Methodology
How DeveloperPortalCost.com sources vendor pricing, calculates Backstage self-hosted TCO, models build-vs-buy economics, and computes ROI driver math. Every number on the site should be re-derivable from a vendor's own published pricing page or a public benchmark study using the formulas shown here.
Vendor sources
Each platform page on the site cites the vendor's own published pricing page or product documentation as the source of record. Backstage cost of ownership math additionally cites the CNCF Platforms Working Group whitepaper and the Spotify engineering blog. ROI driver math cites the DORA and SPACE framework documents.
| Source | Refresh cadence | What we take from it |
|---|---|---|
| Backstage (open source) | On change | The Backstage framework is free. Our cost-of-ownership math takes the engineer-month effort estimate from the Backstage adoption studies the CNCF Platforms Working Group publishes plus the Spotify engineering blog series on internal adoption. Plugin maintenance share (30 to 40 percent of platform team time) is a published range from those sources. |
| Roadie (managed Backstage) | Monthly | Per-developer per-month pricing for managed Backstage. We treat Roadie as the canonical price-point for not running Backstage yourself; the $22/dev/month figure is taken from Roadie's own published pricing page. |
| Port | Monthly | Per-user per-month pricing for the Port internal developer portal. Public starting rate ($30+/user/month) is the headline; enterprise SSO, RBAC, and audit log tiers are gated behind contact-sales and are out of scope. |
| Cortex | Monthly | Per-user per-month pricing for Cortex's service catalog and developer portal product. Published rate $65 to $69/user/month at Cortex Cloud tier. Volume discount tiers are not published; we flag this on the platforms page. |
| OpsLevel | Monthly | Per-user per-month pricing for the OpsLevel service maturity platform. Published Teams tier starts at $39/user/month; Enterprise tier features (SSO, RBAC, audit log) sit behind contact-sales. |
| Humanitec | Quarterly | Internal developer platform pricing reference for adjacent IDP coverage. Humanitec is more an internal developer platform than a portal in the Backstage / Port / Cortex sense, but the cost economics overlap on the build-vs-buy decision and we cite Humanitec on the platforms page as a sister category. |
| Azure API Management developer portal | Quarterly | Azure API Management includes a developer portal module at Standard tier and above. We use Microsoft Learn docs as the source-of-record for what the included portal can do; pricing comes from the Azure API Management pricing page. |
| Kong Konnect / Kong Developer Portal | Quarterly | Kong Konnect Enterprise includes a developer portal module. Source-of-record for feature coverage is the Kong product documentation; pricing is gated for the enterprise tier so we treat it as an extension-cost reference rather than a per-seat figure. |
| Tyk | Quarterly | Tyk's developer portal ships with the API management platform. Self-managed Tyk is free; cloud and self-managed enterprise tiers carry per-environment fees. We cite Tyk in the api-portal-cost page as a build-on-what-you-have option. |
| Apigee (Google Cloud) | Quarterly | Google Cloud's Apigee API platform includes a developer portal in the Standard and Enterprise tiers. The Apigee pricing page on cloud.google.com is the source for the tier inclusion list; we treat Apigee's portal as an extension option for teams already paying for Apigee. |
| CNCF Platforms Working Group | On publication | Platform engineering whitepaper and survey data. We use the CNCF survey as the source for industry-typical platform team size, Backstage adoption rates, and the platform-engineering ROI ranges (185 to 220 percent) cited on the /roi page. |
| Spotify Backstage case study | Reference only | Spotify's own engineering blog series on Backstage adoption. We use this as the reference point for the 99 percent internal adoption benchmark on /backstage-cost and to anchor the realistic 10 percent industry-average adoption discussion. |
| DORA / SPACE frameworks | Annual (State of DevOps) | Source for the DORA four-keys metrics framework and the SPACE developer-productivity framework that the /roi page builds the business case template on. DORA's annual State of DevOps reports are the underlying benchmark dataset. |
In scope
- ●Published per-seat or per-developer per-month pricing from each vendor's public pricing page.
- ●Free-tier inclusions, trial windows, and self-managed open-source tiers (Tyk self-managed, Apigee self-managed Hybrid, Backstage open source).
- ●Custom-build effort estimates in engineer-months at a conservative US loaded engineering rate.
- ●Backstage self-hosted TCO using engineer-months for initial setup plus a published platform-team time-share range for ongoing plugin maintenance.
- ●API management portal extension cost (incremental engineering time to configure the included portal module, not the API gateway tier itself).
- ●ROI driver math using industry-published benchmark ranges for TTFAC reduction, ticket-volume reduction, onboarding day savings, and recovered developer productivity.
Out of scope
- ●Enterprise-negotiated pricing. Volume commitments, contract minimums, custom MSAs, and audit-log / SSO / RBAC uplift tiers gated behind contact-sales are deliberately excluded.
- ●Sales-led pricing for tiers where the vendor does not publish a per-user rate (Port Enterprise, Cortex Enterprise, OpsLevel Enterprise, Humanitec custom).
- ●Cloud regional surcharges. Azure API Management, Apigee, and Kong Konnect pricing varies by region; the comparison uses the US / generic SaaS list price unless flagged.
- ●Compliance add-ons: SOC 2 audit-pack pricing, FedRAMP overlays, HIPAA BAA premiums on the portal SKU.
- ●Internal loaded-engineer-cost estimates. We use a published US loaded engineering rate as a conservative anchor; teams should substitute their own loaded rate in the calculator.
