Rethinking: Cloud & Enterprise Computing

Companies currently spend about 5-6% of their revenues on IT. Many of these companies are now struggling to align their IT to support the business strategy, provide a competitive advantage, and serve as a platform for growth. Exploding number of choices and growing complexity of technology assets making these companies victims of their rapidly obsolescing computing infrastructure. Once these assets offered these companies the competitive advantage and served as barriers to entry but now these IT assets are becoming liability. Supply chain meets the cloud to boost the visibility of collaboration processes with and between third-parties such as suppliers, partners, and customers. If companies fail to deconstruct their IT infrastructure and embrace cloud, somebody will do and make them irrelevant.

CORE AND CONTEXT

The bulk of the economic value of organizations is processed through business and consumer supply chains of products and services across manufacturing and services industries. No matter whether it is retail, healthcare, banking, real estate, manufacturing, insurance, communications or others, there are significant gaps in the point-to-point business processes across business’ operations resulting from underlying Infostructure complexity. These enterprises are trapped with their internal IT (Context) focus and are ignoring the importance of the information and interactions across their supply chains (Core). For many companies this has resulted in loss of profitability and in some cases the elimination of products and services all together.

Professor Hau Lee, well known expert in Supply Chain Management said,

“Companies are great not because they were focused on cost or flexibility or speed but because they have the ability to manage transitions – changing market conditions, evolving technology, different requirements as a product moves through its life cycle. Companies also need to be able to handle one more transition: Crisis Management. Successful companies have been able to grab market share out of crises, which often requires them to work effectively across functional boundaries”.

As the recessionary economic and business climate becomes more challenging for organizations, there are many competing priorities and fewer resources to maintain and manage existing operations. Still, with a once in a generation slowdown, it is also an opportune time to re-evaluate where automation and collaboration of these processes can make significant improvement now and in the upturn. It will not be sufficient to just be internally focused on your segment of the supply chain.

SERVICES IN CLOUD: ARCHITECTURAL TRANSFORMATION

Cloud computing is a new deployment and operational model for making IT management simpler and more responsive to the needs of the dynamic business. Cloud architecture decouples the IT infrastructure from the business services. Cloud computing not only enables rapid innovation, flexibility, and support of core business functions but also enables design, development and delivery of new applications by highly efficient virtualized compute resources that can be rapidly scaled up and down in a flexible yet secure way to deliver a high quality of service.

In the pre-information era, suppliers and manufacturers have market power because they have information about a product or a service that the customer does not and can not have. But, now customer has all the information. Whoever has the information has the power. Power is now shifting to the customer. This means that the supplier, manufacturer will soon cease to be a seller and instead become a buyer for a customer. This is already happening. Peter Drucker put it succinctly in his article, HBR Sep-Oct 1997, “Looking Ahead: Implications of the Present“:

“Increasingly, a winning strategy will require information about events and conditions outside the institution: non-customers, technologies other than those currently used by the company and its present competitors, markets not currently served, and so on. Only with this information can a business decide how to allocate its knowledge resources in order to produce the highest yield. Only with such information can a business also prepare for new changes and challenges arising from sudden shifts in the world economy and in the nature and content of knowledge itself. The development of rigorous methods for gathering and analyzing outside information will increasingly become a major challenge for businesses and for information experts.”

Technology is a critical part of supply chain management because companies need to bring together disparate strands of information to be able to understand and assess situations. They also must have analytical services to be able to quickly and consistently decide on the best course of action. A large number of the larger vendors offer some or all of the pieces needed to support more effective supply chain execution — supply chain management and ERP software for collecting data, data warehouses for staging data, and business intelligence software for creating and managing the reporting, scorecard, and dashboard elements. However, they may not be bringing all of the data together in a way that makes it useful, timely, and actionable. To do so, significant integration and customization are needed, which is very time consuming as well as expensive undertaking. Justifying the long development cycles and huge R&D budgets makes these projects not attractive to the business leaders.

Paul Saffo summarized the state of machines, complexity of tools, and exploding information in his HBR article, “Are You Machine Wise?(HBR, 1997)”

“As our tools become ever more complex and interconnected and more central to the conduct of business, their benefits also become harder to recognize. Furthermore, executives need to know and understand the logic of the work done by machines—and, above all else, the limits beyond which those tools cannot be pushed. Meanwhile, the volume of information continues to expand exponentially, generated by machines conversing with other machines on our behalf. Every business activity leaves behind a wake of information, from data spinning off production-line process controllers to transaction records generated over retail-credit-card networks. And the growing centrality of the Internet for business purposes will only add to the flood.”

