Statistics · Updated October 11, 2026
ERP Implementation Statistics (2026): Cost, Timelines, Failure Rates and S/4HANA Progress
Gartner predicts more than 70% of recently implemented ERP initiatives will fail to fully meet their original business case goals by 2027.
Compiled by Matt Angerer · 28 sources · 14 min read

ERP implementation statistics are easy to find and hard to trust. The same "failure rate" gets passed from blog to blog without a sample, a year or a definition. This page collects the numbers we could trace back to a named study, gives each one its population and date, and shows where credible sources disagree. The aim is to help the people who sign off on SAP and ERP releases build plans on evidence rather than folklore.
Key takeaways
of recently implemented ERP initiatives will fail to fully meet their original business case goals by 2027 (Gartner prediction)
Source: Gartner, 2024 research
of ERP projects in Panorama's 2026 sample (n=170) went over budget
of DSAG respondents still on ECC plan to be on S/4HANA by the end of 2027
Source: DSAG Investment Report, 2026
of organizations have scaled Gen AI in quality engineering enterprise-wide
Source: World Quality Report, 2025
ERP adoption and market momentum
ERP is close to universal in large organizations, but much less common among smaller ones. On the SAP side, contracted cloud revenue is still growing fast, which tells you how many migration and RISE programs are signed and still have to be delivered.
- 46.45% of EU enterprises with at least 10 workers used ERP software in 2025. The figure was 41.08% for small firms (10–49 workers) and 88.71% for large firms (250+ workers). (Eurostat 2025 data, compiled by Business-Software.com)
- SAP's current cloud backlog reached €22.9 billion at the end of Q2 2026, up 27% year over year (26% at constant currencies). Cloud ERP Suite revenue grew 25%. (SAP Quarterly Statement Q2 2026)
- SAP expects €25.8–26.2 billion in cloud revenue for 2026 at constant currencies, compared with €21.02 billion in 2025. (SAP Quarterly Statement Q2 2026)
Source: Eurostat 2025 ERP usage data, compiled by Business-Software.com, 2026
Source: SAP current cloud backlog, SAP Quarterly Statement Q2 2026 (July 2026) and SAP Quarterly Statement Q1 2026 (Form 6-K)
S/4HANA migration progress
The two big SAP user groups show steady progress, with a long tail. Their surveys ask different questions of different populations, so treat the figures below as separate snapshots rather than one series.
- The DSAG Investment Report 2026 surveyed 198 member companies in Germany, Austria and Switzerland in December 2025 and January 2026. Note that SAP, not DSAG, ran the 2025 questionnaire, and the 2026 sample includes fewer large companies than that one did. (IT-Matchmaker on DSAG, 2026)
- 56% of DSAG respondents now run S/4HANA on-premises, up from 44% in 2024. For the first time that is ahead of ECC and the old Business Suite, which fell to 54% from 68%. (DSAG Investment Report 2026, via All About Security)
- Among DSAG respondents still on ECC, 37% aim to switch by the end of 2027, about half by the end of 2030 (accepting extended-maintenance surcharges), and 4% only by 2033. (DSAG Investment Report 2026, via All About Security)
- In ASUG's 2026 Pulse of the SAP Customer survey, a record 55% of organizations reported having deployed S/4HANA or S/4HANA Cloud, but only 34% said they had fully completed the transition. (ASUG 2026 Pulse, as reported by ITChamps, an SAP partner)
- Separate ASUG S/4HANA research released in September 2026 reports 68% adoption across the ASUG community, with value realized in an average of 6.9 months. (ASUG Talks, September 2026)
- For comparison, in the 2024 ASUG Pulse 47% of members were using S/4HANA or starting implementation, and 69% expected to have implemented it within two years. (ASUG 2024 Pulse of the SAP Customer)
Source: ERP systems in use among DSAG respondents (multiple answers possible). All About Security on DSAG, 2026; Security Storage und Channel Germany on DSAG, 2026
Source: ASUG 2026 Pulse of the SAP Customer, as reported by ITChamps, 2026
The 21-point gap between "deployed" and "fully completed" is where test managers spend their time. These are organizations running S/4HANA and legacy environments side by side while data migration, custom code remediation and process cutover continue after the first go-live.
