Enterprise Applications

ERP System: What It Is, What It Covers, and How to Choose One by Capability

An ERP system promises one version of the truth across finance, supply chain and operations. Whether it delivers depends less on the software than on deciding, before selection, which parts of the business should run the standard way and which should not.

By Capstera Business Architecture Staff · Updated

15 min read

ERP stands for enterprise resource planning. An ERP system is a suite of integrated business applications that runs core processes, typically finance, procurement, inventory, manufacturing, order management and often human resources, on one shared database, so a transaction entered once updates every function that depends on it.

In a business without an integrated system, a single customer order can be typed four times: into the sales system, the warehouse system, the invoicing system and, at month end, a spreadsheet that reconciles the other three. Each copy drifts. Finance and operations argue about whose numbers are right, and both are partly correct. An ERP system's original promise was to end that: one order, entered once, flowing through stock allocation, picking, shipping, invoicing and the general ledger without being re-keyed. That promise is real, and ERP systems now sit at the center of most mid-sized and large organizations. They are also among the largest and riskiest technology programs those organizations undertake, and the reasons they go wrong are mostly business decisions, not software defects.

Key Takeaways

  • An ERP system integrates core business processes on one database, so data is entered once and shared across functions.
  • Modules map to end-to-end value streams such as order-to-cash, procure-to-pay and record-to-report, not to departments.
  • ERP is strongest for capabilities the business should run the standard way; competitive differentiation often lives in specialized systems around it.
  • Decide which capabilities are differentiating before selection, using a capability heat map. It is the most useful input to scope, fit-gap and customization decisions.
  • Most ERP problems are business problems: unclear scope, process redesign skipped, customization used to preserve old habits, and data migration underestimated.
  • Cloud ERP shifts spending from capital to operating budgets and pushes organizations toward standard processes, which is a feature when the capability is not differentiating.

What Does ERP Mean?

ERP means enterprise resource planning. The name is historical; today's systems do much more than plan resources.

The lineage starts in manufacturing. Material requirements planning (MRP) systems, which spread from the 1960s and 1970s, calculated what materials a factory needed and when. Manufacturing resource planning (MRP II), in the 1980s, extended that to capacity, shop-floor control and the related finances. Gartner is generally credited with coining "enterprise resource planning" in 1990 for the next step: integrated systems that covered finance, distribution and human resources alongside production, for any kind of enterprise, not only factories.

What defines an ERP system is integration rather than any particular function. Modules share master data, such as customers, suppliers, products, employees and the chart of accounts, and transactions flow between them automatically. A goods receipt updates inventory, matches the supplier invoice and posts the liability to accounts payable without anyone keying it twice. That shared data model is what gives finance and operations the same numbers, and it is also what makes changing an ERP system harder than changing a stand-alone application.

What Does an ERP System Do? Modules and Value Streams

ERP vendors sell modules, but the work flows through them end to end. Reading the modules as value streams is the clearest way to see what an ERP system actually runs.

The same process names recur across vendors and consultancies: order-to-cash, procure-to-pay, record-to-report, plan-to-produce, hire-to-retire. Each is a value stream, a sequence of stages from a trigger to an outcome, and each draws on several modules. That is why an ERP implementation organized by module, with a separate team for each, can deliver modules that work and value streams that do not.

The table below shows the common value streams, the modules they draw on, and the business capabilities involved. Coverage varies by vendor and edition; check specifics against each product's documentation.

ERP value streams, the modules they use, and the capabilities involved
Value streamFrom trigger to outcomeTypical ERP modulesBusiness capabilities involved
Order-to-cashCustomer order to payment receivedSales order management, inventory, warehouse, shipping, billing, accounts receivableOrder Management, Inventory Management, Fulfillment, Billing, Collections
Procure-to-payNeed identified to supplier paidPurchasing, supplier management, goods receipt, invoice matching, accounts payableProcurement, Supplier Management, Receiving, Accounts Payable
Plan-to-produceDemand forecast to finished goodsDemand and supply planning, MRP, production orders, quality, costingDemand Planning, Production Planning, Manufacturing, Quality Management
Record-to-reportTransaction to published financial statementsGeneral ledger, fixed assets, consolidation, financial reportingFinancial Accounting, Asset Accounting, Financial Close, Reporting
Hire-to-retireRole approved to employee departureCore HR, payroll, time, talent (often a separate HR suite)Workforce Planning, Talent Acquisition, Payroll, Employee Administration
Acquire-to-retire (assets)Asset requested to asset disposedFixed assets, maintenance, project accountingAsset Management, Maintenance Management, Project Accounting

Types of ERP Systems

Two decisions shape most ERP choices: how it is deployed, and how much of the business it is expected to cover.

