Revenue Journey Assessment

Give it a website. It returns a scored, evidence-backed read of how that business turns a stranger into a repeat customer — and where the journey leaks.

Scored checks across five stages — awareness, consideration, decision, conversion, retention — each with the evidence that produced the score and a plain-language recommendation when it comes back low. No account, no API key, no setup.

What you get

One row per run, in the default dataset. The headline numbers:

Field	What it is
overall_score	0–10, the five stages weighted for this kind of business
awareness_score … retention_score	0–10 per stage
weakest_stage / strongest_stage	where to spend the next hour, and what already works
business_class	detected from the site: local, professional_services, ecommerce, saas, content_creator, influencer
naics_code / naics_title	the industry the site reads as
assessment_detail	every check: score, evidence, and recommendation

The full row shape is published in the actor's dataset schema — see the Storage tab on any run.

The evidence, not just the number

Every check carries what it actually saw:

{
  "label": "Booking / Demo / Consultation CTA",
  "stage": "decision",
  "score": 0,
  "max": 3,
  "evidence": "No booking, demo, or consultation CTA detected",
  "recommendation": "A prospect just decided they want to work with you, but there is no booking mechanism on your site. They will book with whoever makes it easiest.",
  "needs_review": true,
  "review_reason": "Crawl degraded — this hub-derived zero may be a false negative"
}

A score you cannot check is a score you cannot act on, so the evidence ships with it.

Scored on what a website can show

Stage weights follow the business class. A local service business is scored heavily on awareness and conversion; a SaaS on decision. The same site scores differently depending on what it is, which is the point.

This actor reads the website and nothing else. No Google Business Profile, no Google Maps, no Bing listing, no search-rank lookup.

That is a scope boundary, not a gap in the crawl, and the scoring respects it: the checks that need those sources are not in the rubric here at all. They are not scored zero and quietly averaged in — a business with a strong Google presence is not penalised for a lookup this actor never makes. You are reading a website audit, scored out of what a website can tell you.

If you need Google Business Profile, review, and SERP scoring alongside this, that is a different product — see richardnorwood.com.

Honest about a bad crawl

A slow, blocked or single-page crawl produces weak readings, and the run says so rather than reporting them as findings. assessment_detail.audit.crawl_quality carries the status and the reason:

{
  "status": "degraded",
  "reasons": ["Only the homepage was crawled — sub-page signals (forms, case studies, pricing) may be missing"],
  "pages_crawled": 1,
  "ttfb_ms": 1561
}

and every affected check is flagged needs_review: true with the reason attached. A zero that means "we could not see" is never presented as a zero that means "it is not there."

Input

Only businessUrl is required.

{
  "businessUrl": "https://www.example.com",
  "brandName": "Example Co",
  "businessClass": "local"
}

Field	Notes
businessUrl	Required. Must start with http:// or https://
brandName	Optional. Defaults to the name detected on the site
businessClass	Optional override. Leave it out and the classifier decides
proxy	Defaults to Apify residential proxy, US
maxConcurrency	Default 5
maxRequestRetries	Default 3

A run takes roughly one to three minutes depending on how many pages the site exposes.

Getting the results

The run's Output tab links straight to the row. Via API:

https://api.apify.com/v2/datasets/<defaultDatasetId>/items

Or ?format=csv for a flat export — assessment_detail is a nested object, so JSON is the richer read.