systematic viewpoints

I’ve spent over half of the last 10 years helping enterprises get greater use of their ERP systems. Having been by turns a graphic designer, IT and development manager, user experience advocate and close ally of business, marketing and communications professionals and strategist mine is a particularly multidisciplinary approach.

I sense the beginnings of a change coming about, although I think it will be some time before it’s fully manifested in products and ultimately in the workplace. I’m still trying to hash this nascent trend out, so bear with me and please do call me out or remix these thoughts.

How did we end up here?
If I had to describe a typical ERP deployment (necessarily a fiction, there’s no such thing) , it would have the characteristics of an installation - scaled to the usage estimates, tuned to perform acceptably but not optimally under real-world conditions, configuration changes only, no customizations allowed by IT.

It took longer and cost more than projections. Business requirements were gathered but often ended up being deferred so the critical path could be cleared of dependencies that would incur further costs and/or delays, worsening the tension that already existed between the business audience and IT. A launch is achieved with one or two key business functions being supported. ‘Features’ are rolled out over multiple releases until all the intended functional solutions are live.

Now What?
What happens next is highly variable. Frequently budgets have been strained to the point where planned change management activities are scaled back or even eliminated in favor of some form of training. This is often remote and offered for a limited time after a launch event. Recorded training is available for new employees - if they can find it.

Professional users in the functional areas begin to struggle with the gaps between local procedures and the methodology of the system as delivered. Specific pain points arise: inconsistent data sources, multiple screens to perform single tasks, you name it. Workarounds abound - job aids and cheat sheets are circulated, and a body of underground tacit knowledge required to successfully perform job functions begins to arise. Eventually metrics begin to suggest that the ROI is not being met, and the blaming begins.

What’s to be done?
How it plays out depends on how the people responsible for the systems are rewarded. I’ve just re-read an interview with Donald Norman from 2000 where he took the usability profession to task for not understanding how business people typically get promoted, and emphasizing long-term benefits to the wrong audience. His point was if a manager gets a very narrowly defined task completed without making a mess of their P&L sheet for the year, they get promoted. Usability? Service quality? Benefit realization? Not my job - that’s for the next person to achieve.

Companies are frequently motivated to address problems arising from ERP deployments because senior management relies on them for critical processes and key data and they are not achieving the desired results. They assign that ‘next person’ to improve the system. Sometimes they call in folks like me.

Over time and through many engagements we’ve identified a spectrum of possibilities that improve in varying ways the business results that ERP supports, depending on a given company’s appetite for change and customizations. It’s not about user-centric design, although that’s a key component. It’s about tasks and goals and how people get through complex, lengthy processes. It’s about how the systems support the strategic goals of a company. Sorry to say, no system delivers that out of the box.

Vendors know the truth.
This challenge is very clear to ERP vendors. Their interfaces are brittle and monolithic; corporate IT experiences so much pain customizing and maintaining them that they have very compelling arguments against modifications. SaaS companies like Salesforce.com and Workday are invading their turf.

Oracle knows this, but they’re too busy rationalizing their product lines to be able to address it head-on yet.
SAP knows this and even though they provide tools for IT to tweak interfaces they are not used in may enterprises for the reasons above.

Change is coming…maybe.
One of the biggest challenges in any system is how to design for large numbers of people across many disciplines. Many of today’s applications try to accommodate just about everyone, creating extraordinary complexity. This applies as much to Microsoft Office products as it does to ERP. Word and Outlook are ‘feature-rich’ to the point of being ridiculous for must folks.

Other paradigms for improving the interface to ERP have been in play, most prevalent being the dashboard. They can be terrific for information consumers but they are often implemented with limited interactivity for decision support. A very compelling set of demonstrations was given at SAP’s Munich TechEd event showing interfaces and widgets that begin to decouple interactions and data manipulation from the ERP interface. Oracle and SAP both have dedicated groups looking at ways to exploit the best of Web 2.0 technologies and interfaces to the business solutions.

I’m not sure whether folks can cope with widgets floating around their computer desktops, monitoring data, work lists, or enabling faster/simpler transactions. But in general people prefer use-specific interfaces and devices over multipurpose ones. I commonly use the kitchen as a case in point. Your own kitchen probably has a range/oven, a microwave and some form of toaster-oven. 3 devices, all specialized interfaces for making food hot in a chamber.

Folks like Don Norman have envisioned more embedded computing and fewer general purpose systems in the future. In the last year specialized computing products have bloomed in the consumer space: digital picture frames at Target, iPhone and iPod Touch, Chumby. Perhaps the general public’s embrace of Web 2.0 interfaces (which seem to tend towards the single-purpose) is beginning to create sufficient demand that the product managers for ERP systems can contemplate adding them to feature sets. For some interesting insight into the dynamics of that process, see “Why 2.0 Didn’t Start in the Enterprise” by Paul Pedrazzi.

