AI Content Reporting in Antradus AI: What WordPress Users Need to Track

AI content reporting gives you a clear, post-by-post view of words, images, tokens, and estimated cost inside WordPress, and on Antradus AI it exists to answer a practical question fast: what did this content run up in usage, and what should you bill against it as of 2026?

You publish ten posts in a batch, a client asks for a usage breakdown, and suddenly “we used AI” is not enough. You need counts, costs, and a paper trail. That is exactly where Antradus AI turns a vague workflow into something you can manage.

The documentation source for this feature is https://antradusai.com/docs/reporting/. Antradus AI is a WordPress plugin built to help publishers, site owners, and agencies create and manage AI-assisted content inside their own WordPress environment, and its reporting area focuses on showing what was generated and what that generation likely cost.

What is AI content reporting and why does it matter now?

AI content reporting is the usage ledger inside Antradus AI that shows what each post consumed and produced, including generated words, created images, token counts, and an estimated cost per post.

That matters in 2026 because content teams are not only asking whether output is fast. They are asking whether it is profitable, trackable, and easy to explain to a client or internal finance lead. A tool that generates content without showing usage leaves a gap. A reporting view closes it.

According to the Antradus AI documentation for Pro v2.0.0, last updated on 2026-07-24, the reporting feature applies to Antradus AI Pro v2.0.0 and lives in Antradus AI → Reporting inside WordPress. That detail matters because it tells you where to find the feature and which version the documentation refers to.

The other big point is storage. The documentation states that all reporting data stays in your own database and that nothing is sent anywhere. For WordPress site owners handling client work or sensitive editorial activity, that is not a small detail. It means the reporting view is based on data retained within your site’s own setup rather than pushed out to a separate reporting service.

“All data stays in your own database — nothing is sent anywhere.”

Antradus AI documentation, Reporting page

What does the Antradus AI reporting page actually show?

The Antradus AI reporting page shows a row-level breakdown of usage tied to individual posts, with separate columns for post name, words, images, tokens, and estimated cost.

Each column serves a specific purpose. Post identifies which article the usage belongs to, and the documentation says that includes bulk posts. If you generate content in volume, that prevents the familiar mess where usage is visible in aggregate but impossible to assign back to a specific deliverable.

Words records how many words were generated. The documentation adds a useful technical note here: the count works correctly for Chinese, Japanese, and Korean as well. That is more important than it sounds. Multilingual publishers often run into poor counting rules, especially when a system assumes space-separated words. Antradus AI explicitly calls out support for CJK word counting, which makes the report more credible for multilingual work.

Images shows how many images were generated for that post. If image creation is part of your workflow, this saves you from stitching together separate text and image usage records by hand.

Tokens shows the exact token counts reported by your provider. That wording is specific. The token number is not described as a rough guess from the plugin; it is based on what the provider reports.

Estimated cost is calculated from two pieces: tokens multiplied by your model prices, plus images multiplied by your image price. So the report is not locked to a default pricing assumption. It uses the numbers you set, which is exactly what an agency or operations manager needs if provider rates vary across models.

How does AI content reporting calculate cost?

AI content reporting calculates cost by multiplying token usage by the model prices you enter and adding image totals based on the image price you set in the reporting settings.

The documentation says prices are pre-filled and fully editable. On the Reporting page, you look for the model-prices section, then enter your provider’s current per-token or per-million-token rates for the models you use, plus your per-image price. After that, costs recalculate from your numbers.

That setup solves a common problem. AI providers do not all bill the same way, and pricing changes over time. If a reporting tool hard-codes rates, it goes stale fast. Antradus AI avoids that by letting you maintain your own pricing inputs directly inside WordPress.

Say you use one model for drafting, another for editing, and image generation on top. A single blended estimate would hide the difference. A price table you control makes the report fit your actual stack, not someone else’s assumptions.

The documentation also names one special case: OpenRouter ★ FREE models show as zero cost correctly. That is a practical touch because free-tier usage can distort internal reporting if the system insists on charging every action. Here, zero-cost usage remains visible without pretending it incurred paid model expense.

“Free models (OpenRouter ★ FREE) show as zero cost, correctly.”

Antradus AI documentation, Reporting page

How can agencies use AI content reporting in real client work?

Agencies can use AI content reporting in Antradus AI to turn generation data into a usable internal cost sheet and pair it with exportable delivery reports for clients.

The documentation is unusually direct on this point. It says the feature matters for agencies because you can set your real rates once and the report becomes a true cost sheet you can place next to client invoices. That is not fluff. It describes a real operating workflow: configure actual provider costs, generate client work, then compare delivery volume against underlying generation expense.

Antradus AI also connects reporting to another feature named Bulk Publishing → 📊 Reports CSV export. The documentation suggests a simple split: batch report for the client, cost report for you, with margin visible per project. That is one of the clearest business uses in the source material.

For a WordPress agency running content packages, that can cut a lot of manual reconciliation. Instead of opening provider dashboards, copying token usage, checking image counts, and matching them back to post titles, you can review post-level records in one place. Then you can export the client-facing batch report separately while keeping the cost math internal.

And because the report includes bulk posts, it fits high-volume workflows rather than only one-off article generation. If you publish at scale, that distinction matters. A reporting screen that breaks the moment you move from one post to fifty is not much use.

What are the limits and gotchas of AI content reporting?

AI content reporting in Antradus AI is a management view, not the final billing authority, and the documentation is clear that you should treat provider invoices as the source of truth.

The most important limitation is right there in the wording: estimated cost. The report estimates cost from the pricing values you enter. Providers still bill through their own dashboards, using their own metering and invoice rules. If there is ever a discrepancy, the provider’s invoice wins.

The documentation also notes that usage is attributed to a post when there is one, but utility calls are tracked too. One example given is topic suggestions. That means not every recorded action is a neat one-post, one-generation event. Some usage happens around the publishing workflow rather than inside the final article body. If you are auditing costs closely, that is good to know upfront.

Another practical note: old rows can be purged from the page if you want a clean slate. That helps with housekeeping, especially on long-running WordPress installations where reporting tables can get cluttered over time. But it also means you should think before clearing historical records if you rely on those rows for monthly comparisons or client disputes.

So yes, AI content reporting gives you useful financial visibility. But no, it is not an accounting ledger in the strict sense. It is an operations tool that helps you manage content production with better numbers.

What should you do with the reporting feature next?

You should open the Reporting screen in Antradus AI, enter your actual model and image prices, and check whether the estimated cost lines match the way your WordPress content operation is billed in practice.

Start with the basics. Confirm you are using the feature documented for Antradus AI Pro v2.0.0. Go to Antradus AI → Reporting. Review the columns for post, words, images, tokens, and estimated cost. Then update the model-prices section with your current provider rates as of August 2026, not old assumptions.

If you run client work, compare one week of reports against your provider dashboard and one invoice cycle. That quick audit tells you whether your internal pricing setup is realistic. After that, use the report as your day-to-day management view and the provider invoice as the final billing record.

One more move. If you use bulk workflows, pair the reporting screen with the CSV export from Bulk Publishing reports. That gives you a cleaner split between what clients need to see and what you need to monitor behind the scenes.

For WordPress teams trying to keep AI production measurable, Antradus AI makes that job much simpler, and the source documentation at https://antradusai.com/docs/reporting/ is the right place to review the exact feature details before you roll it into your process.