Deployment. On-premises ERP runs on infrastructure the organization owns or controls, with licenses bought outright; it offers the most control and the most responsibility for upgrades. Hosted or private cloud ERP runs the same kind of software on someone else's infrastructure. Software-as-a-service (SaaS) ERP is a multi-tenant service the vendor runs and upgrades on its own schedule, typically several times a year, which limits deep customization and keeps every customer close to the standard product.

Scope. A single-suite strategy runs as much as possible on one vendor's ERP. A composable or best-of-breed strategy keeps a core ERP, usually finance and supply chain, and connects specialized applications around it for areas where a specialist product fits better. Neither is right in general. The answer depends on which capabilities the organization should run the standard way, which is a business architecture question rather than a technology one.

The vendor market is broad. Large-enterprise suites include SAP S/4HANA, Oracle Fusion Cloud ERP and Microsoft Dynamics 365; mid-market products include Oracle NetSuite, Infor, Epicor and others; and some vendors, such as Workday, are strongest in finance and human resources. Products and editions change frequently, so evaluate current offerings rather than reputations.

ERP deployment models compared
On-premisesHosted or private cloudSaaS (multi-tenant cloud)
Who runs the infrastructureThe organizationA hosting providerThe vendor
Who controls upgradesThe organizationThe organization, with the providerThe vendor, on its release schedule
Room to customizeMostConsiderableLeast; extensions through supported interfaces
Typical spending profileMostly capital up front, then supportMixedMostly operating, as subscription fees
Best suited toHighly specific requirements, strict control needsExisting customized systems moving off owned hardwareCapabilities that can run close to standard

Why ERP Programs Go Wrong

ERP programs have a reputation for overrunning, and the reasons are well understood. Almost none of them are about the software itself.

  • No agreed scope. "Replace the old system" is not a scope. Without a list of capabilities in and out, scope grows module by module.
  • Customizing to preserve old habits. Each customization feels small; together they recreate the legacy system on a new platform and make every future upgrade expensive.
  • Process redesign treated as optional. Installing new software on unchanged processes moves the old inefficiencies onto a more expensive system.
  • Data migration underestimated. Customer, supplier and product master data accumulates duplicates and gaps over years; cleaning it is slow, unglamorous and on the critical path.
  • Organized by module, not by value stream. Teams deliver working modules while the end-to-end flow, such as order-to-cash, breaks at the handoffs.
  • Business ownership missing. An ERP program led entirely by IT makes business decisions by default, usually by keeping whatever the old system did.
  • Benefits never defined. Without a business case that names the capabilities and measures that should improve, nobody can say afterward whether it worked.

How to Choose an ERP System by Capability

The most consequential ERP decision happens before any vendor is invited: deciding which capabilities should run the standard way.

Start with a capability map of the scope under consideration, at enough depth to separate, say, Procurement into sourcing, contract management, purchasing and supplier performance. Then rate each capability on two dimensions: how much it differentiates the organization in its market, and how well it performs today. The result, a capability heat map, sorts the scope into three groups.

Commodity capabilities, such as general ledger, accounts payable and payroll for most organizations, should run as close to the ERP's standard process as possible. The organization gains nothing by doing them differently, and every deviation costs money at each upgrade. Fit-to-standard workshops, which demonstrate the standard process and ask the business to justify any change, are the right method here.

Differentiating capabilities, such as a distributor's availability promising or a manufacturer's configure-to-order process, deserve strong native support or a specialist product connected to the ERP. These are the capabilities to test hardest in vendor demonstrations, with the organization's own scenarios rather than the vendor's script.

Weak but important capabilities are where the business case lives. They are the reason to change and should be named in it, with the measures that should improve.

This view also changes vendor evaluation. Instead of scoring hundreds of feature requirements equally, weight the evaluation toward the differentiating capabilities and the end-to-end value streams. A vendor that wins on general ledger features and loses on the one process that sets the business apart is the wrong choice, however its total score adds up.

  1. Map the capabilities in scope, to the level where standard and specialist needs separate.
  2. Rate each for differentiation and for current performance; publish the heat map.
  3. Decide the stance for each capability: adopt the standard, configure, extend, or keep a specialist system.
  4. Write vendor scenarios for the differentiating capabilities and the end-to-end value streams.
  5. Evaluate vendors on those scenarios first, and on feature lists second.
  6. Run fit-to-standard for commodity capabilities; require a business case for every customization.
  7. Build the program business case on the weak but important capabilities, with measures and owners.

A Worked Example: A Manufacturer Replaces Its ERP

An illustrative example. The company and its decisions are invented to show how the capability view changes an ERP program.

