Prefer: AI Visibility Tracking and GEO Content

Prefer is an AI visibility tool for brand, SEO and content teams. It brings recurring answer monitoring, citation analysis, competitor gaps and content work into one web workspace, helping a team decide what to improve after seeing how AI search describes its business.

What does Prefer actually do?

Imagine you sell project management software. A prospective customer asks ChatGPT which tools suit a small team. Prefer runs selected questions across supported engines, records whether they name your brand, and shows which competitors and sources appear instead. This makes repeated manual checking easier to organize and turns missing mentions into work you can investigate.

The workflow is relevant to generative engine optimization (GEO) and answer engine optimization (AEO): observe answers, inspect evidence, improve content and sources, then measure again. In-house teams need someone who can edit the website and review recommendations; agencies should also compare client, project and reporting capacity.

Read visibility, mentions and citations separately

Self-serve plans cover ChatGPT, Gemini, Perplexity, Google AI Overviews and Google AI Mode. Prefer stores full answers, named brands, citation URLs and sentiment. Views can be filtered by engine, topic, persona, region and date.

Visibility Score is the percentage of scoped answers naming your brand. Share of Voice compares your mentions with all brand mentions, while Citation Share compares links to your domain with all citations. A brand mention does not necessarily include a link to your website. Sentiment describes the AI answer's tone rather than a customer survey.

Delta compares the previous tracking run, which may have happened earlier the same day. These figures depend on your prompt set and filters; they do not measure every real customer conversation, website session or sale. Check the official metric definitions before reporting changes.

Find the sources behind competitor visibility

Source analysis separates owned pages, competitor sites, earned coverage, reviews, roundups, directories and communities. In Competitors > Sources, select a rival and inspect domains with high citation share. Open a prompt's exact answer and citation links to verify the evidence.

The result can inform an owned comparison page, a correction to an outdated review, or a relevant directory submission. Inspecting the source gap first helps avoid producing another article when the missing information actually lives elsewhere.

Prioritize actions and review content drafts

Action Center presents impact-scored cards with evidence and steps, organized on a Kanban board. Its categories cover comparisons, off-page sources, community work, crawler access, schema, entity consistency and analytics. Some recommendations come from tracking; technical suggestions can also come from an audit or templates, so inspect their provenance.

Content Studio offers new drafts, URL-based page optimization with accept/reject diffs, and Content Gap reports based on the latest run. Research and reference-page tools help build a brief before drafting. Scores assist coverage and structure review; a person must still check facts, examples and claims.

Current detailed documentation says Action Center work is manual. It also says the WordPress connection UI is ready but the actual publishing step is not live. Export or copy reviewed content into your CMS for now. The Content Studio documentation is useful when checking release status.

A practical first monitoring cycle

1. Prepare your brand domain, positioning, competitors, target region and customer persona.
2. Create a workspace with Google or an email magic link. Review suggested topics and prompts, keeping questions customers would ask before choosing a supplier.
3. Run tracking and use Monitor > Overview for the baseline. Read prompt details to check the full response and sources.
4. Accept one evidence-backed action, edit the relevant page manually, and publish the change.
5. Run tracking again with comparable prompts and filters. Completed action cards can show citation changes after a later run; use your own analytics to check visits and conversions separately.

The official quickstart explains setup and navigation. A small, well-chosen prompt set gives a more interpretable baseline than a large collection of generic questions.

Free audit, trial and paid capacity

Checked on October 4, 2026, the free audit is a one-time snapshot without a credit card, separate from recurring tracking. There is no free-forever full plan. Paid plans offer a seven-day no-card trial capped at 30 tracked prompts.

Monthly brand subscriptions are USD $24 Starter, $129 Growth and $399 Scale. They allow 30/150/500 prompts, 15/25/50 AI articles per month, 3/10/25 competitors and 1/2/5 projects respectively. A prompt uses one slot across included engines. Annual billing is prepaid and its monthly equivalent is not the monthly checkout price.

Brand plans require a tier upgrade for more capacity rather than automatic overage billing. Upgrades apply immediately; downgrades apply at renewal, and cancellation is available through billing. Enterprise and managed work are custom quoted. Review current pricing for taxes, final totals and agency terms.

Connections, privacy and suitability limits

Google Search Console and Analytics connections are described as read-only, with revocable access and no sale or sharing of that data with other customers. Slack and CSV/PDF exports are Growth and above; warehouse sync is Enterprise. Public pages disagree about some MCP/API entitlements, so confirm the actual account access before building an integration.

The privacy policy also covers account, domain, device, analytics, cookie and provider data. It does not justify a blanket zero-retention, no-training or security-certification claim. Prefer's Trust page does not guarantee rankings, citation placement or revenue. Qwen, DeepSeek and Claude API additions are Enterprise features; regional engine and native interface coverage should be tested separately.

Frequently asked questions

Is Prefer free or subscription based?

The one-time audit is free. Recurring tracking requires a subscription after the trial, with plan-specific capacity limits.

How can I check whether AI mentions my brand?

Run representative prompts, then read the full answer and citation links in prompt details. Compare engines separately and keep the same scope.

Which plan capacity should I compare?

Check tracked prompts, monthly articles, competitors and projects, alongside monthly versus annual billing. These limits determine the practical cost for your workload.

Official resources

Use the Prefer documentation to verify workflows, release status and permissions. Evaluate your own answers and source evidence during the trial rather than treating an example dashboard as proof of future results.