Implementation timelines
Panorama's numbers suggest mid-market ERP projects have become much shorter. Two caveats apply: the 2024 and 2025 figures are averages and the 2026 figure is a median, and Panorama's respondents are mid-sized organizations rather than global SAP estates.
- Panorama's 2026 ERP Report covers 170 organizations surveyed from January 2025 to January 2026. Median annual revenue was $200.5 million, 56.5% were multinational, and the median project timeline was 9 months. (Panorama 2026 ERP Report)
- Panorama's 2025 report found the average project timeline fell from 15.5 months to nine months compared with the previous year's report, which Panorama links to high SaaS adoption. (Panorama press release, March 2025)
- 22.3% of projects in Panorama's 2026 sample finished late. The leading cause of schedule overruns was organizational issues, not technology. (XoroSoft on Panorama 2026; KORE1 on Panorama 2026)
- Panorama also points out that technical work can run to plan while approvals, sign-offs and cross-functional alignment fall behind, which stretches timelines even when scope stays the same. (Panorama 2026 ERP Report)
- Historical: Panorama's 2015 report (562 organizations) recorded an average duration of 15.7 months, and 75% of projects overran their timeline. (Panorama 2015 ERP Report)
Source: Panorama press release, March 2025; Panorama 2026 ERP Report, March 2026
Budgets and cost overruns
Overrun rates depend heavily on who is being measured. Panorama's mostly mid-market sample shows a minority of projects going over budget. McKinsey's work on large technology programs shows a large majority overrunning. Both can be right.
- Panorama 2026: 50.6% of projects finished on budget, 19.4% came in under, 22.9% finished slightly over and 7.1% well over. That makes 30% over budget in total. (XoroSoft on Panorama 2026) Panorama's own release describes this as "more than a quarter" and names additional technology needs as the leading cause. (Panorama, March 2026)
- McKinsey 2026: According to McKinsey ERP cost benchmarks, most large enterprises spend $100 million to $1 billion to migrate their ERP systems, and one-off implementation costs usually come to a multiple of the subscription cost. (McKinsey, 2026)
- McKinsey/Oxford: 65–80% of large tech programs exceed their planned budget or timeline. (McKinsey, 2026)
- Historical: In McKinsey and Oxford's study of IT projects over $15 million, the average project ran 45% over budget and 7% over schedule and delivered 56% less benefit than planned. (McKinsey, 2012)
- Historical: 55% of respondents in Panorama's 2015 report exceeded their budgets. (Panorama 2015 ERP Report)
Source: Panorama 2026 ERP Report (n=170), as reported by XoroSoft, 2026
Several of the causes Panorama found behind overruns are things a test program can catch early: technical problems, data issues and scope that grows after fit-gap.
Source: Causes cited by over-budget organizations (multiple answers). Panorama 2026 ERP Report, as reported by XoroSoft, 2026
Data problems showed up even more often in late projects: 36.8% of organizations that ran late cited data issues. (XoroSoft on Panorama 2026)
Failure, value and the business case
There is no single ERP failure rate. Published figures run from roughly one in five to more than seven in ten, because each study defines "failure" differently.