CLOUDABILITY OF ERP/SCM SERVICES

It has taken good 10 years for companies to embrace enterprise resource planning and supply chain management. This is primarily due to high implementation and licensing costs of the software. In my view, the adoption of cloud computing services in a supply chain and enterprise resource planning many be faster than the former uptake patterns of on-premise enterprise resource planning software. More and more companies are already collaborating with their suppliers, vendors, and partners using the Internet or VANs. It doesn’t make any economic sense to own and operate their own internal data centers to run these applications.

In the same way that ERP/SCM applications have not been employed to automate 100% of enterprises’ business processes, organizations are likely to use a hybrid approach, public and private cloud services where appropriate. Initially, lower-level cloud-based services such as accessing compute power or storage capacity over the internet (infrastructure as a service) and exploiting platform as a service for use in tactical and emerging applications. Software as a service models will be embraced for standardized application areas such as finance, payroll, logistics, human resources (context) that do not provide organizations with competitive advantage.

These companies may also pursue the concept of “private” cloud computing to create their own “private cloud” datacenters. Individual business units (or partners) then pay the IT department for using industrialized or standardized services in line with agreed charge-back mechanisms. This approach is less threatening than a wholesale move to the public cloud, but should make it easier to plan the gradual migration to cloud services.

Joe Wienman enumerated for enterprise adoption of Cloud Computing. Joe Wienman wrote,

“Cloud services are definitely of use for extranet communities…we are seeing it in a variety of areas in AT&T’s businesses. For example, AT&T’s Sterling Commerce unit is a “cloud provider” for supply chain visibility and optimization, and our AT&T Telepresence Solution provides benefits through extranet connectivity, where there is a network effect. And, with networking costs and transaction costs coming down, and enabling technologies such as RFID, sensor networks, electronic product codes, etc., supply chains will continue to benefit from neutral and authoritative cloud services, e.g., chain of custody for tagged pharmaceuticals. And, when two giants are part of the supply chain, e.g., a large retailer and a large consumer packaged goods manufacturer, where should the data reside? If it’s at the retailer, then the manufacturer can access it, but needs to build separate interfaces for other retailers, etc., so the order(n) vs. order(n squared) economics come into play, driving functionality into the cloud.”

CLOUD AND BUSINESS IMPLICATIONS

Economies of Scale: Cloud redefines economies of scale, allowing small companies to enjoy the low unit cost for scaling out their computing infrastructure – traditionally companies with huge data centers only been able to offer rich information to their customers.

Compressed Transaction Costs: Transaction costs along the supply chains are getting lower and they continue to decline sharply. Lower transaction costs are allowing companies to significantly enhance the richness of the information combined with interactivity(soon may be augmented realty), that would have been too costly to capture and process in absence of Cloud like models.

Your Success Depends on Quality of Decisions You Make:A real-time enterprise derives competitive advantage from responding to changing business conditions and opportunities faster than the competition. Often, decision-making depends on computing, e.g., business intelligence, risk analysis, portfolio optimization and so forth. Gartner Research  introduced Corporation Performance Management(CPM) in 2001 which describes all of the processes, measures, metrics, key performance indicators(KPIs), Key Risk Indicators(KRIs), and systems needed to measure and manage the performance and risk of an organization.  Many companies from the Global 3500 and major public sector organizations have embraced the vision of CPM. In spite of its wide adoption across various industry and establishment of sector for itself, CPM failed to deliver on its promise and business practitioners downsized the scope to risk management tool due to the complexity and time it takes to implementation of human intensive data collection and analysis process.

Since an ideal cloud provides effectively unbounded on-demand scalability, for the same cost, a business can accelerate its decision-making combining human and machine intelligence. We see CPM re-emerging as a real-time decision support system. So far, few organizations have figured out how to turn the oceans of data available to them into islands of insight about their best opportunities for growth. Therein lies largely untapped potential for companies to accelerate their growth and separate from the competition (Cloudonomics Law #7).