- ●Migration costs from one developer portal vendor to another. Lock-in cost is mentioned on /ongoing-costs as a risk factor but is not modelled per-vendor.
Calculation framework
Engineer-month effort
Custom-build effort uses an engineer-month anchor of 168 hours per engineer per month at a $100/hour conservative US loaded rate. So one engineer-month = $16,800. Custom-build estimates show effort in engineer-months (e.g., 6 to 12 engineer-months for a standard portal) and let the calculator translate to dollar terms. Teams should substitute their own loaded rate; the formula is shown.
Backstage self-hosted TCO
Backstage TCO formula: setup engineer-months (3 to 6 months for a 2 to 3 person team) + ongoing platform-team time share (30 to 40 percent of the platform team's annual capacity for plugin maintenance, security patches, framework upgrades) + plugin development time (4 to 8 weeks per significant plugin). At 20 developers and a 2-engineer platform team, year 1 lands at $150K to $300K range, with the $150K being the optimistic case (3-month setup, low plugin development burden).
Per-seat SaaS TCO
Per-seat SaaS TCO formula: published per-user rate x team size x 12 months. Volume discount of 15 to 30 percent applied at 100+ seats per the published vendor enterprise-tier patterns. At 50 engineers, OpsLevel runs $39 x 50 x 12 = $23,400/year list, $20K to $23K after typical volume discount; Cortex at $65 to $69 lands $39K to $41K list. Enterprise-tier uplifts for SSO, RBAC, and audit log are out of scope.
API management portal extension
Incremental cost for extending an existing API gateway's developer portal module. Engineering effort: 2 to 4 weeks of customisation (theme, branding, signup flow, sample-app gallery, API catalog mapping). At $100/hour loaded, that is $16K to $32K one-time. Ongoing maintenance burden is shared with the gateway maintenance the team is already doing. Net incremental annual cost: $0 to $50K, dominated by feature-gap fills.
ROI driver math
Four measurable drivers, formulas on /roi: (1) TTFAC reduction = hours saved per new developer x developer hourly cost x new developers per year; (2) ticket volume reduction = annual tickets x reduction percentage x average ticket cost; (3) onboarding day savings = new hires per year x days saved x daily loaded cost; (4) recovered productivity = engineers x hours saved per week x 48 weeks x hourly cost. Industry benchmark ranges (40 to 70 percent TTFAC improvement, 30 to 50 percent ticket reduction, 2 to 5 days saved per hire, 30 to 60 minutes per developer per week recovered) sourced from CNCF Platforms Working Group survey and the DORA State of DevOps reports.
Break-even inflection points
Three published inflection points on /build-vs-buy: 50 engineers is where commercial SaaS beats custom build on a 3-year TCO basis for most teams; 100 to 200 engineers is the band where self-hosted Backstage starts to beat managed Backstage if the platform team capacity exists; 200+ engineers is where custom builds become economically defensible when developer experience is a strategic differentiator. These are point estimates from the engineer-month math; the calculator on /calculator lets teams substitute their own inputs.
Refresh cadence
The site is re-verified against each vendor's pricing page on the first business week of each month. The visible "Prices verified" label and the dateModified field in every page's Article JSON-LD read from a single constant (LAST_VERIFIED_DATE) so the on-page text, the schema, and the footer are always in lockstep. Cosmetic date refreshes are structurally impossible: bumping the date is a single-line change that touches every page at once.
Out-of-cycle refreshes trigger on:
- ●Vendor-announced rate change (e.g., Port adjusts the published per-user rate, Cortex moves the Cloud tier price).
- ●New plan tier introduced or removed (e.g., OpsLevel adds a Pro tier, Roadie shifts the developer-tier inclusions).
- ●New billing model (per-environment fees, per-component fees, audit-log uplift moving into the Standard tier).
- ●Backstage major release that materially shifts the platform-team time-share (e.g., a breaking framework upgrade that adds 2 to 4 weeks of work).
- ●New entrant in the developer portal category that crosses a meaningful adoption threshold.
Refreshes that move per-seat rates by less than 5 percent are batched into the next monthly pass. Refreshes that introduce a new billing model or shift a major adoption-cost assumption (Backstage framework breaking change, Backstage maintenance share moving outside the 30 to 40 percent published range) ship as soon as the change is confirmed against the source.
Limitations
Calculator and TCO outputs are estimates. Production developer portal pricing depends on enterprise agreements, volume commitments, audit-log / SSO / RBAC tier uplifts, and per-environment fees that are out of scope here. Always verify with the vendor before purchasing.
Backstage cost of ownership is sensitive to platform team capacity and plugin development ambition. The 30 to 40 percent platform-team time share is a published industry range; teams with deeper plugin ambitions (custom plugins per-quarter) sit at the high end. Teams that adopt Backstage as-is with no custom plugin development sit at the low end. The /backstage-cost page shows both ends of the range.
ROI percentages are an industry band sourced from CNCF and DORA published research, not a guarantee. Real outcomes depend on adoption rate, baseline TTFAC, and existing developer support process maturity. The ROI page shows the formulas so teams can model their own assumptions.
Corrections process
Spotted a stale price, a missing tier, or a vendor change we have not caught yet? Email [email protected] with the page URL and the source you would like cited. Substantive corrections (per-seat rate changes, free-tier inclusion changes, new billing models) are typically actioned within five business days. Non-substantive corrections (typo fixes, link rot, structural edits) batch into the next monthly pass.
See also the about page for the site's editorial position, disclosures, and full coverage map.