- Gartner: By 2027, more than 70% of recently implemented ERP initiatives will fail to fully meet their original business case goals, and as many as 25% of those will fail catastrophically. (Gartner; Gartner, 2024 research)
- Gartner: 75% of ERP strategies are not strongly aligned with overall business strategy, according to a Gartner survey. (Gartner)
- McKinsey: Only 25–35% of large tech programs achieve their targeted EBITDA and cash-flow impact. (McKinsey, 2026)
- Prosci: Prosci's 2025 ERP study defines failure as delivering less than 70% of expected benefits. On that basis it reports failure rates of 11–31%, averaging close to one in five. (Prosci 2025, as reported by XoroSoft)
- Panorama 2026: Among eligible respondents, 87.3% realized the productivity benefits they expected, but only 56.3% realized the expected inventory benefits. (XoroSoft on Panorama 2026)
- Historical: McKinsey reported in 2019 that three-quarters of ERP transformations failed to stay on schedule or within budget, and two-thirds had a negative return on investment. (McKinsey, 2019)
How to read the conflict: Gartner and McKinsey measure against the original business case and focus on large programs. Prosci uses a benefits threshold. Panorama asks mostly mid-market organizations about specific benefit areas. "73% of ERP projects fail" is often attributed to Panorama, but we could not trace it to any current Panorama report, so we left it out.
Deployment models and investment plans
Cloud dominates new mid-market ERP selections. Among large SAP customers in German-speaking Europe, on-premises and private cloud still get most of the money.
- 73.5% of ERP projects in Panorama's 2026 cross-industry survey used cloud deployment. (XoroSoft on Panorama 2026)
- In the 2024 ASUG Pulse, the 62% of members using or planning S/4HANA in the cloud split as follows: 40% private cloud, 16% managed cloud and 6% public cloud. (ASUG 2024 Pulse)
- IT budgets are rising at 38% of DSAG companies, down from 43% two years earlier. (Inuit on DSAG, 2026) 43% plan higher SAP software spend (46% two years earlier), and the share reporting a declining SAP budget rose from 19% to 28%. (All About Security on DSAG, 2026)
Source: Share of DSAG respondents planning high or medium 2026 investment, by ERP option. DSAG press release, February 2026
Testing and quality engineering
No current independent study isolates testing effort on SAP programs, so we use the broader World Quality Report 2025 (17th edition, published November 2025 by OpenText, Capgemini and Sogeti). Its findings line up with the overrun causes above, especially on test data.
- 60% of organizations struggle with secure, scalable test data, and 58% report difficulty adopting AI-powered tools. (World Quality Report 2025-26)
- Use of synthetic data in testing grew from 14% in 2024 to an average of 25% in 2025, and it ranks as the top Gen AI use case. (World Quality Report 2025-26)
- 94% of organizations review production data, yet nearly half struggle to turn those insights into quality strategies. (World Quality Report 2025-26)
- 89% are piloting or deploying Gen AI-augmented QE workflows (37% in production, 52% in pilot), but only 15% have scaled enterprise-wide. (Capgemini/OpenText, November 2025)
- Organizations report an average 19% productivity gain from Gen AI in QE, but one-third have seen minimal gains. (Capgemini/OpenText, November 2025)
Source: World Quality Report 2025-26, Capgemini/OpenText/Sogeti, November 2025
Adoption is leveling off. The share of organizations not using Gen AI in QE at all fell sharply between 2023 and 2024, then rose again in 2025.
Source: Gen AI non-adopters in quality engineering. World Quality Report 2025 press release, November 2025
People, skills and change management
- 36% of ASUG member organizations say keeping the internal skills needed to manage new technology is a significant challenge. Another 22% struggle to keep knowledgeable staff through turnover. (ASUG 2026 Pulse of the SAP Customer, via ASUG, 2026)
- 59% of ASUG members name S/4HANA expertise as the most important skill for their organization's success. (ASUG, 2026)
- Budget constraints are now the top challenge for SAP customers, cited by 61% of 2026 ASUG Pulse respondents. That is 7 points higher than the year before. (ASUG 2026 Pulse, as reported by ITChamps)
- Fewer than 1 in 4 organizations in Panorama's 2026 sample gave organizational change management an intense focus. (KORE1 on Panorama 2026)
- Quality engineers rank Gen AI as their top skill (63%), just ahead of core QE skills (60%). (World Quality Report 2025-26)
Source: ASUG 2026 Pulse of the SAP Customer. ASUG, 2026 (skills figures); ITChamps, 2026 (budget figure)
AI in ERP programs
AI is being added on top of transformations that are already underway, not replacing them. Most organizations are still piloting.
