Ahrefs Brand Radar April 2026 Updates: What Grok and Bigger API Limits Change

Ahrefs’ April 2026 release brought eight product updates, including Grok coverage in Brand Radar and a 4× increase in API units for Lite plans. This guide explains the Ahrefs Brand Radar April 2026 updates and, more importantly, how to use them to measure brand mentions, AI citations, share of answer, and competitor gaps with less guesswork.

The headline features are easy to summarize: a new AI platform, better citation reporting, dashboard connectivity, and more API capacity. The harder question is operational: which buyer prompts should we monitor, what should count as a visibility win, and how do we separate a page being retrieved by an AI system from it actually being cited or driving brand preference?

Ahrefs Brand Radar April 2026 updates at a glance

Ahrefs grouped its April 2026 release across Brand Radar, Site Explorer, Keywords Explorer, Google Search Console reporting, Social Media Manager, and its API. For AI-search teams, Brand Radar is the center of gravity because it gained Grok, redesigned citation reporting, historical citation-position views, and a Looker Studio connector.

The confirmed changes included:

• Grok in Brand Radar for both Ahrefs’ prompt dataset and custom prompts.
• A rebuilt Cited Pages report with a citation funnel: Found in versus Cited in.
• Historical views for citation visibility and average citation position over time.
• A Looker Studio connector for Brand Radar metrics such as mentions, impressions, and share of voice.
• Larger self-serve API unit and row limits, plus POST endpoints for Brand Radar.
• A Site Explorer Organic positions report and Crawled pages report.
• Historical total search-volume charts for keyword lists in Keywords Explorer.
• New Google Search Console filters, Portfolio-level GSC Insights, and a redesigned Social Media Manager overview.

Not every change has equal importance for every team. A content team trying to understand why a competitor appears in answers will likely care most about the citation funnel and custom prompt support. An agency building recurring client reporting may care more about the connector and API changes.

The release is also a useful reminder that AI search measurement is broader than citation collection. A citation is a visible source link or attribution. Visibility includes whether a brand is named, recommended, compared, omitted, or displaced by a competitor in the answer itself. We explain that distinction in more detail in our guide to AI citations vs. AI visibility tracking.

Grok in Brand Radar: availability, coverage, and caveats

The most discussed April addition was Grok, xAI’s AI assistant, in Brand Radar. At launch, Ahrefs said Grok was available for both its existing Ahrefs prompts and user-defined custom prompts. That matters because the two datasets answer different questions.

Ahrefs prompts versus custom prompts

Ahrefs prompts provide broad discovery: they help identify where a brand appears across a large prompt corpus. Custom prompts provide depth: we define the exact questions buyers ask, select platforms and locations, then track the resulting answers on a chosen schedule.

For example, a B2B cybersecurity vendor might use broad discovery to surface mentions across category questions, then track a focused set of custom prompts such as:

1. “What are the best SOC 2 compliance platforms for startups?”
2. “Is [brand] a good alternative to [competitor] for vendor risk management?”
3. “Which compliance platform integrates with Jira and Slack?”
4. “What is the most affordable GRC tool for a 100-person company?”

Grok support means these prompts can be observed in one additional AI environment rather than leaving a gap in the reporting set. But we should not treat “platform added” as equivalent to “all collection is permanently available.” Ahrefs’ more recent help documentation says that, following changes in Grok policy, it was temporarily unable to gather new Grok data. As of August 29, 2026, that is the practical limitation to verify in an account before promising ongoing Grok tracking to stakeholders.

What Grok changes in competitor analysis

Grok can produce a different competitor set, framing, or recommendation order than ChatGPT, Gemini, Perplexity, or Google AI results. That is useful only if we preserve the engine-level breakdown. A blended “visibility score” can hide the fact that a brand wins in one model and is invisible in another.

We recommend recording, at minimum, four values per prompt and platform:

• Brand mentioned: Was our brand named at all?
• Brand recommended: Was it presented as a viable or preferred option?
• Citation status: Did the response cite our domain or a page we control?
• Competitor gap: Which named competitors appeared when we did not?

That approach creates a usable share-of-answer view rather than a pile of screenshots. For a deeper framework, see our guide on measuring which brands win buyer answers in AI search.

The Cited Pages report now shows a citation funnel

The strongest Brand Radar update is arguably not Grok. It is the rebuilt Cited Pages report and its separation of Found in from Cited in.

Ahrefs defines Found in as responses where a page was found during the system’s search or grounding process, whether or not that page was ultimately cited in the answer. Cited in is the number of responses where the domain was actually cited by the AI.

That distinction surfaces a useful middle state: the AI appears to have discovered or used a page, but did not expose it as a citation. Previously, teams could easily collapse those two realities into a binary “we were cited” or “we were not cited” report.

Why found-but-not-cited is a real optimization signal

Suppose a pricing page is found in 40 responses but cited in only 5. That is not proof that the page is poor. It is evidence that the content may be relevant enough for retrieval but not structured, specific, trustworthy, or directly useful enough to earn visible attribution.

