# Chronometrica vs Streamers: YouTube Metrics With Precision Labels

Choose Chronometrica when your YouTube comparison needs each count’s availability and precision beside its value. Choose Streamers YouTube Scraper when the same job needs its documented search filters, playlists, subtitle collection, or optional AI tools. The choice depends on the output your report requires and the wider collection job. It does not depend on a proven difference in count accuracy. See the [Chronometrica contract](https://apify.com/chronometrica/youtube-metrics-scraper) and [Streamers contract](https://apify.com/streamers/youtube-scraper).

This guide is for a creator or analyst who already has a small list of similar public videos. The goal is a table that links each count to its source and shows which records meet a stated comparison rule.

This is a Chronometrica guide. It uses public product contracts checked on October 20, 2026. The input example below passed a schema check, which checks the input’s structure. It was not run. The worked numbers are fictional.

## Choose the collection contract your report needs

An Actor is a hosted data collection program. Its output contract describes the data it returns and how to read that data. For a comparison report, that contract matters because a number alone may not tell you whether you can use it.

Suppose your report accepts only full public counts for a like-to-view calculation. You need to know whether each required count was available. You also need to know whether the source showed a full number or a rounded label. A value of 100,000 and a display of 100K can require different treatment under that rule. The view count must also be greater than zero before you can use it in a ratio.

| Decision point | Chronometrica YouTube Metrics Scraper | Streamers YouTube Scraper |
| --- | --- | --- |
| Main checked fit | Reports that require explicit metric availability and count precision | Wider collection with search filters, playlists, subtitles, or optional AI tools |
| Relevant documented output | Public counts with metric status structures | Public video and channel counts |
| Reader’s responsibility | Apply the report’s eligibility rule to each metric | Define equivalent availability checks for the report |

These are documented fits, not results from a comparison run. See the [Chronometrica documentation](https://apify.com/chronometrica/youtube-metrics-scraper) and [Streamers documentation](https://apify.com/streamers/youtube-scraper).

Chronometrica documents a flat row: one record with fields for the post, its counts, and collection time. Relevant fields include `inputUrl`, `postUrl`, `postId`, `title`, `publishedAt`, `views`, `likes`, `comments`, `channelSubscribersCount`, and `scrapedAt`. Its `postMetricStatus` and `channelMetricStatus` structures describe metric availability and precision. Missing counts stay missing. This explicit contract is the reason to choose it for a report with strict count rules. [Chronometrica output contract](https://apify.com/chronometrica/youtube-metrics-scraper)

Streamers documents direct URLs, keyword search, and separate result limits for regular videos, Shorts, and streams. Those controls can support a wider collection job. [Streamers input schema](https://apify.com/streamers/youtube-scraper/input-schema)

The main switch condition is therefore practical: choose the collector whose documented output and collection modes fit the whole job. This guide does not establish that a competitor count is wrong or that it lacks all quality information.

## Read availability and precision before calculating

Availability answers whether a metric is present. Chronometrica documents states such as `present`, `unavailable`, and `not_attempted`. An unavailable count and a count that was not attempted are different cases. Keep that difference when you review the results. [Chronometrica metric status contract](https://apify.com/chronometrica/youtube-metrics-scraper)

Precision describes how fully the public source expressed the count. The documented labels include `exact`, `display_rounded`, and `likely_rounded`. A rounded display gives less detail than a full count. Preserve that label beside the value so that a later calculation does not hide the source limit. [Chronometrica precision contract](https://apify.com/chronometrica/youtube-metrics-scraper)

Here, “exact” concerns the public count as shown. It does not certify the platform’s underlying records. It also does not turn public views into unique people, watch time, retention, or sales.

Your report needs a rule before it needs a formula. For example: “Include a video in the exact-count ratio only when public likes are present and exact, and public views are present, exact, and greater than zero.” Present means the value is not missing or `null`. A record that meets this rule is eligible. A record that does not meet it can still remain in the source table with a clear reason.

Zero is a known value, but division by zero is undefined. Keep a zero-view record in the source table and exclude it from ratio coverage. Do not report its ratio as zero.

Keep the source value, availability, and precision together. If you copy only the number into another sheet, the next reader may use it without knowing that the source was rounded.

## A fictional example: three videos, one eligible ratio

All three records in this section are fictional reader data. They are not outputs from a current run, a product test, or a customer report.