Create and Stage Rich User Experiences:Using Cloud, enterprise can take advantage of Cloud to reduce the latency of critical business applications (Cloudonomics Law #8).

Availability and Reliability at Fractional Cost: The reliability of a system with n redundant components, each with reliability r, is 1-(1-r)^n. So if the reliability of a single data center is 99 percent, two data centers provide four nines (99.99 percent) and three data centers provide six nines (99.9999 percent). For enterprises to achieve this level of availability, it not only takes huge capital investment, but also drives their operational cost. Instead, enterprises can leverage Cloud to achieve extremely high reliability architecture with only a few data centers (Cloudonomics Law #9).

CLOUD AND BUSINESS CONSEQUENCES

Process Optimizations: Though Y2K provided an opportunity to replace/optimize the old transaction systems with more efficient models, many enterprises have been quick to replace them with standard software – primary goal was Y2K compatibility. Cloud provides a unique opportunity to optimize key enterprise services — business process management, end-to-end visibility of demand-supply patterns, business activity monitoring, business analytics and data warehouse.

Process Standardization: Globalization, supply chain management, and restructuring demand standardization of services with clear interfaces. Standardized services are critical for collaboration, co-ordination, and co-creation with business partners and alliances.

Shared Services: Many enterprises are today utilizing shared services like UPS/FedEx for transportations/Logistics, ADP for payroll processing. Cloud enables these enterprises to explore more opportunities for shared services enabling them to focus more on their core competencies.

Enterprise Messaging Services: Last one decade many standards for information exchange across enterprise applications have evolved like EDIFACT, cXML, Rosettanet etc. Cloud will take these building blocks to the next level by enabling the globally scalable and reliable messaging infrastructure relieving them from expensive VANs used by enterprises today. It makes sense for today’s VAN providers to provide similar services in the cloud at fractional cost.

Integration Services: Even after a decade of huge investments into Enterprise Application Integration services, still integration is the major barrier for enterprises to launch new services. My hope is that Cloud offers a platform to simplify the integration through standardization of service interfaces. Instead of investing into customization and support of these integration services, VAN service providers can offer these integration services, if still required to talk to legacy systems.

Communities of Co-Ops: Cloud enables greater number of cooperating services between the members of a business community (suppliers, partners, customers).

Labor intensive Data Warehouses to Machine Intelligence powered Knowledge Graphs: ERP, SCM, and CRM process measurement generates an unprecedented flood data. Enterprise value is buried in this data. Most of the enterprises can’t afford to have their own IT infrastructure to make meaning out of this data. Cloud enables enterprises to turn their labor and resource intensive Data Warehouses to real-time knowledge graphs powered by Machine Intelligence and Algorithms to provide services to enrich and contextualize this data. Cloud provides an opportunity to bring scale and access required to deliver machine intelligence to the enterprises. This transformation would also help enterprises to effectively implement corporate performance and risk intelligence applications powered by machine intelligence and algorithms. Welcome to algorithms driven corporate governance and decision making!

CONCLUSION

The economic downturn and globalization is changing the way enterprises operate. Changes are becoming increasingly more radical. Enterprises are being broken down into components and reassembled along different lines. The feeling of uncertainty has never been great as it is now. Cloud computing going to play a critical role in simplifying the operations of Supply Chain networks and communities by taking advantage of cost structures and cloudonomics offered by the Cloud. Cloud makes new business solutions possible. This might means new or improved products and services, additional sales channels, more optimal means of procurement, new ways of customer, supplier collaboration, more effective management, and new information services.

ACKNOWLEDGEMENTS

I had the good fortune to be in a good place at the right time and to learn from others who willingly shared their experiences. I am most grateful to the many people who have offered me a helping hand, encouragement, and inspiration along the way. I would also like to acknowledge the years of wisdom many of you has shared with me on cloud computing, issues, benefits, and challenges. My sincere appreciation goes to Joe Weinman for his helpful insight and perspectives on Cloudonomics and Supply Chain. He has generously allowed me to use his ideas and spared his valuable time to review this post and provided me his valuable feedback. I have incorporated number of his Cloudonomics laws and some of our email conversations into this article.

