable I should have left that role a couple of years earlier. With hindsight, I should have done my best to join VMware at the time, but that is a different story.

What was particularly odd about the supposed ‘no double bubble’ mantra is that Oracle did form overlay sales teams – there were x86 and Sparc overlay product sales teams, for example. There was no reason I can see why the desktop sales team could not have continued as it had previously, with double bubble for the desktop guys and hardware field sales reps. Had we done so, I firmly believe the product would have lived on successfully.

Oracle is a powerful brand whose name gets you direct access to senior people. If the hardware account reps had continued to be goaled on the desktop products, there would have been revenue growth coming in from the dedicated and passionate desktop sales community who had backed the product for years and made a career out of it. Fix the management layers above and you’d have had a real success story. (I hate to use the words ‘passionate’ and ‘community’, but they do fit in this case).

As it was, you had a few lonely reps trying to flog Sun Rays and compete with the noise in the top accounts against all the other Oracle sales guys in there. The targets remained the same but our extended salesforce had gone. How can the Middle East and Africa rep do direct sales across that huge region? It would be a challenge even for a rep with one country to handle. (To his credit, I believe the UK sales guy did very well, as he had been able to grow his business via a more direct model during his time in a previous role before transferring to the desktop team).

That ‘no overlay’ decision, I believe, is the main reason why revenues never grew to the level Oracle would have wanted them to be. I’m surprised things latest as long as they did, but now it’s all over.

Sad.

Advertisements

Eye of the beholder

22/07/2013

Riz, my brother-in-law, took this snap of Zara while we visited Milton Keynes’ famous concrete cows yesterday.

Here’s the original and a couple of crops – you can see me and Riz reflected in Zara’s eye, Riz with his camera and me pushing the buggy.

DSC02576.JPG

DSC02576-002.JPG

DSC02576-001.JPG

DSC02576-003.JPG

Mum, Dad and Zara at Milton Keynes’ most iconic location. Dad needs haircut and a shave.

DSC02579.JPG

Who to hire

19/07/2013

Tim has an excellent blog post here – it’s long, but I encourage you to read all of it. Insightful stuff.

I remember being surprised when I first joined Sun and learnt that the guy who managed a team of highly technical security geeks earnt less than any of the people who reported to him. My naive assumption was that whoever is ‘the manager’ would naturally earn more than his team members. Of course, there is no way why this has to be the case, with situations like the security consultant team neatly highlighting the situation.

The manager’s set of skills was completely different to the security geeks’. The security geeks would all have made awful managers. Their skills were rarer than their manager’s.

I found Tim’s example of companies only hiring ‘the best’ very interesting. Ending up with highly intelligent people who all get bored and hate their jobs but get paid a fortune.

At the time I was leaving university, I was surprised to see some people being hired by some of the top management consultancies and being paid an absolute fortune to perform a role that was basically equivalent to that of a junior programmer. You had top Oxford graduates who had not studied computer science and who had no previous programming experience doing a crash course in C++ and then spending six months straight living in a Premier Inn on the outskirts of Huddersfield programming ERP systems for corporate customers. As time went on some of them had more interesting roles, but it involved a huge amount of work – and I wonder how much brain power. You don’t spend 18 hours a day at work, six days a week, for a year, with every minute filled making intelligent decisions or insightful comments.

This might sound like sour grapes, as I didn’t make it that far through the interview process, but I am still glad I didn’t end up going down that route.

Back home soon

17/07/2013

I leave for a long holiday on Thursday afternoon.

Hoping to take it easy and take a proper break too. Lots of reading, lots of blogging and lots of time with this one –

IMG-20130717-WA002.jpg