How we score
This page explains how Best Process Automation scores the 14 platforms it ranks, how its situational shortlists and cost estimates are built, and what it did not do.
The rubric
Every option gets a score from 0 to 10 on each criterion. The overall score is the weighted average, shown to one decimal. Positions follow the unrounded average, so when two options show the same score, the one with the higher unrounded average ranks first. An exact tie goes to the higher score on the heaviest criterion.
| Criterion | Weight | What we check |
|---|---|---|
| Ease of use | 20% | How quickly a non-developer gets a working automation: builder, templates, AI assistance. |
| Power | 25% | Branching, loops, error handling, code steps, AI agents, webhooks and APIs. |
| Integrations | 15% | Breadth and depth of connectors, including legacy and desktop systems. |
| Price and value | 20% | Entry price, free plan, and cost as volume grows. |
| Security and governance | 10% | SOC 2 or ISO 27001, SSO, audit logs, data residency, self-hosting. |
| Support and ecosystem | 10% | Documentation, community, partner network and response times. |
Where the facts come from
Every price was read on the vendor's own pricing page on 8 November 2026 and checked again on 9 November 2026, in US dollars unless the vendor publishes another currency: n8n prices in euros. Make shows prices by the visitor's country; this site quotes its US price list. Where a vendor's own pages disagree, the review says so and cites both.
- Prices come from each vendor's own pricing page and carry the date we checked them.
- Product facts come from the vendor's documentation, security pages, app-store listings or changelogs.
- Other rankings and AI answers are never used as a source for a fact.
- When a fact is not published, the page says so instead of guessing.
How often it changes
Once a day, a script fetches every cited source page that its robots.txt allows and compares the text, prices and quoted passages with the previous copy; a change, a missing page or a redirect puts the facts that cite that page on a recheck list. Every change to a score or a ranking position is dated in the changelog.
How evidence becomes a score
Two criteria use a fixed scale, so the same evidence always gives the same number.
- Integrations. The vendor's own count sets a base: 9,000+ is 9.0, 3,000+ is 8.0, 1,000+ is 7.5, 500+ is 7.0, 300+ is 6.5, under 100 is 5.5, and 7.0 when no count is published. Desktop or screen automation (RPA) for apps with no API adds 1.0. A way to reach systems inside a private network (an on-premises agent or gateway, self-hosting, VPC peering or tunnels) adds 0.5. A platform without desktop automation stops at 9.0.
- Security and governance. Five items count, on any plan: an audit report (SOC 2 or ISO 27001), SSO, audit logs, a choice of data region, and self-hosting or deployment in your own cloud. Three items score 7.5, four score 8.0 and five score 9.0. An ISO 27001 certificate, or a FedRAMP, HITRUST, PCI DSS or TISAX attestation, next to SOC 2 adds 0.5.
The other four criteria use anchors:
- Ease of use. 9.0: a no-code builder and an AI assistant on every plan, with a free start. 8.0 to 8.5: plain-language builders that a person supervises at first. 7.0 to 7.5: a visual builder with concepts to learn, or a start that needs IT or a sales call. 6.5 and below: setup that needs developers or servers, or free building time capped by a trial.
- Power. 9.5: every rubric item without major limits, plus orchestration of long-running processes across people, systems and agents. 9.0: every rubric item with few limits. 8.0 to 8.5: every item, with caps on loops or run time, features in beta, or code and agents outside the main builder. 7.5 and below: items missing or sold separately.
- Price and value. 8.5: a lasting free plan, an entry price near $25 a month or less, and unit prices that fall at volume. 7.5: free for some buyers, then priced per user. 6.0: a published entry price, with no lasting free plan or steep cost at volume. 5.5: published prices with one-time free credits, per-seat growth, unpublished unit costs or quote-only upper plans. 3.0 to 4.5: paid prices only from sales; a free plan that runs production work sets the top of that range, and no free plan the bottom.
- Support and ecosystem. 9.0: a published critical-response target of one hour or less, a partner network, and a community or academy. 8.5: response targets slower than an hour, or stated only as goals. 8.0: docs, a community and partners, with no published response times. 7.0 to 7.5: no published response times and limited channels below the top plan, no partner network, or a response target of four hours or more on weekdays only.
Each review ends every criterion note with a sentence that shows how its number follows from these scales.
Situational shortlists
Each shortlist states its requirement, keeps the platforms that meet it, and lists them in overall score order, up to five. Platforms that meet the requirement but fall below the fifth place are named on the page. When only one platform meets the requirement, as on the Microsoft 365 page, the platforms after it are alternatives for what it lacks, also in score order.
Cost estimates
Cost tables are estimates from published list prices; a vendor's quote can differ. Each table states the workflow, the monthly volume, the billing period and how each vendor counts a run. A platform is left out of a table when its published prices cannot turn a run into a cost, and the table says which.
What we did not do
No platform was installed or tested for this ranking. Scores rest on each vendor's published documentation, pricing, security and support pages, cited on every review. When a review says a feature works a certain way, it reports what the vendor's documentation says.