Cloud ERP adoption – Eric Kimberling reveals surprising data and misconceptions


Eric Kimberling of Panoroma Consulting has some interesting views – and data – on cloud ERP adoption. But when we got into a back-and-forth about cloud ERP versus SaaS, it led us to the age-old ERP issue of over-customization. How do you advise clients determined to head down a problematic path?

fingers-crossedMy last piece with Eric Kimberling, on project failure and ERP vendor rankings, provoked debate and rightly so. Any form of vendor ranking veers into subjectivity, something Kimberling acknowledged. As for project failure, everyone is bloody well sick of the topic, but alas, we can’t leave it alone until failure rates finally improve.

One potent area we missed was cloud ERP adoption. That’s an issue that Kimberling’s firm, Panorama Consulting, addressed head-on in their annual ERP report. The short version: In 2017, Kimberling expects cloud ERP adoption rates to exceed on-premises in his survey for the first time. But – there are some qualifiers – including the type of cloud deployment.

In a follow-on call, Kimberling told me how his views on cloud ERP have changed. In short, he is less of a cloud ERP skeptic but not an evangelist yet either. He also shared surprises, which can be traced back to Panorama’s distinctions between cloud and SaaS ERP.

Cloud ERP adoption is rising – with caveats

A few caveats: Panorama’s annual ERP reports don’t tend to have a huge sample size (2016 was comprised of 215 respondents). The data gets more interesting with year-to-year comparisons; Panorama’s first ERP report came out in 2009. Kimberling’s insights also stem from his client base, which centers around midmarket manufacturing companies. These companies don’t tend to be early tech adopters. That means Kimberling’s views on cloud are likely to be less religious than some:

Most of the clients we work with are not bleeding edge type companies that are wanting to be the first to adopt cloud technology. These are midmarket manufacturers… they just want to make their businesses better. We’re a little more mainstream in that way, whereas a lot of the industry analysts are maybe more on the forefront, talking to early adopters embracing these trends.

ERP deployment options may be shifting, but overall ERP adoption continues. In this year’s survey:

  • 81 percent of organizations are either in the process of implementing ERP software or have completed implementation.
  • 14 percent of organizations are in the process of selecting software.
  • 5 percent are in the process of upgrading software.

As for cloud adoption, this year’s numbers showed the highest levels of cloud ERP adoption to date:

  • On-premise – 56 percent
  • SaaS – 17 percent
  • Cloud ERP – 27 percent

Defining SaaS ERP versus cloud ERP – in Panorama’s terms

You may be wondering about the SaaS -versus-cloud distinction. Panorama defines “cloud ERP” as a hosted solution, an off-premises deployment. Whereas SaaS ERP would be the classic multi-tenant SaaS definition, a growing category of cloud ERP solutions that includes the likes of NetSuite, Plex, Intacct, FinancialForce, Workday, and Acumatica (SaaS ERP is a big ol’ umbrella; many of those cited would not view each other as direct competitors, and rarely see each other in deals. SAP, Oracle, and Infor also have SaaS ERP offerings).

The biggest distinction between cloud ERP and SaaS, as Panorama has defined it? With cloud ERP, or “hybrid” or “hosted” if you prefer, a company can customize the code base to their heart’s content. SaaS ERP implementations do not allow code customization and usually require a rethink and/or simplification of processes within their configuration choices. As an advocate of “cloud native SaaS” as defined by my colleague Phil Wainewright, and a skeptic of hosted ERP or a “lift and shift” to cloud approach, I was intrigued to hear Kimberling’s views.

Cloud ERP – cost savings versus security

One benefit of cloud ERP that Kimberling is tracking – cost savings:

From Panorama’s 2016 ERP Report

In my experience, cost savings is not the main benefit customers cite from a cloud ERP move. And: when you do a full tally of upfront versus subscription-based, year-over-year cost savings are not always significant. Kimberling responded:

In our data, we do see some cost savings, but it’s not a key driver or a key reason why people are moving to cloud ERP. It’s more about the headache, and just not wanting to manage an infrastructure – which I totally get.

When it comes to barrier to cloud ERP adoption, security concerns still rank highest for Panorama:

  • Perceived risk of security breach – 29 percent
  • Insufficient information about cloud offerings – 16 percent
  • Perceived risk of data loss – 9 percent

This led Panorama to emphasize, under the heading “Misconceptions”:

In our experience, we have found that the above concerns are misconceptions. In fact, a number of cloud providers are enhancing their security to meet customer needs, especially in the financial sector.

