A Fortune 50 client requested a complete NetSuite ERP system in less than five months. Not many of the program staff believed it was possible and that was a fair assumption. The lowest estimate for enterprise ERP implementation is 6 months but it can last up to 18 months. The average is around 14 months, and there are large multi-national programmes that can be multi-year projects.
We were live on time. The first closing at the end of the month was within the new system, while the program was completed within budget. This article details how that happened, and some lessons for other finance leaders.
Why Most ERP Implementations Slip
ERP project failure facts are troubling. Industry researchers such as Panorama Consulting always report that most ERP projects go over budget and over time. The top two reasons are scope expansion and poor planning. According to a NetSuite study of the market, the first attempt failure rate is about 50 percent. Many projects are significantly over budget and are up to half a month late.
By the end of the first sprint, the slip typically will become noticeable. It's just after month four or five, when the program team is finally ready to admit it.
This was something I experienced with a previous Finance IMO engagement. We were connecting Salesforce, NetSuite and an HR application called UKG. The Salesforce and UKG parts were delivered by third party vendors. In two sprints it was evident that those teams were much bigger in scope than the time frame. Any design decision made in an upstream application alters the design of the ERP.
I immediately became a client leader and put all parties on a phone call. Final go-live design agreed, scope changes limited, front-end changes would not have an impact on the back-end agreed. That's when the problem was raised early – and calmly – that saved that program.
Where ERP Program Time Actually Goes
Traditional ERP deployment plans are executed in a linear fashion.Traditional ERP deployment processes follow a linear path. Requirements give rise to design, design gives rise to configuration and configuration gives rise to testing. Each handoff has a buffer as the receiving team wants to review past decisions. Those buffers contain approximately three months when it is paid monthly for a 12-month plan.
Instead, we re-built the governance over parallel work streams:
- Requirements validation and chart-of-accounts design went hand in hand.
- While business process mapping was still ongoing for workstreams which did not feed into the integrations, integration build began.
- There was no month 4, planning started in week 3.
All this is not new theory. It is feasible due to the speed of decision making. The IMO was able to solve cross-workstream issues within hours, rather than weeks.
The Escalation Path That Kept Decisions Moving
Every issue was rated based on severity from 1 to 5, 1 being the highest. The pace was set:
- The IMO committee held two meetings per week, to be held on Tuesdays and Thursdays.
- Executive steering committee meetings were held on Fridays.
Each issue was referred to the IMO Committee first and a Resolution path was drafted on the call. Approximately 90% of the problems were fixed on the spot. Those issues that had no clear solutions were moved up to the steering committee (SC).
This enabled leadership to keep focused on what was really important that required executive attention. It also provided the actual discussion time on those issues and not an agenda that was too crowded. Tickets that were given the highest priority were resolved the quickest.
Scope Discipline in the First 30 Days
The most common reason for ERP overruns is by scope expansion and this typically happens in the first month, when users are still refining the scope of the system. After the customization is added to the build backlog, it takes political capital to take it off, which few program leads want to invest.
This engagement involved a fit-gap analysis which took around 3 weeks to complete. There was one thing that determined the result. Any gap needed to be addressed with process change, manual process, or deferred phase-two enhancement. Everything else was absorbed in the standard configuration. That was the most important governance ruling made for the timeline.
It also involved having open discussions. Changing a reconciliation process to fit into the system, vs. Customizing the system to support a legacy process, requires executive support. If you can't customize it, it's the easy way to go—and customization is where timelines go wrong.
Reporting was clearly a good example of that. For complex reporting, most ERP systems require a lot of customization. Early sprints revealed that there were a number of reporting requirements which would require just that. I suggested that the client leadership should use a data warehouse tool and Power BI. The ERP data was ingested into the warehouse and processed into the desired reports and dashboards. All reporting requirements were on time and NetSuite was close to standard.
Manual Workarounds Are a Feature, Not a Failure
Short-term manual workarounds are needed for accelerated ERP go-lives. This is a trade-off, not a bug. The danger is that short-term solutions become long-term solutions. That's what makes a successful go live a finance team still conducting three extra offline reconciliations per month, two years later.
The fix comes as a documented post go-live roadmap agreed prior to cutover. This NetSuite engagement had all the workarounds assigned a sunset date and named owner. The list was reviewed quarterly by the steering committee for the first year. Everyone was aware of its passing, its perpetrator, and its end date.
What CFOs Should Ask Before an ERP Program
I’m not working with the CFOs who want a quicker timeline, in and of itself. They seek synergy capture in a carve-out, Day 1 readiness in an acquisition, or to unburden a legacy system that does not allow them to do what they want. The importance of compression is that there are true costs to the extra month of runtime. These range from consulting costs, lost synergies, and finance firm time diverted from business.
Then, the question is not if an ERP can be live in five months. Most (if not all) programs shouldn't, and most don't. So the question is, does the governance structure promote three elements?
- Parallellism: Workstreams that move together, rather than waiting in line.
- A strong fit-gap rule from the beginning (scope discipline).
- The speed of decision making – the clear escalation path with real authority.
If those three factors have been taken care of, the following timeline ensues. If they are not there, it's not something that can be made up by additional resources.