How does this impact the enterprise?
I see a shift away from the massive interface, the all-in-one portal and the soup-to-nuts dashboard in favor of compact, customizable and intelligent widgets, applets and services that can be called upon demand or pegged to a corner of the screen. I see a move away from the browser and the page paradigm that demands information architectures and navigation, towards a set of easily grabbed tools that I can use in combination or snap together like Lego blocks to solve my here and now business problem, and move on. The browser will still have it’s place because it’s a great interface for linear processes, but it will stop trying to be everything to everyone. I’m almost reminded of the March 1997 issue of Wired magazine, which breathlessly declared the death of the browser. I still have my copy.

When I watch the Demo Jam video I think that it’s some of the better thinking I’ve seen in this space in quite some time, but realistically speaking these innovations aren’t ready for general availability. Enterprises are often years away from major upgrades of ERP; in fact the days of the sweeping upgrade are probably past for many organizations. It’s incremental change that will be coming, so I don’t expect the landscape to change drastically in the next few years. But it’s an exciting trend and when these innovations begin to creep into the enterprise, I fully expect demand for more to rise.

Oracle sues SAP for “corporate theft on a grand scale”, according to Reuters.

Tomorrow I’m reviewing an upgrade approach with my technology team (the one I used to manage) for moving the PeopleSoft Enterprise Portal from 8.46 to 8.9. I need to dynamically mix transactions and content with awareness of context, person, function and process. Having measured the delivered functionality against my needs for almost a year I’m pretty sure that I’ll decide to pass.

I’ve been in a love/hate relationship with this product since I deployed 8.3 three years ago. From the very start we needed to modify it to do things that it can’t easily do. I run one of the largest implementations on the planet and have pressured Oracle to be more forthcoming with me about the road map for the Portal line and Fusion, but I don’t get a lot to go on.

Right now I have a very expensive, slightly intelligent link farm. Bottom line is I’m not inclined to take the time and budget for an incremental upgrade that isn’t going to bring me a whole lot of business value. The most likely scenario for 2007 is I’ll tweak the UI and look to replace it entirely in 2008-9.

Computerworld reports that “Oracle Corp. is already claiming to be ahead of plan” on delivering Fusion, saying that they are halfway along.

We spent the morning of our second day identifying what, by application, we percieved as the major business benefits from what’s delivered in the PeopleSoft upgrades. This was particularly compelling for our regional HRMS folks as they were able to compare and contrast approaches. There was a really positive flow of ideas, and you could practrically feel the group come together into a working team. This was possibly the best take-away from the event. I’ve seen time and again that the power of a motivated, smart group is incredibly more effective than individual efforts, no matter how informed.

If we expected to hear anything more specific from Oracle about Fusion, it wasn’t forthcoming. They’ve rebranded exisiting middleware as Fusion, but nothing that gave the HR product landscape any more substance was brought out. In terms of timing, they told us that the first Fusion applications are slated to begin appearing in 2007, the first Fusion ‘application suite’ for HR is targeted for 2008 and to expect it to be well beyond then before it’s sufficiently stabilized for enterprise-scale environments.

Our last day was a wrap-up, and we caught our planes back to our respective continents and cities. It was a US holiday Monday, but already emails are flying aorund following up on specifics that were raised. This morning our team met and we’re working on our summaries, follow-ups and thinking hard about how to maintain the energy from this event.

This event should mark the last of my travels for this season…I hope. I’ve been away from home a lot and it gets very tiring to be slogging aroung like that.

Long day but it was actually a good expreience. We’re having a strategy meeting about upgrading our full PS environment - 4 HRMS instances, sPro, EPM and Enterprise Portal - to 8.9. We started with sharing a definition of our target state. These are the carve-em-in-stone principles we want to deliver and/or achieve, regardless of requirements or technology. They’re simple and high level:

• a quality user expreience,
• timely and accurate data sourcing and feeds management,
• powerful and practical reporting and
• maximal system and service consolidation.

These four objectives contain many characteristics or best practices, each of which in turn will expand out to identified gaps between the as-is and terget state. The first order of business was to get agreement on the objectives and characteristics. Having that, we then went around the room with each functional owner speaking to the top of mind gaps in their areas, which we managed to filter into a single list of key gaps from a global perspective. Some of these will take years to address, others hang low but now we all agree on what they are. Everybody has to go hame and produce an exhaustive catalog.

The next stage was to identify what we wanted to achieve with 8.9 from a purely business side. How are we improving processes, enhancing services, shortening cycles, etc. The plus is that we now have a common ste of standards by which to measure any aspect of this effort - if it doesn’t deliver on one of the target state characteristics, we can fairly ask why we’re doing it. Further, with each of the 4 regional HRMS managers present in the room, talking to their goals, there was ample opportunity for sharing and comparing that was exercised in a really positive way.

By 6:30 PM we broke to go to dinner after an spontaneous whiteboard session between EMEA and Asia-Pacific on tree standardization, inspired by the warehouse and North America discussions on their respective approaches. Much lubrication followed and people were voicing their wonderment that we hadn’t started like this 3 years ago when we took this effort on on a global scale. Honestly, we didn’t have the luxury of strategy back then. We were racing a clock and a leaky budget, and we had to get the ‘plumbing’ in place. We made a lot of compromises and there’s no question that what we’re doing now is unraveling some of that.

