Crozdesk Software Scraper - SaaS Product Data
Pricing
from $2.00 / 1,000 results
Crozdesk Software Scraper - SaaS Product Data
Export Crozdesk software references with separate community review excerpts, product URLs and source provenance. Reviews are not product descriptions.
Pricing
from $2.00 / 1,000 results
Rating
0.0
(0)
Developer
Ben
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
0
Monthly active users
5 days ago
Last modified
Categories
Share
Crozdesk Software Scraper
Read software references from Crozdesk’s public homepage while keeping community review excerpts separate from product descriptions. Each accepted card identifies one software product through its name and Crozdesk item URL.
Start with the saved example
Open the public example to inspect its input and output fields. Run a bounded request and open its dataset to see the returned records.
{"searchUrls": ["https://crozdesk.com/"],"maxResults": 3}
The September 11 homepage capture starts with review cards for Afina Browser, KORONA POS and Maya AI. These are references in the Recent Community Reviews section. The sample does not establish a complete software directory, popularity order or a vendor-authored description.
What the fields mean
The homepage cards contain review snippets. The Actor places that text in review_excerpt and leaves description null, because those cards do not supply standalone product summaries. It excludes the “All … User Reviews” link text and does not combine neighboring reviews. The excerpt can end in an ellipsis exactly as the source presents it.
The Actor requests the supplied page over HTTP. It does not sign in, execute page JavaScript, scroll, open product details or make purchases. Browser-only content may therefore differ from the fetched response. Increasing the result cap cannot recover records that are absent from that response.
Dated output example
This record comes from the September 11, 2026 source capture. Keep the source wording and its observation date together when reusing it.
{"source": "crozdesk-software-scraper","kind": "software","index": 1,"title": "Afina Browser","url": "https://crozdesk.com/software/afina-browser","description": null,"review_excerpt": "I work as a personal accountant, so I spend a lot of my day in large Google Sheets and financial dashboards. In my pr...","input_url": "https://crozdesk.com/","source_url": "https://crozdesk.com/"}
| Field | Meaning |
|---|---|
title | Software/product name on the individual card. |
url | Absolute source item URL. |
description | Product-card summary, or null when the card supplies another text type. |
review_excerpt | Optional separately labeled community excerpt from Crozdesk. |
source, kind | Existing source and record-type identifiers. |
index | One-based position among accepted records for this input URL. |
input_url, source_url | Requested URL and final fetched URL after redirects. |
company, raw | Optional original structured-response fields; not inferred from a card. |
Download JSON to retain nulls and optional fields. CSV or Excel can support a manual shortlist. Save the run or dataset ID with each export, because page URLs alone do not identify its collection time. A typed field describes the output contract; it does not guarantee that a source supplies every optional value.
Input limits and repeated collection
searchUrls is the list of source URLs to request. maxResults applies separately to each URL, from 1 to 1,000; the schema default is 25. Two URLs with a limit of three can return up to six records. The Actor removes duplicate item URLs within each response and keeps the first accepted occurrence. The same item in different requested pages can appear more than once, preserving each page’s provenance.
There is no automatic pagination or cross-run state. Save dated exports separately when comparing changes. Match exact item URLs within the same source and keep your internal notes outside the source-text fields. A changed title or missing row in a capped response is a review question, not automatic evidence of a product launch or removal.
Pricing
Charges apply to exported rows plus the existing memory-based Actor start event. Current result charges per 1,000 rows are:
| Plan | Result charge per 1,000 |
|---|---|
| Free | $2.5 |
| Bronze | $2.25 |
| Silver | $2.125 |
| Gold | $2 |
| Platinum | $1.875 |
| Diamond | $1.75 |
The Free start rate is $0.00005 per GB with a one-unit minimum; paid plans retain their existing discounts. The unchanged 4,096 MB default uses four start units. The Pricing tab shows the current terms. Set a total charge cap for bounded runs; requesting fewer results limits output, not the full source response download.
Questions before using the export
Why is description null? A community review is a different kind of statement from a product summary. The verified homepage cards supply review excerpts, so the Actor records them in review_excerpt. Filling description with that text would conceal its origin.
Does it download every review? No. It reads the excerpt present in the homepage response. It does not follow the review link, paginate through review history or claim a full review body. Repeated references to the same software within one response keep the first accepted card.
Does an excerpt establish a product feature? It records what a source reviewer wrote. A statement about inventory management or spreadsheet work is not an independent product test, verified requirement or vendor guarantee. Keep evaluation notes separate from the excerpt.
Are reviewer identities included? This card export omits reviewer names, profile images, ratings and review dates. It retains the software name, item URL and available excerpt. The run completion time identifies collection time; it is not the date when the review was written.
Will a software detail page use the same parser? The accepted starting point is the homepage review-card layout. Category pages and individual product pages have not been independently qualified in this release. Supplying a different layout may produce no accepted rows; inspect its source and response before scheduling it.
What happens when a page fails? Failed fetches fail the run, while rows already saved from earlier URLs remain available. Inspect the final status and server log before treating a multi-URL export as complete. A nonempty dataset can still be partial.
Can I schedule collection? Apify supports saved Tasks, schedules, webhooks and dataset APIs. Check a small real example before enabling recurring work. The Actor itself does not retain comparisons, send alerts or decide whether a downstream import should overwrite an older snapshot.
What if the page layout changes? A page without recognized cards can return no rows. Inspect the final source URL and report a reproducible issue with the public input and run ID. Never include credentials or private account data in a public issue.
May I reuse the text? Keep source attribution and respect applicable terms and rights. The export records source statements; it does not transfer ownership or validate their accuracy. Use your own evaluation for purchasing, compatibility and compliance decisions.
Related source comparison
AppSumo Software Scraper reads product-card summaries. Its descriptions have a different source role from Crozdesk community excerpts; preserve that distinction in a combined catalog.
The software-catalog identity guide explains why matching names and shared websites do not establish a single product identity across platforms. Its downloadable offline example needs no new extraction run.
Keywords: Crozdesk software, public software references, community review excerpts, software product URLs, source attribution, JSON export, CSV export, review shortlist, software research.