Cloud ERP versus SaaS ERP adoption – why the difference?

So what about the growth of cloud ERP (the hosted stuff), from 11 percent in 2015 to 27 percent in 2016? From the report:

The significant increase in the use of cloud ERP, from 11 to 27 percent, makes sense considering the increasing number of vendors providing this option. Both SAP and Oracle are investing more research and development in cloud technology than in on-premise software, and other vendors are quickly following.

So I asked Kimberling, why is cloud ERP on the rise in his surveys, but not necessarily Saas?

Even with our own clients, we’re seeing a lot of companies that are saying, ‘We want the best of both worlds, we don’t want the whole multi-tenant SaaS thing. We want the flexibility to do whatever we want to with the software. But: we don’t want to deal with all the crap that comes along with on-premises solutions, so let’s go with cloud, where we own the licenses. We can do whatever we want. We can bastardize it if we want – for better or worse – and we can have someone else deal with the headache of managing the infrastructure behind it, and the application support.

The ability to customize a hosted solution factors heavily for Kimberling’s clients:

I think the SaaS ERP model’s a little bit stuck right now. We don’t see a lot of movement there yet in terms of growing exponentially like overall cloud is. You’ve certainly got vendors like Plex, NetSuite, Workday, and Salesforce that have a pretty strong niche within their respective areas, but as a whole, SaaS ERP is not growing out with the pace the whole Cloud movement is.

I asked Kimberling: in any given five cloud ERP adoptions he sees, how many are SaaS? Answer: just 1 in 5. In smaller companies – or companies that don’t have legacy processes -the ratio changes:

Yes, that does shift quite a bit when you look at smaller organizations, or companies that maybe don’t have very mature processes or systems in place yet, that ratio is a little bit more balanced. Maybe not quite 50/50, but maybe it’s more three to two than four to one.

Kimberling sees SaaS ERP trending in those cases:

These types of companies have less to lose. They don’t have as much baggage with the old processes, or a set in stone way of doing things that will be difficult to change.

Kimberling underscored that we are talking SaaS ERP here, not line of business SaaS like HCM Or CRM:

Remember, I’m talking full-blown ERP. If we’re talking to companies that just want a CRM system, or just want an HCM system, that’s more in favor of SaaS – I’d say 3:2 SaaS.

The problem of over-customization – and client advisory

We had an in-depth talk on the dangers of bringing a “customize the code base” mentality into the cloud – and the problem of client advisory. Check this exchange:

Eric Kimberling: It seems too easy to say, “I’m just going to buy the software licenses, and I’ll have someone else deal with all the crap and I have all the flexibility I want.” It sounds good, but to your point, there’s a lot of risk there. And there’s a lot of downside you’re going to experience when you go through the implementation.

Jon Reed: Doesn’t this go back to a lot of the advice you’ve been giving for years on avoiding project failure? It seems like one of your key arguments is: don’t get caught in the over-customization trap – use this opportunity to rethink your processes.

Eric Kimberling: A lot of executives just don’t want to deal with it. Because of what you just said: that’s a lot of work up front. Their view is, “I want to do is go out and get an ERP system, and now you’re making it all complicated by telling me I need to rethink my business processes.”

Jon Reed: Your point brings out that tension of being a consultant. I think that the best consultants are advisors. Sometimes that means telling people what they don’t want to hear. But there’s a point where you do have to back off from that.

Eric Kimberling: We face that challenge. We have clients where we feel like they’re doing it wrong, or they’re going down a wrong path. Sometimes we have to catch ourselves and say it’s not really about right or wrong. We can talk about where the risks are and where the odds of success are going to lie. If they choose to go down a path that we think are minimizing their odds of success, we’ll tell them that. We have to try and guide our clients through that, and we’re not always successful, by the way. Sometimes our clients push back, and say “That’s great, I hear what you’re saying, but we’re different. We need it this way, and here’s why we go down that path.” You’re absolutely right about that.

My take

I do believe, in the long run, that SaaS ERP will become the dominant ERP deployment model across SME. Large enterprise will likely go in a different SaaS direction, opting for LOB-based SaaS with open integrations, allowing them to mix and match their own. For large enterprises, it will be data control and trans-national regulation that cause some SaaS holdout, not flexibility. Some of the SaaS downsides Kimberling noted, such as lack of flexibility and integration, are now being mitigated by extensible solutions, a platform of add-on apps, and API-enabled services.