The reader wants an exact-count public like-to-view ratio. This ratio expresses likes per 100 public views. Likes are the numerator, the number above the division line. Views are the denominator, the number below it. Multiplying by 100 gives a percentage:

`public like-to-view ratio = public likes / public views × 100`

For an eligible ratio, likes must be present and exact. Views must be present, exact, and greater than zero.

Video A has 100,000 exact public views and 2,000 exact public likes. Both counts meet the rule, and the view count is greater than zero:

`2,000 / 100,000 × 100 = 2.00%`

Video A is eligible for the exact-count comparison. Its result describes the relationship between two public counts. It does not show what share of unique viewers liked the video.

Video B has a displayed view label of 100K and 2,100 exact public likes. Using 100,000 in the calculation would give about 2.10%. But the view input is rounded. Under the declared rule, Video B is excluded from the exact-count set and ratio coverage. The reader could keep the estimated figure in a separate group, with its rounded input clearly marked. That estimate does not meet the ratio eligibility rule.

Video C has 300 public likes and no public view count. Its ratio is unknown. The reader must not replace the missing view count with zero. Zero is a value; missing means the required value is unknown. A known zero-view count would also prevent division, but for a different reason: division by zero is undefined.

One of the three supplied videos meets the exact-count rule, including the requirement for views greater than zero:

`1 / 3 × 100 = 33.33% eligible coverage`

Eligible coverage is the share of supplied records that meet the report’s rule. It is not a scraper success rate. Video B is excluded because of rounded views. Video C is excluded because views are missing. Those reasons belong beside the records. Any zero-view rows would also be excluded from ratio coverage, with their own reason.

This example supports one exact ratio. It does not support a precise performance ranking between A and B. To expand the eligible comparison, the reader needs suitable exact counts, including a view count greater than zero. Clearly marked estimates can remain separate. The same eligibility rule must apply to every record.

## Use a small, repeatable comparison method

### 1. Fix the question and supplied list

Write down what you want to compare before collection. “Which supplied videos have the highest public like-to-view ratio?” is a question that this method can address. “Which thumbnail caused more people to watch?” needs different evidence.

Keep topic, format, video age, and observation time as similar as the question requires. Do not add or remove videos after seeing the ratios without explaining the change. For help choosing the list, use the guide to [YouTube reference panel design](/en/articles/youtube-reference-panel-design/). This comparison begins after that list is chosen.

### 2. Name the required counts and rule

List each count needed for the calculation. A public like-to-view ratio needs likes and views. A comparison of public view totals needs views. Channel subscriber counts may give context, but they do not automatically belong in either formula.

State the numerator, denominator, and unit before you calculate. Then state the eligibility rule in plain English. For the exact-count ratio, require present, exact likes and a present, exact view count greater than zero. Exclude missing, rounded, and zero-view records from ratio coverage.

If estimates are useful, keep them separate from eligible ratio results. A known zero view total can remain in a count comparison, but it cannot serve as a ratio denominator. Set these rules before you see which records qualify.

### 3. Collect a small known list

Start with supplied video URLs so you can compare the requested list with the returned records. Keep post identity and collection time. In Chronometrica output, `postId` and `postUrl` help identify the video, while `scrapedAt` records the collection time. [Chronometrica output fields](https://apify.com/chronometrica/youtube-metrics-scraper)

A known list also gives you a clear starting total for coverage. Keep that original list available during review. Do not let the set of returned rows silently become the definition of what you intended to collect.

### 4. Read all three result locations

Chronometrica sends only `ok` rows to the default paid dataset. A dataset is a stored collection of records. Other outcomes go to the separate Unbilled results dataset with the same row fields. `OUTPUT` reports `datasetRows`, `unbilledRows`, status counts, and metric coverage. [Chronometrica results contract](https://apify.com/chronometrica/youtube-metrics-scraper)

Read all three locations. Keep unresolved inputs separate from resolved records that lack a required metric. A successful processing run does not mean every input produced an eligible record. An `ok` row also does not mean every requested metric is present or exact. Check the metric status fields and the required values against your report’s rule.

### 5. Apply the rule to each required metric

For every resolved record, check each count used by the formula. Confirm availability first. Then check precision. For a ratio, also confirm that the exact view count is greater than zero. A full like count cannot make a rounded, missing, or zero view count eligible.

