Internal tools, B2B portals, workflow automation and the integration work that connects them. Enterprise software rarely fails on features: it fails on adoption, permissions and the legacy system nobody wants to touch.
Approvals, routing and process orchestration
Self-service with audience-scoped content
Operations dashboards and admin consoles
Connecting ERP, CRM and legacy databases
Agents embedded in real business workflows
Generation, approval and retention
Offline-capable tools for staff off the desk
Multi-language, multi-region deployments
SAML, OIDC and group sync against the existing identity provider
Permissions that survive reorganisations
Talking to systems that predate REST, without destabilising them
Who changed what, when: recorded, not reconstructed
The tool people actually use beats the more capable one they avoid
Where data physically lives, when that is contractual
Cutting over without a weekend of downtime
Language, RTL and regional formatting handled structurally
Ranges, not quotes. Scope, integrations and compliance work move these, so the number we give you after a discovery call is the one that means something.
A single-workflow internal tool with SSO runs $60,000-$110,000. A business platform spanning several workflows and system integrations runs $110,000-$220,000. A programme replacing a legacy system, or operating multi-region, runs $220,000-$400,000+. Integration count and the state of the systems you are integrating with drive the range: a documented REST API is a fraction of the cost of an undocumented legacy database.
That is usually most of the work. Salesforce, SAP, NetSuite, Workday and similar have documented APIs; older internal systems often mean direct database access, SOAP, or a scheduled file exchange. We audit what actually exists during discovery (including whether the system can take the additional load) because an integration approach chosen on assumption is the most expensive kind of mistake here.
Yes: SAML 2.0 and OIDC against Okta, Microsoft Entra ID or an internal provider, with SCIM provisioning where you want accounts and group membership to flow automatically. Tying permissions to existing directory groups rather than maintaining a second list of users inside the application is usually worth the extra setup.
Prototype with the people who will use it before the build, pilot with one team before rolling out broadly, and treat low adoption as a design defect rather than a training problem. Internal tools fail far more often on workflow mismatch than on missing features. If the tool is slower than the spreadsheet it replaces, it loses, regardless of what it can do.
Three to five months for an internal tool, six to ten for a business platform, ten to eighteen for a full programme. Data migration and the pilot phase are the parts most often underestimated; both reward being run properly rather than quickly.
Tell us what you are trying to ship and we will tell you what it takes — scope, timeline and the parts that usually go wrong.