- A February 2026 ASUG survey of 142 members found 41% actively piloting AI, 39% still building foundational knowledge and 24% in active deployment. (ASUG, as reported by ITChamps)
- 43% of DSAG companies already run AI use cases in production, mostly outside the SAP core. (IT-Onlinemagazin on DSAG, February 2026)
- Panorama's 2026 report found business intelligence was the most significantly deployed digital initiative, at 55.3% of organizations. (Panorama, March 2026)
- The top barriers to scaling Gen AI in quality engineering in 2025 were data privacy risks, integration complexity and hallucination or reliability concerns. Half of respondents also lack AI/ML expertise. (Capgemini/OpenText, November 2025)
Source: Top challenges to scaling Gen AI in QE. World Quality Report 2025 press release, November 2025
What this means for your sign-off
- Plan for the business case, not just go-live. Gartner's 70%-plus prediction is about missing the original business case, not about systems failing to start. Make sure your exit criteria trace back to the benefits the program promised, such as inventory, close time or order cycle, and not only to defect counts. Panorama's 87.3% versus 56.3% gap between productivity and inventory benefits shows where value tends to slip.
- Treat data as a test stream in its own right. Data issues appear in 29.4% of over-budget and 36.8% of late Panorama projects, and 60% of World Quality Report respondents struggle with secure, scalable test data. Budget for several mock loads and reconciliation cycles. Our guide to test data for S/4HANA covers copies, masking and synthetic data.
- Size your plan for a long hybrid period. Only 34% of ASUG organizations have fully completed the S/4HANA transition, and about half of DSAG's ECC users are aiming for 2030. Expect regression testing across ECC, S/4HANA and integrations for years, not months. See what 2027, 2030 and 2033 really mean.
- Protect sign-off time in the schedule. Panorama traces schedule overruns mainly to organizational issues, including approvals and sign-offs that lag technical work. Put UAT and go/no-go decision dates in the baseline plan with named owners. The template library includes a UAT Sign-Off Form and a Go/No-Go Readiness Checklist.
- Govern AI in testing before you scale it. With only 15% of organizations scaling Gen AI in QE and 60% worried about hallucinations, set review gates for AI-generated tests now. Our AI-DLC guide for ERP teams covers how to do this.
Frequently asked questions
What percentage of ERP implementations fail?
It depends on the definition. Gartner predicts that more than 70% of recently implemented ERP initiatives will fail to fully meet their original business case goals by 2027, with up to 25% failing catastrophically. Prosci, which counts a project as failed if it delivers less than 70% of expected benefits, reports 11–31% (as reported by XoroSoft).
How long does an ERP implementation take?
The median project in Panorama's 2026 sample of 170 mostly mid-market organizations took 9 months (Panorama, 2026), and 22.3% finished late (XoroSoft on Panorama). Large multi-country SAP programs usually take much longer and run in waves.
How often do ERP projects go over budget?
30% of Panorama's 2026 respondents went over budget (XoroSoft on Panorama). For large technology programs, McKinsey puts the share that exceed planned budget or timeline at 65–80% (McKinsey, 2026).
How much does a large ERP migration cost?
McKinsey's ERP cost benchmarks show most large enterprises spending $100 million to $1 billion to migrate their ERP systems, with one-off implementation costs usually a multiple of the subscription cost (McKinsey, 2026).
How many SAP customers are still on ECC?
Among 198 DSAG member companies surveyed in December 2025 and January 2026, 54% still use ECC or the old Business Suite, down from 68% in 2024. 37% of those still on ECC plan to switch by the end of 2027 (DSAG Investment Report 2026).
Are testing teams using generative AI on ERP programs?
Mostly in pilots. The World Quality Report 2025 found 89% of organizations piloting or deploying Gen AI in quality engineering but only 15% at enterprise scale (Capgemini/OpenText, 2025).