We would inspect that page alongside the pages cited instead and ask practical questions:

• Does the page answer the prompt’s central comparison or only describe the product?
• Are claims supported with specifications, methodology, pricing context, or original data?
• Does a competitor page offer a clearer list, table, definition, or independent validation?
• Is the page technically accessible and crawlable for the relevant AI search process?
• Does the answer cite third-party sources because the buyer question requires evidence beyond a vendor’s own claims?

This is why we should avoid reducing AI citation tracking to a single count. It is better to model a sequence: eligible content is discovered, relevant content is used, a source may be cited, and a brand may or may not be recommended. Each stage has a different fix.

The report also moved the former Cited Domains view into an All domains tab alongside Yours, Competitors, and Other. That layout is more helpful for diagnosing whether a gap is owned-content-related, competitor-related, or driven by publishers, review sites, communities, and other third-party sources.

Citation history is useful only when the prompt set stays stable

Ahrefs added two chart modes to the Cited Pages report: Found in and Position. Found in shows a solid line for total responses where a page was found and a dashed line where it was cited. Position tracks average citation position over time, with line thickness indicating how many citations support the average.

That gives teams a historical view instead of forcing them to compare separate snapshots manually. Yet trend charts have a condition: the underlying prompt set, location, platform, and measurement cadence must remain clear.

A drop from position 2.1 to 4.8 can be meaningful. It can also be noise if the result rests on a small number of citations, if the platform changed response behavior, or if the prompt inventory was altered mid-period. Ahrefs itself signals sample strength through line thickness, which is a sensible cue not to overinterpret fragile averages.

We recommend using a reporting note for every trendline:

Citation-position features apply to chatbot models, according to Ahrefs’ April release notes. Google AI Overviews and AI Mode do not expose citation positions in the same way, so they retain the older views. That is a platform constraint, not a reporting failure.

A practical prompt-level workflow for Brand Radar

The release supports a more mature workflow, but the tool cannot choose the right prompts for us. We should build prompt tracking around buyer intent and decisions, not around a legacy keyword list alone.

Start with 30 to 100 prompts grouped into a small set of repeatable intent clusters. For a project-management software company, clusters might include “best for,” “alternative to,” “integration,” “security requirement,” “pricing,” and “implementation.” For each prompt, choose the platforms and markets that matter commercially.

Step 1: Separate discovery from monitoring

Use broad Ahrefs prompts for category discovery and custom prompts for the priority questions we need to defend or win. The two are complementary: discovery finds unexpected opportunities; custom tracking keeps important buyer questions under observation.

A keyword can suggest a topic, but it rarely captures the full wording, constraints, and comparison logic present in an AI question. We cover the measurement implications in AI search monitoring prompts vs. keyword lists.

Step 2: Capture answer-level outcomes

For each prompt-platform result, store a consistent set of fields:

• Prompt and intent cluster
• AI engine and selected location
• Our brand: mentioned, recommended, or absent
• Named competitors and their apparent role
• Controlled-domain citation, if any
• Third-party cited sources
• Answer position or ordering where observable
• Date, run identifier, and raw response record

The raw response matters. A monthly dashboard may say visibility rose 8%, while the actual answer reveals that the increase came from a disclaimer, a passing mention, or a negative comparison. Metrics should direct analysis, not replace it.

Step 3: Turn gaps into actions

Prioritize gaps where all three conditions are true: buyer intent is valuable, competitors are repeatedly named, and there is a plausible content or evidence change we can make. Avoid publishing pages just because an AI response cited a competitor once.

For each priority gap, choose one action: improve an existing page, create a missing comparison or use-case page, strengthen independent evidence, repair accessibility, or pursue authoritative third-party coverage. Then re-measure against the same prompt cluster over multiple runs.

Bigger Ahrefs API limits change automation, not measurement design

Ahrefs expanded API capacity across self-serve plans in April 2026. Lite plans received 4× the previous API-unit allowance and 10× the previous row limit per request. Standard increased from 150,000 to 400,000 units and from 25 to 250 rows per request. Advanced increased from 500,000 to 1 million units and from 100 to 500 rows per request.

Those are meaningful operational improvements for agencies and larger in-house teams. More units and rows make it easier to pull larger report sets, refresh dashboards, and reduce the need for awkward pagination patterns.

Ahrefs also released POST variants of Brand Radar endpoints so more complex entity inputs—such as brand variations and brand websites—can be passed in JSON request bodies. New endpoints were also introduced for Brand Radar report management, including creating reports, adding prompts, and monitoring results.

What the API expansion does not solve

More capacity does not automatically make AI visibility reporting reliable. We still need to manage prompt duplication, platform differences, API cost allocation, incomplete runs, and changes in model behavior.

A sensible automation design therefore includes:

1. A canonical prompt registry with a unique ID for every prompt.
2. A fixed brand-variation and competitor dictionary.
3. A separate table for raw answers and another for calculated metrics.
4. Scheduled data-quality checks for missing platforms, failed runs, and unexpected response changes.
5. Reproducible calculations for mention rate, citation rate, and competitor share of answer.