Comments ( 8 )
  • Vanessa Alvarez says:

    I’m so glad you wrote this post…ERP/SCM systems are perhaps one of the more important components of an enterprise, and probably the most consuming in terms of resources. There are some great points that outline the challenges enterprises face with respect to ERP adoption. I do agree with the point that cloud comp services in ERP/supply chain mgmt could potentially leap right over on-prem uptake. For new deployments it’s a (somewhat) simple decision. However, a challenge to overcome will be how to help enterprises leverage their existing on-prem ERP systems already in place, while transitioning to leverage the efficiencies of cloud services.

    Having gone through a couple of on-prem SAP implementations, I can absolutely see the benefits that could be achieved. Many enterprises roll out ERP deployments in phases, meaning that different aspects (e.g regions) are rolled out, while others continue on legacy. Maintaining these in parallel is a painful process, and often go awry, leaving a bad taste in executives’ mouths, not to mention a budget that ends up out of control.

    Assuming that an enterprise does decide to move away from on-prem and choose to go this route, the challenge then becomes bringing customers/partners on board, integrating them into an enterprise’s ecosystem. Different partners, different systems…this is where standardization AND flexibility will be key; cloud services will need to be flexible enough, as customers/partners can change, and so do requirements. However some level of agreed upon standardization (set by customer most likely) is needed. Again, I do believe cloud services can enable that flexibility.

    Ultimately, I believe most enterprises will adopt a hybrid approach with respect to ERP/SCM, and take advantage of benefits of both cloud services and on-prem. Some functions of an ERP system will be delivered through cloud, other functions will be delivered on prem, depending on the level of resources needed to perform those functions. I don’t necessarily think enterprises will want everything delivered through cloud services, but they will need to decide what functions can be delivered optimally from either.

    Of course, then there is the challenge that goes along with a technological transformation of this nature, and that’s the cultural/organizational aspect of it. A change of this nature requires an organizational change; IT, finance, accounting, LOB become much more integrated and fluid. The role of procurement evolves; levels of accountability increase…all changes that become very disruptive. In order for this kind of transformation to happen, it must be an enterprise-wide effort.

    That’s just my 2 cents…

  • John Furrier says:

    Great post. Education and the resource advantage will make cloud for the enterprise a great thing. It reminds me of client server in the early days. The change was facilitated by first a bottom up (grassroots) acceptance then top down business drivers. I think that you’re right on the money here. On site IT resources today remind me of the minicomputer and cloud is client server. On premise datacenters won’t go away but will morph to new topologies and technologies.

    Winners will do more with less and create value not just cost reduction. There has to be a business benefits and there is one. Some old value proposition: productivity and service delivery.

  • Surendra Reddy says:

    I pulled in the twittercussion on how long it would take enterprises to adopt cloud computing for Core and Context applications. Do you think 5 years is too short or 20 years to too long for Cloud adoption?

    randybias: The problem with the assertion that enterprises will adopt mostly public cloud is that, while true, it’s 20 yrs out. What about b4 then?

    randybias: @A_F @samj The transition starts now and will take 20 years just like mainframe to client server. There is too much momentum in cli/server

    sureddy: @randybias very interesting view! It only took 10 years to dramatically change manually driven enterprises to ERP/SCM based ones.

    sureddy: @randybias Cost has been major barrier in addition to cultural barriers for first and second generation IT movement.

    sureddy: @randybias now if IT leaders can’t think ahead of the business, their role will get marginalized. So, time scale are radically changing.

    sureddy: @randybias Globalization, Economic turmoil, and pressure to create more value, competition, information empowered customer& many drivers

    samj: @sureddy @randybias exactly – law of accelerating change. 20 years out is quite probably post-singularity so good luck with predicting that

    samj: @sureddy @randybias that too… there is extreme pressure for a change & “digital native” competitors will devour incumbents

    randybias: @samj @sureddy Some change accelerates and some does not. Given that cloud is more a people change than a tech change, 20 yrs likely.

    randybias: @samj @sureddy regardless there is no way it’s 5. You just can’t grow it that fast and enterprises don’t turn that fast.

    randybias: @samj @sureddy 5 years is an eyeblink in Internet time. Not possible to create culture change that fast in an enterprise

    valb00: @randybias @samj @sureddy Precisely. Cloud is a generational shift, driven by ‘millenials’. They’re in control now, so 5-10 years IMO

    randybias: @valb00 @samj @sureddy 5 yrs is much too short. I could be talked into 10, but it took 15 just for the Internet to mature.