I believe that with a vision, communications and a strong group working together we’re going to realize some cool achievements. With some blood and guts spilled, no doubt.

Orace is telling us about 8.9 and Fusion tomorrow. I’ll have to see what I can share, we’re under non-disclosure.

If I thought I’d be able to take it easy and catch up when I got back from travelling, I was sadly mistaken. There’s been a blur of activity, a lot having to do with our recrutiment workstream. They’ve been working to get Taleo out as a global standard for internal and external postings. Again, due to the size and complexity of our organization, deploying any global standard is a hige challenge. Recognizing that, the team was charged with creating something that links to not only Taleo but any de facto job board used within the company. I’m not confident that we even have a definitive list.

To the horror of the workstream lead I proposed that we proceed with a target state definition of an integrated career management environment and use that undoubtedly compelling vision as leverage to convince people around the company that the heavy lifting and pain that it will take to adopt this particular standard will be worth it in the end. I understand their pain, as an escaped technologist I know it would be in their intrests to simplify, not amplify. What I’m fighting is the emergence of a fat new silo created by lashing together a bunch of old silos. In any case I think she ultimately agreed so long as she can deliver something in March to satisfy the basic request. But now I have her attention while I create a picture of a critical part of the overall target state for self-service users.

Noted Jakob Nielsen’s Alertbox post “Why Ajax Sucks (Most of the Time)” I’m in the love/hate camp with regard to Jakob. On the one hand I very nearly engaged him to speak to senior management when I was creating the business case for the latest version of the corporate intranet, on the other hand I sometimes find his positions unnecessarily orthodox/purist. In this case there are some spot-on issues - I watched focus group users struggle with some of the Ajax elements in the intranet prototype, especially around the use of the back button. This is something that needs fixing, like it or not. That said, I believe that the Ajax approach has much goodness, especially as it will apply to a thoughful self-service environment.

I’ve also been asked to put together a short paper for internal use on Fusion as a follow up to Open World. Given the current state, that should be brief.

Sunday I travel yet again, we’re taking our road show to Colombia, I come back Tuesday and then out again Thursday for Dallas and Mexico City. That’s the end of the intranet prototype tour, and I’m glad for it. I love to get out now and again but I’m just plain tired.

My technology team is a week away from moving a Tools upgrade for the PeopleSoft Portal to production. With 300,000 users and 18 integrated applications, it’s been quite a dance. The toughest part is getting a green zone when we can absolutely take the Portal (and most of the integrated apps) down and make the changes, then rouse tech and business people in the middle of the night on a weekend to test and approve. Since we’re global there’s only an hour and a half during per week that is outside of someone’s normal working hours, which of course isn’t enough for this task. But this is Comp season, so lots of people are working weekends. Overall, it’s not pretty.

Next we want to upgrade the application itself from 8.4 to 8.8 or 8.9. I wonder what happens after that - Fusion, I guess - so that could be our last major PeopleSoft upgrade.

Third-party companies are seeing opportunities to provide better support for companies considering Oracle’s new Extended and Lifetime support models, as described in InformationWeek.

Back in New York! So what are my after thoughts? First, the size of this event was challenging, the general estimate I heard was 37,000 attendees. That made initial introductions a little like speed dating - “Are you Oracle? PeopleSoft? JD Edwards? Apps? DB?” It made it hard to figure out who to network with. I was thinking about ways to make it easier for us. At the trade show, EMC (the storage and enterprise content management company) was handing out big, numbered pins that people were wearing like license plates. The hook was that somewhere out there was another attendee with the same number. Find them and win something! That made me think that Oracle should put RFI tags in the show badges that would light up when someone with a similar customer profile approached. Or maybe just color code them somehow - keep it simple.

If I didn’t say it already, I’ll say it again - the messages were remarkably consistent:

• Oracle is moving to standards-based platforms and products
• Oracle will provide flexibility and choice between their and other company’s products when architecting business solutions
• Fusion will be best-of-breed from their entire product portfolio
• We’ll support you as long as you want until Fusion, and even after that
• Oracle has a laser focus on the customer’s needs

I have little direct background with Oracle, they were always a commodity component (the standard database layer) in my applications. So I’m taking their claims at face value until proven otherwise. With my arms folded, I guess. I took away some good connections with individuals who are doing the same things, that I can share ideas with. And for all the good messages and rock-star keynotes what I ultimately I saw for the here and now were basic, incremental changes to the HRMS, EPM and Portal applications, and a lot of hesitant customers wondering how this will play out. If they do it well, we could get a lot closer to the kind of systems and usability that we hoped we would get (and usually didn’t) when we purchased PeopleSoft apps. If not, I expect a lot of folks will take them up on their ‘lifetime support’ deal.

Now it’s back to day-to-day efforts. Today it’s getting a plan and support together for rounding up some standalone intranet sites that provide mandatory training and attestations and getting them into a LMS and front-ending it through the portal.