The Looker Studio connector helps distribution, not diagnosis

Ahrefs’ new Brand Radar Looker Studio connector provides historical charts for AI mentions, impressions, and AI share of voice, plus distributions across brands and platforms. This is useful for teams that already use Looker Studio to combine organic search, paid media, social, and client-reporting metrics.

A connector can reduce reporting friction. An agency can create one template, filter it by client or market, and give stakeholders a consistent monthly view. A central dashboard can also make platform-level differences visible: for example, a brand could lead on Perplexity but trail on ChatGPT and Grok.

But dashboards should remain a layer above the evidence, not the evidence itself. A report showing declining share of voice should provide a path to the affected prompts, raw answers, competitor mentions, and source pages. Without that drill-down, the dashboard explains what changed but not why.

Ahrefs’ connector documentation lists fields including platform, brand, mentions, impressions, share of voice, and date-based history. Teams should define their internal metric dictionary before exposing these figures to executives. In particular, “impressions,” “mentions,” and “share of voice” should never be presented as interchangeable measures.

The other April 2026 Ahrefs changes matter for supporting research

Although Brand Radar dominates the AI visibility story, the other April updates can support the work around it.

Site Explorer Organic positions shows every ranking URL for organic keywords, unlike the Calendar-oriented positions view that focuses on daily movement. That can help identify which existing pages already have conventional search visibility before we decide whether to improve them for AI answer coverage.

The new Crawled pages report lists pages AhrefsBot has crawled and sorts them by the latest crawl attempt. We can use it to audit page discovery, spot crawl issues, and monitor competitor content that may become relevant before it ranks prominently.

Keywords Explorer historical total search volume for multiple keywords or saved lists adds demand context. It does not measure AI visibility, but it can help distinguish a declining AI mention rate from a broader seasonal shift in category interest.

In Google Search Console reporting, Ahrefs added a Word Count filter and a Rank Tracker status filter, while GSC Insights became available in Portfolios for combined reporting across properties. The Social Media Manager Overview was redesigned to show top-line KPIs, scheduled posts, channel status, and top posts by views.

These are adjacent capabilities, not substitutes for prompt-level AI monitoring. We should avoid making causal claims such as “a higher word count caused more AI citations” without answer-level evidence.

Hosted Brand Radar versus a local-first visibility workflow

Hosted platforms such as Brand Radar are attractive when a team wants broad coverage, packaged reporting, and a centralized interface. Ahrefs offers broad prompt discovery, custom prompt tracking, citation data, API access, and connector-based distribution in one ecosystem.

A local-first tool solves a different set of priorities. With AI Visibility Tracker, we run a desktop workflow using the customer’s own API key to track how ChatGPT, Claude, Gemini, Perplexity, and Grok answer real buyer prompts. That model can offer more direct control over prompt inputs, run history, raw outputs, spend, and data handling.

The trade-off is straightforward:

• Hosted suite: faster access to bundled datasets, dashboards, and surrounding SEO tools.
• Local-first workflow: more control over API credentials, experiment design, raw-response retention, and reproducibility.
• Best operational choice: depends on whether broad index coverage or controlled prompt experimentation is the immediate priority.

For brands that already use a hosted platform, a local-first tracker can serve as an independent verification layer for the prompts that matter most. For teams starting with a tight list of high-value buyer questions, it can be the primary measurement system. Our practical framework for selecting metrics is covered in AI search measurement: a practical system for measuring brand visibility.

FAQ

What changed in Ahrefs Brand Radar in April 2026?

Ahrefs added Grok support for its prompt dataset and custom prompts, rebuilt the Cited Pages report, introduced Found in and Cited in citation-funnel metrics, added historical citation-position tracking for chatbot models, and launched a Brand Radar Looker Studio connector. The release also included expanded API capacity and new Brand Radar API endpoints.

How does Grok work in Brand Radar?

At the April 2026 launch, Ahrefs made Grok available in Brand Radar for both Ahrefs prompts and custom prompts. It could be included through the All Indexes add-on or purchased separately, while custom-query tracking required monthly checks rather than that add-on. Later Ahrefs documentation noted a temporary inability to gather new Grok data due to policy changes, so availability should be confirmed before relying on it.

Is Grok available for custom prompts as well as Ahrefs prompts?

Yes—Ahrefs’ April 2026 release specifically stated that Grok was added for both Ahrefs prompts and custom prompts. Custom prompts let teams specify buyer questions, AI platforms, locations, and refresh frequencies. However, operational availability can change, and Ahrefs later flagged temporary restrictions on collecting new Grok data.

How much did Ahrefs increase its API limits?

Lite plans received four times the prior API-unit allowance and ten times the previous row limit per request. Standard increased from 150,000 to 400,000 units and from 25 to 250 rows. Advanced increased from 500,000 to 1 million units and from 100 to 500 rows per request.

How does the Grok addition affect AI citation tracking and competitor analysis?

Grok adds another answer environment where brands may be mentioned, recommended, omitted, or displaced by competitors. The benefit comes from comparing results by engine and prompt, not blending everything into one score. Track whether your brand is named, whether it is cited, which competitors appear, and which pages or third-party sources shape the answer.