Sources
- Quarterly Statement Q2 2026, SAP SE, July 23, 2026
- Quarterly Statement Q1 2026 (Form 6-K, Exhibit 99.1), SAP SE / SEC, April 2026
- DSAG Investment Report 2026: Companies are investing more selectively, DSAG, February 26, 2026
- DSAG-Investitionsreport 2026: SAP bleibt gefragt, All About Security, 2026
- SAP Investments 2026: Targeted Spending, Security Storage und Channel Germany, 2026
- DSAG: Mehr Budget, mehr S/4HANA, IT-Matchmaker News, 2026
- DSAG-Investitionsreport: KI etabliert sich, Cloud auf dem Prüfstand, Inuit, February 2026
- DSAG-Investitionsreport 2026: KI wird zur strategischen Bewährungsprobe für SAP, IT-Onlinemagazin, February 2026
- 2024 Pulse of the SAP Customer Research, ASUG, 2024
- ASUG Campus Connect Brings Career Opportunities to University Students (cites the 2026 Pulse), ASUG, 2026
- ASUG 2026 SAP S/4HANA Research Reveals 68% Community Adoption and Value Realized in 6.9 Months, ASUG Talks, September 2026
- SAP AI in 2027: What Enterprises That Migrated Early Are Now Doing Differently (reports ASUG 2026 data; SAP partner), ITChamps, 2026
- The 2026 ERP Report, Panorama Consulting Group, March 2026
- Panorama Consulting Group Releases Latest Study of ERP Implementation Outcomes, Panorama Consulting Group, March 4, 2026
- The 2025 ERP Report press release, Panorama Consulting Group via EIN Presswire, March 25, 2025
- 2015 ERP Report (historical), Panorama Consulting Solutions, 2015
- ERP Implementation Statistics 2026 (secondary report of Panorama 2026 data; ERP vendor), XoroSoft, 2026
- Manufacturing ERP Statistics 2026 (secondary; ERP vendor), XoroSoft, 2026
- ERP Implementation Failure Statistics (secondary report of Prosci data; ERP vendor), XoroSoft, 2026
- ERP AI Adoption Report 2026: Mid-Market Benchmarks (secondary report of Panorama 2026 data), KORE1, 2026
- ERP Statistics 2026 (compiles Eurostat 2025 data), Business-Software.com, 2026
- Enterprise Resource Planning to Optimize Operations, Gartner, accessed October 2026
- What IT Leaders Must Do to Avoid Disappointing ERP Initiatives, Gartner, 2024
- The end of ERP as we know it? Five ways AI is disrupting ERP, McKinsey & Company, 2026
- Agile in enterprise resource planning: A myth no more (historical), McKinsey & Company, 2019
- Delivering large-scale IT projects on time, on budget, and on value (historical), McKinsey & Company with University of Oxford, 2012
- World Quality Report 2025-26, Capgemini, Sogeti and OpenText, November 2025
- World Quality Report 2025: AI adoption surges in Quality Engineering, but enterprise-level scaling remains elusive, Capgemini/OpenText, November 13, 2025
Last updated: October 11, 2026. Statistics are refreshed as new research is published.
Free SAP testing templates
Master test strategy, UAT sign-off, cutover runbook, go/no-go checklist and more.
Get the weekly newsletter
New statistics, practitioner stories and templates, once a week.
More statistics
- AI in Software Testing Statistics (2026)
- SAP ECC End of Maintenance Statistics (2026): Deadlines, Holdouts and Migration Risk
- IT Project Failure Statistics (2026): What the Data Says About ERP, S/4HANA and AI Programs
- SAP S/4HANA Adoption Statistics (2026): 39 Sourced Numbers on Where ECC Customers Really Stand
- ECC to S/4HANA Migration Statistics (2026): Adoption, Deadlines, Cost, Risk and Testing
- ERP Vendor Satisfaction Statistics (2026): NPS, User Groups and Peer Ratings
How this page is made: statistics are gathered from published research with AI-assisted search, every figure is linked to its source, and the page is refreshed as new data is published. Found an error? Tell us.