Kimberling met me halfway on that point:

One of the potential assumptions that could change would be if SaaS were to continue down the road they’re going, which is to make better integration and personalization tools. The more flexible SaaS becomes, the more likely it is that you’ll see more companies go that route versus the (hosted) cloud.

Kimberling used to be a cloud skeptic, which dates back to the days of ASP overhype. Now Kimberling is persuaded by the data – and client feedback. But he still stands against cloud krishnas who would advocate cloud as the only approach. I don’t have a problem with that. Customers should be the arbiters, and when was the last time customer needs could be generalized?

Image credit - Lying businessman holding fingers crossed behind his back © Bits and Splits - Chart from Panorama ERP report credited above.

Disclosure - SAP, Oracle, NetSuite, Plex, Salesforce, Financial Force, Infor and Workday are all diginomica premier partners.

    Comments are closed.

    1. says:

      Thank you for interesting article and sharing statistics.

      I want to add one more reason why companies that don’t have very mature processes (often SMEs) prefer SaaS over more customizable solutions. As usually SaaS offer less customization and configurations, they have stricter structure, which can be used as a canvas or a framework to setup company structure and business processes. Often in young companies everyone does everything. And when company starts to grow, this needs to be changed. Well designed software can be one of sources of information how to organize the company.

      1. Jon Reed says:

        Alex thanks for that. I probably could have gone into more detail on what I perceive as SaaS ERP benfits but the article was already getting pretty long 🙂

        I would agree that young companies can find the structure of a SaaS implementation process beneficial in terms of structuring their business process. I’d go so far as to say that even larger companies can benefit – IF that particular SaaS provider is rich in functionality and processes for their industry, and if the consulting partner is well-versed in the best industry configuration options.

        From there it an be a very helpful exercise to impose discipline around going vanilla for more standard processes, configuring nuances in processes where needed, and then thinking about areas where they may some true differentiation. In SaaS that might be accomplished through some kind of extensibility framework, of having a partner build an app or service on the platform, or even tweak an existing app in the app store. That way the differentiations are both limited and more easily updateable – with little or no custom code to worry about.

        To me that’s a big part of the value prop of modern ERP, but again, not all customers have the exact same needs so I try to keep an open mind…

        – Jon

    2. Mr. SaaS says:

      Great article Jon. I think an area that was missed though was the upgrade dilemma with Cloud ERP. Sure someone else has to maintain the infrastructure and ‘maintenance’ but customizations WILL require a significant amount of testing when a new version is available. The more customization, the more testing and we all know, the more code fixes (and retesting) will be necessary to ensure that past enhancements work with new functionality. For the end user, they really do not care that the solution is in the ‘cloud’ (i.e. hosted). What they care about is whether the software works when they try to do their work. That is why SaaS is so superior in that, code customizations DO NOT happen. Changes are made via ‘configuration’ within a framework that the vendor ensures works when EVERYONE is migrated to the next version. And when everyone is migrated to the same version, you now have a single community of users that a user can collaborate with and share ideas with which provides for a better overall experience.

      1. Jon Reed says:

        Great comment. And yeah – like I said to the last commenter, this article was plenty long enough. But yes, the difficulty upgrading is one of the main reasons I am anti-customization and one of the big reasons excessive customization is frequently tied to project failure and/or problems as Eric referred to. So, my issue with customizations is largely based on what you just articulated. Thanks, that adds to the article to have this here.

        – Jon

    3. Strongly feel threats of over customization should be overblown, alas, it is not. Now expired NDA that quite a few Microsoft accounting/ERP ISV will recognize is “Project Green”. Where high bandwidth, high firepower Microsoft architect’s worked out how to support upgradeable customizations. Makes the very successful App exchange look underpowered by comparison. My main point is ERP software industry as whole is not investing in software innovation at the same level as Facebook, Google do. Living on past glory. Someone will clean-up. Microsoft, apparently has dropped the ball (but you never know). Frank Scavo has some Project Green intelligence on his site. Some how the patents came up during Oracle’s purchase of PeopleSoft (Google search).

      1. Jon Reed says:

        “My main point is ERP software industry as whole is not investing in software innovation at the same level as Facebook, Google do. Living on past glory”

        SAP, Infor and Oracle would all vocally and strongly disagree with you. 🙂 I think the answer lies somewhere in the middle. Keep in mind that Facebook and Google have no legacy customers and in fact both companies feel free to cancel programs users love at will. You can’t do that with your ERP install base.

        I would phrase it as: the big ERP players are all making high investment gambles on next gen product lines, but customers are still deciding if those new offerings are the innovations they are looking for.