    valb00: @randybias @samj @sureddy Our generation would take 10-15 years, but the younger one will do it in 5-10 🙂

    sureddy: @randybias @valb00 @samj If companies fail to deconstruct their IT infra & embrace cloud, somebody will do & make them irrelevant

    sureddy: @randybias @valb00 @samj Context functions are already moving & then Core –
    valb00: @sureddy @randybias @samj Completely agree with Context already being on the move, followed by Core going to virt (5yrs), then cloud (10yys)

    samj: RT @sureddy: If companies fail to deconstruct their IT infra & embrace cloud, somebody will do & make them irrelevant < - my point exactly

    AAinslie: RT @sureddy: If companies fail to deconstruct their IT infra & embrace cloud, somebody will do & make them irrelevant (via @samj) +2

    drelu: [Agree] RT @samj @sureddy: If companies fail to deconstruct their IT infra & embrace cloud, somebody will do & make them irrelevant

    derik66: RT @sureddy: @samj If companies fail to deconstruct their IT infra & embrace cloud, somebody will do & make them irrelevant < -EA drives tha

    randybias: @sureddy @valb00 @samj yes and no. Agree that companies that don’t embrace cloud run risks, but IT not only barrier to entry in industries

    randybias: @sureddy @valb00 @samj For example, Kaiser Permanente is 3.5B IT shop. Need cloud but will upstart use cloud to dethrone? No. Other barriers

    randybias: @sureddy @valb00 @samj This is not a black/white situation. Critical that biz embrace cloud now, but it will be a much longer process.

  • Jake Kaldenbaugh says:

    Surendra,
    Great post. I agree as well that context is on the move and core is the next big area to transition. I think that transition will happen through a change in perspective. In the past up to present day, the focus has been on the application with the data it has created viewed as a by-product. You see this within the enterprise because each application has its own data store and data management processes. I think the Internets are unveiling a new way to think about how to construct the business with the data being the central focus and applications being the by-product. Companies will move to unified data stores (they’re doing so already) but it will take a long time. I’ve long argued that companies should view their data as a single instance and generate their applications as thin layers off the data. Today it’s the other way around (for most of the enterprise market).
    Great, thought-provoking post. Good to see you still stirring it up! We should get that lunch.
    Cheers,
    Jake

  • Randy Arthur says:

    Surendra:

    I think your points concerning process standardization and enterprise messaging services are where the real payoff will be. Many organizations find changing the culture and the people too difficult, so they buy software the forces the culture to the way the software app wants to do things. To the extent that a cloud delivered service instantiates a best-practice process (or at least a good enough process at the right price) you will find companies improving their business processes by dint of adopting cloud SaaS.

    Integration of the supply chain has been rolling along, but there is still quite a way to go. Standard message bus and the protocols that ride over this bus will make supply chain integration that much better, and will bring the B2B integration only the “big boys” can afford to a much wider constituency.

    As for the Twitter discussion on the speed of adoption – I think it is at least 7-10 years before most Fortune 500 corporations can say that they have at least one mission critical application running in the cloud. Even if an organization had an appetite to get 100% of its context apps into the cloud, it would be a minimum of 4 years to migrate environments because of the problem of stranded capital.

    I think that in 15 years when I am right about to retire, most of what we consider IT today will be in the cloud with maybe only about 10-20% of on-premises equipment. It will be a complex and difficult environment to manage because while each individual component may be simple and commoditized, the overall system will be very complex. Making the aggregate of all the best in breed services deliver the promised value will be the challenge to get right before things will really take off.

    UNIX and whatever Windows system admins will be just as scarce as mainframe system programmers are today (and mainframes will still be around) doing niche work. Only a few people will really understand system programming and the interaction between hardware and software by then. The platforms will not need that much tuning and the abstraction level will be so high – only a handful of truly specialized application environments will require the level of complexity and investment we take for granted today.

    Anyway – great article – enjoyed reading it thoroughly.

    ~Randy

  • Debbe Kennedy says:

    Great article! Enjoyed the comments from others very much.