A mid-sized maker of industrial pumps runs a fifteen-year-old ERP system so heavily customized that the vendor no longer supports its version. The first instinct is a like-for-like replacement: take every customization, write it into the requirements, and invite vendors to match it. The requirements list runs to more than a thousand lines, each weighted equally.

The business architecture team proposes a different starting point. Its capability heat map shows that most of the scope is commodity: financial accounting, accounts payable, payroll, fixed assets and most of purchasing. Two capabilities are genuinely differentiating. Configure-to-order, which turns a customer's pump specification into a bill of materials and a price within a day, is the reason customers choose the company. Aftermarket service, which tracks every installed pump and its spare parts, is where most of the margin comes from.

The program changes shape. For the commodity capabilities, the team commits to the standard process and retires most of the old customizations through fit-to-standard workshops. For configure-to-order, vendor demonstrations must process three of the company's real specifications end to end, live. For aftermarket service, the team keeps a specialist field-service product and makes integration with the new ERP a selection criterion. The requirements list shrinks, the shortlist changes, and the business case is built on the two capabilities customers notice, with measures for quote turnaround time and spare-parts availability, rather than on replacing an unsupported system.

ERP and Business Architecture

ERP implements capabilities; it does not define them. Keeping that distinction clear is the business architect's main contribution to an ERP program.

A capability model gives the program a reference that does not depend on the old system or the current organization chart. Without one, ERP configuration tends to follow today's departments and today's workarounds, hard-wiring them into an expensive platform. With one, the program can configure toward the target operating model: which capabilities are centralized, which are shared services, which differ by region and why.

After go-live, the same model keeps the application portfolio honest. Mapping applications to the capabilities they support shows where the ERP now covers what a specialist system used to, and which systems can be retired; it also shows where several systems still support the same capability years after the ERP was meant to consolidate them.

ERP decisions are also investment decisions with a distinctive spending profile. Cloud ERP moves most of the cost from capital to operating budgets, and implementation costs are treated differently under different accounting frameworks, which can change how the options look to the people approving them.

Go further

Frequently Asked Questions

What does ERP mean?

ERP stands for enterprise resource planning. An ERP system is integrated software that runs core business processes, such as finance, procurement, inventory, manufacturing and order management, on one shared database, so information entered once is available to every function that needs it.

What is an ERP system used for?

To run and record the core transactions of a business: taking and fulfilling orders, buying and paying suppliers, managing inventory and production, paying employees, and closing the books. Because all of these share one database, management sees consistent figures across functions.

What are the main modules of an ERP system?

Financial accounting, procurement, inventory and warehouse management, sales order management, manufacturing and planning, and often human resources and payroll. Together they support value streams such as order-to-cash, procure-to-pay and record-to-report.

What is the difference between ERP and CRM?

ERP runs back-office operations: orders, inventory, purchasing, production and finance. CRM manages customer engagement: leads, opportunities, marketing and service interactions. Most organizations run both and integrate them, so an order won in CRM flows into ERP for fulfillment and billing.

What are examples of ERP systems?

SAP S/4HANA, Oracle Fusion Cloud ERP, Oracle NetSuite, Microsoft Dynamics 365, Infor and Epicor are widely used examples. Some vendors, such as Workday, focus on finance and human resources. Product names and editions change often, so evaluate current offerings.

Is cloud ERP better than on-premises ERP?

Neither is better in general. SaaS ERP shifts spending to subscriptions, keeps the system current through vendor-managed upgrades and limits customization. On-premises ERP gives more control and more responsibility. SaaS suits capabilities that can run close to standard; deeply specialized requirements may need more flexibility.

How do you choose an ERP system?

Start with a capability map of the scope and rate each capability for how much it differentiates the business and how well it performs today. Adopt the standard process for commodity capabilities, test vendors hardest on differentiating capabilities with your own scenarios, and build the business case on the important capabilities that perform poorly today.

Why do ERP implementations fail?

Mostly for business reasons: unclear scope, customizing the new system to preserve old processes, skipping process redesign, underestimating data migration, organizing the program by module rather than end-to-end value stream, and not defining the benefits that should result.

Pro Tips

  • Before writing requirements, publish a capability heat map of the scope and get the business to agree which capabilities are differentiating. It settles most later arguments about customization.
  • Ask vendors to run your end-to-end scenarios, such as order-to-cash for your most complex order, not their scripted demonstration.
  • Start master data cleanup the day the program is approved. It is almost always on the critical path, and it does not depend on the vendor choice.
  • Require a short business case for every customization: which capability it serves, why the standard process will not do, and what it will cost at each upgrade.
  • Organize testing and go-live readiness by value stream, not by module, so the handoffs are tested as hard as the screens.