        I also really enjoyed my time at Intacct and Acumatica shows – not all ERP players are the same. And in Microsoft’s defense they are doing cool stuff with Azure.

        We’ll see. I hear your point though. Thanks for backing me up on the high costs of overcustomization.

        – Jon

        1. says:

          I strongly agree with @Jon Reeds point..

    4. Jon,

      Interesting observations on the Cloud ERP topic but I wonder if the conclusions are skewed based on the midmarket manufacturing focus that the survey respondees have.

      Having said that I am seeing the continuation in SME’s basing most of their decision making on the application fit, first and foremost and then the deployment and licensing model. Not to say that these don’t weigh heavily on the mind of the selection teams or individuals but they are still the secondary concern and appear to be the icing on the cake as they make the solution easier to deploy and more affordable.

      This is where the multi cloud model that Acumatica espouse makes sense and even the Hybrid approach presented by SAP with SAP Business One and partners like Enterpryze (who I now work for) is also alive and well for companies that have made major investments in their ERP solutions already.

      This is proving to be an area where there’s a big opportunity to build solutions that ease the transition to a cloud deployment model by taking specific functional set and deploying through apps and alternative web clients to the core solutions – this opportunity exists for the vendors and ISV’s however it appears that the ISV’s are more agile and better able to address this.

      Of course, in an API driven world, the issue of code based customisation does seem to become a lesser concern for companies deploying these solutions – as long as the API’s are contract based like those based on RESTful services and not just a middleware layer on an outdated COM based stack.

      And as a further note, Acumatica (and MYOB Advanced here in Australia) have a great approach with this – having done multiple deployment myself now, the fact that a lot of the required adaptation for SME’s is around business process, reporting and ecosystem integration – the built in “code free” automation, UI customisation and reporting tools are making the customisation issue less prevalent.

      1. Jon Reed says:

        “Interesting observations on the Cloud ERP topic but I wonder if the conclusions are skewed based on the midmarket manufacturing focus that the survey respondees have.”

        Thanks Richard for your comment, which adds some variable scenarios. I didn’t draw conclusions based on Kimberling’s data, what I did was share his perspective and offer my own, which is biased towards multi-tenant SaaS, of the more flexible variety I referred to in the conclusion.

        “Of course, in an API driven world, the issue of code based customisation does seem to become a lesser concern for companies deploying these solutions”

        Sure, it’s less of a concern, but a far better approach is the extensibility companies like Acumatica offer to make changes, but on top of the core code without impacting the core code and making future upgrades problematic. Acumatica happens to call this “customization,” but it’s entirely different than the customization I”m critiquing here. I tend to call it “extensibility” to distinguish it though in some cases like adding or changing fields the customization word is more accurate as long as we understand the differences.

        I got into that further in my piece and podcast with Acumatica’s Jon Roskill on multi-cloud, which you may enjoy 🙂

        – Jon

        1. Thanks Jon

          My point on the skewing was more about the Panorama report than your observations…my language wasn’t clear on that point sorry

    5. Randy Martin says:

      I think the entire article boils down to this statement

      Even with our own clients, we’re seeing a lot of companies that are saying, ‘We want the best of both worlds, we don’t want the whole multi-tenant SaaS thing. We want the flexibility to do whatever we want to with the software. But: we don’t want to deal with all the crap that comes along with on-premises solutions, so let’s go with cloud, where we own the licenses. We can do whatever we want. We can bastardize it if we want – for better or worse – and we can have someone else deal with the headache of managing the infrastructure behind it, and the application support

      However, in my opinion, moving from on premise to hosted while still preserving the ability to customize isn’t really an evolution at all, minor at best. The pain and headaches we have are not caring for the infrastructure, it’s in the maintenance of an increasingly complicated application landscape because of the “Layers” of customization. Hosting doesn’t really address that. There is a price to “We can do whatever we want” and “We want the flexibility to do whatever we want with the software”. Trust me, it pains me to say that because large parts of my experience and career have been built on the mantra of “We can make this system do whatever you need”. It seems clear to me that in the push to minimize ERP cost companies will be have standardize their processes around Saas offerings. The key will be evaluating the particular SaaS solution to ensure it is robust enough to meet the need.

      The other big key to the SaaS model of Cloud is the elasticity of demand. Paying for what you use and growing/contracting as you need. Now what we don’t know just yet is if the licensing models are delivering on that. Some software companies are still stuck in the old licensing model where you pay for X number of licenses whether you use them or not.