    I’m coming in by invitation of @sureddy on Twitter to share a few thoughts for consideration from perhaps a little different vantage point…

    Below is a backdrop from which I offer a few observations that I strongly believe have proven to be critical to cultural/paradigm shift transformations:

    Over the years, working across many corporate functions, including IT, dealing with cultural change and paradigm shifts, one conclusion I formed has remained true: Regardless of the particulars or function, dealing with cultural change in the context of a paradigm shift is pretty much the same.

    Most of us, think our circumstances are so very unique. This is true, but the process of change on both fronts is universal. This doesn’t diminish the significance of the change itself and all its inherent issues that need to be resolved. Change is difficult. However, in my experience, engaged leaders and often very simple strategies work far more effectively than big grandiose campaigns and/or trendy methodologies that often make change seem complex and contrived.

    The great news is this universal reality provides a wealth of time-tested KNOWLEDGE to drawn from that can be hand-picked, customized and adapted for specific circumstances and issues. It is often a deliberate step by step strategy that works best — part visible; part invisible behind the scenes. I assume many of you, if you reflect on what has worked for you, works, would agree.

    So when I think and observe the ENTERPRISE COMPUTING transformation in progress, I see IT going through what most other functions are doing. Reinventing themselves for a whole new global marketplace reality. It is exciting for those who can see it and experience it and dream about it. It is suspect to those who “don’t yet get it.” I think the result will change the culture in some dramatic ways, including…
    *** the way we think about competitors,
    *** mutualistic collaborations some can’t imagine with old thinking,
    *** a demand that everyone master the art of putting differences together,
    tapping into new combinations people and technology that create
    innovation far greater than any one could do on their own, and
    *** the development of an open, adaptable working environment that is focused on discover of the next new idea.

    OBSERVATIONS/IDEAS about CULTURAL/PARADIGM SHIFT TRANSFORMATIONS

    • The FIRST CRITICAL steps are to effectively…1) establish THE NEED FOR CHANGE in every mind, 2)engage sponsors’/people’s COMMITMENT to be part of it in a big way; 3) create experiences that immerse people in the “the end in mind” — enough that they can see where they are now — and havea fingerprint on developing the path and outcome of where you are leading them.
    • Most people go yah, yah, yah about #1, but few execute it with superior results achieved. We(and consultants ☺) tend to put everything through “the complexor” — big new words, complicated flow charts, graphs with statistics, surveys and analysis, processes that make most people doubt the change and themselves — all the same ol’ things we’ve all seen before. Then we wonder why senior leaders don’t get it; people aren’t engaged, and change doesn’t take root as we hoped. ACCEPTANCE of a cultural change/paradigm shift or any change is undoubtedly the most overlooked, under-valued, and/or ignored success factor in most strategies related to cultural/paradigm changes.
    • SPECIFIC to ENTERPRISE IT…My observation (offered humbly) is like other organizations, IT tends to talk about its changes in language it understands. When others don’t get it, those people blame the messenger, the idea, timing et al… in other words, those who don’t “get it,” throw up all kinds of “gorilla dust” or create obstacles that most often create a detour or derail progress. Unfortunately, again like others in this situation, the problem is rarely diagnosed as a “failure to execute” #1 and #2 above, but it is. If people don’t get it, it is because it was not effectively introduced — meaning “you didn’t get the order”. It only takes a few cynics, victims, or bystanders — or threatened peers/senior leaders to cause this. Sometimes it comes from consultant-rich involvement that has a bunch of fanfare, but doesn’t really create the ongoing results and organizational engagement and ownership needed to fuel next steps and new levels of acceptance and investment at the organizational level to support success.

    What do you see from your vantage point? Would welcome learning about alternate perspectives.

    Best…
    Debbe

    @debbekennedy
    founder, Global Dialogue Center/Leadership Solutions Cos.
    author, Putting Our Differences to Work…

  • Twitted by AndiMann says:

    This post was Twitted by @AndiMann {Finally got around to reading @sureddy – “Rethinking: Cloud & Enterprise Computing” http://bit.ly/4VMUc3. Some excellent ideas, worth time.}

  • kare@sayitbetter.com says:

    Will able you, S, join the conversation about how Silicon Valley could help Sacramento (state gov) to upgrade its ancient IT systems – including moving some stuff to the cloud? Might be as valuable to SV companies as getting firms to up their IT spending… See Vivek’s post?

Leave A Comment

Your email address will not be published. Required fields are marked *