Keep eligible exact-count results, rounded estimates, and unknown results separate. Mark a zero-view ratio as undefined and exclude it from ratio coverage. Do not put it in the unknown-view group, because its view count is known.

Add a brief reason when a record is excluded. “Views rounded,” “views unavailable,” and “views zero; ratio undefined” describe different limits. Preserve the original status information so a later reviewer can check your decision without repeating the collection.

### 6. Report coverage against the supplied list

Count the records that meet the full rule. For this ratio, that means exact, present likes and exact, present views greater than zero. Divide the eligible record count by the number of records on the original supplied list, then multiply by 100. State both the fraction and the percentage.

Zero-view rows stay in the original supplied-list total but do not enter the eligible record count. Treat missing and rounded view rows the same way for coverage, while keeping their exclusion reasons separate.

This step shows how much of the intended comparison remains. A report based on one eligible video from a larger list has a different scope from a report where every supplied video qualifies. Explain the exclusions beside the result. Do not label this coverage as collection accuracy, product reliability, or a success rate.

### 7. Calculate and describe only the declared measure

Use the same formula and eligibility rule for every included record. Calculate eligible ratios only after confirming that the view count is exact, present, and greater than zero. Retain source links beside the results. Keep an estimated result visibly separate so sorting a table does not place it among eligible exact-count results without explanation.

A public count difference or ratio can describe the observed records. It cannot, by itself, explain why one video received more attention. Titles, thumbnails, publication timing, channel size, and other differences may still matter. A matching topic does not remove all of those differences or create a causal experiment.

### 8. Check important results at the source

Open the public source page when a result will affect a decision. Check the video identity, count display, and observation time. Public counts can change, so a later page view may differ from the saved record.

Use your own YouTube Studio data when the question requires private measures such as retention or watch time. Keep those measures clearly named if you use them beside public peer counts. Do not present a public ratio as a substitute for a private measure that you cannot observe for the other channel.

## Make the first collection easy to inspect

The public [YouTube Video Performance Tracker Task](https://apify.com/chronometrica/youtube-metrics-scraper/examples/youtube-video-performance-tracker) provides a starting point. A Task is a saved input preset.

The following input was checked against the current public schema. It was not run. The URL is a schema example, not a checked live metric result. Replace it with one relevant public comparison video:

```json
{
  "startUrls": [
    "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
  ],
  "maxItems": 1
}
```

Make the first collection easy to inspect

The fields above match the [Chronometrica input schema](https://apify.com/chronometrica/youtube-metrics-scraper/input-schema). `maxItems` is an overall cap on saved rows. It does not prove full source coverage. A single direct-video input keeps this first check focused on the record you supplied.

Before you expand the list, confirm that you can find the video identity, collection time, required count, and its status information. Then check where the record appears and read the run summary. For a ratio, confirm that the exact view count is present and greater than zero. This small check shows how to build your comparison table without assuming that a returned row is ready for every calculation.

## Start with one count you can explain

For a report that requires count availability and precision, choose the [Chronometrica YouTube Metrics Scraper](https://apify.com/chronometrica/youtube-metrics-scraper). Collect one relevant public video. Inspect one count’s availability and precision before adding it to your comparison table.

Keep the source link, collection time, value, and status together. Once you can explain why that count meets your rule, extend the same check to the remaining supplied videos. For ratio eligibility, require an exact, present view count greater than zero. If the wider collection job requires the other Actor’s documented extra modes, use that workflow and retain equivalent checks in your report. The final table should show which results are eligible exact-count calculations, which are estimates, which remain unknown, and which have an undefined ratio because views are zero.

## Sources

- [Chronometrica YouTube Metrics Scraper: output fields, metric status, and result handling](https://apify.com/chronometrica/youtube-metrics-scraper)
- [Chronometrica input schema](https://apify.com/chronometrica/youtube-metrics-scraper/input-schema)
- [YouTube Video Performance Tracker Task](https://apify.com/chronometrica/youtube-metrics-scraper/examples/youtube-video-performance-tracker)
- [Streamers YouTube Scraper: documented collection features and output](https://apify.com/streamers/youtube-scraper)
- [Streamers input schema](https://apify.com/streamers/youtube-scraper/input-schema)
- [Reader question about comparing similar YouTube creators](https://www.reddit.com/r/YouTube_startups/comments/1tihdnb/how_do_you_compare_your_youtube_analytics_with/)