Ahrefs plans compared against the add-ons priced on top, and whether the AI half is worth buying

The five plan figures are the smaller half of the decision, because Brand Radar, Report Builder, traffic reports, project boosts and every seat after the first are each priced separately, all read on 31 October 2026.

AI Knows Us · Updated 31 October 2026

Ahrefs is worth buying when the thing you actually need is its link index, and the plan price is the smaller half of that decision. Read on 31 October 2026, the pricing page publishes five plan figures in US dollars and then prices another eight things on top of them, including almost all of the AI visibility half of the product. Two teams can both say they pay for Ahrefs and be 500 dollars a month apart, so the useful question is not the sticker, it is which of those extra lines you will switch on.

Everything below was read in the served HTML of ahrefs.com/pricing and ahrefs.com/brand-radar on 31 October 2026. Both returned HTTP 200 and both bodies were saved. Figures move, so carry the date with the number whenever you quote one of these back to a salesperson.

The plan ladder as published on 31 October 2026

PlanPrice a monthUsers includedEach extra userAI prompt checks a monthTracked keywords
Starter29 dollarsnot enumerated on the pagenot statednot statednot stated
Lite129 dollars140 dollars, up to 2 more150750
Standard249 dollars160 dollars, up to 5 more3002,000
Advanced449 dollars180 dollars, up to 10 more6005,000
Enterprise1,499 dollarsfrom 3100 dollars, unlimitedfrom 2,500from 10,000

Two lines in that table do more work than the prices. Every tier below Enterprise includes exactly one user, so a three person team on Lite is 129 plus 80, and the same team on Standard is 249 plus 120. And the AI prompt quota at the bottom of the ladder is 150 checks, which is a sample rather than a monitoring programme, for reasons the arithmetic below makes plain. The page also states Pay annually, save up to 17%, and Enterprise carries the line Annual commitment required, so the cheapest route to the top tier is a year.

The eight things priced on top of the plan

This is the part a plan comparison usually misses, and it is where the bill actually moves. Each of these is published as a separate monthly figure beside the plan ladder.

Add-onPublished priceWhat the figure buys
Brand Radar AI Visibility Index199 dollars a month83 prompts a day, plus 2,500 checks a month, overage at 0.020 dollars a check, all platforms with Claude available
Custom prompt package, Basic50 dollars a monthplus 2,500 checks a month, overage 0.020 dollars a check, example given as 80 prompts daily on 1 platform
Custom prompt package, Growth100 dollars a monthplus 7,000 checks, overage 0.015 dollars, example given as 100 prompts daily on 2 platforms
Custom prompt package, Scale250 dollars a monthplus 25,000 checks, overage 0.010 dollars, described as the lowest per check rate
Report Builder99 dollars a month50 reports, 500 widgets, scheduling
AI Content Helperfrom 99 dollars a monthcontent grader and inventory
Traffic reports99 dollars a month, or 299 for traffic and markettraffic sources including AI traffic, then market share and audience overlap at the higher figure
Project Boost20 dollars a month a project, or 200 for Boost Maxpatches, IndexNow submission, batch AI in Site Audit, always on auditing at the higher figure

One more figure is worth recording because it is printed without a unit. The Brand Radar page carries the banner 699 dollars a month for all models beside the line Starts at 50 dollars a month, and does not state the prompt count that figure buys. If you are going to track every model, that is the number to ask about before the call ends.

What one AI check buys, in Ahrefs own units

Ahrefs publishes the unit, which is rarer in this category than it should be. The pricing page states that tracking one prompt on one platform in one location uses 1 check. The Brand Radar page is more specific again: each prompt is checked daily on every selected platform, 5 prompts on 1 platform equals 150 checks a month, and Claude consumes 8 checks per update.

So the arithmetic is yours to do before you pay anybody. Five prompts tracked daily on one platform consumes the entire Lite quota. The same five prompts tracked daily on Claude consumes about 1,200 checks a month, which is eight times that quota, and at the published overage rate of 0.020 dollars a check that is roughly 24 dollars added to the bill, or 100 dollars for the Growth package that covers it with room to spare. Custom Prompts also lets you pick the cadence per platform, daily, weekly or monthly, so the honest way to size this is prompts multiplied by platforms multiplied by cadence, with Claude counted eight times.

Worth separating from that: the Rank Tracker line says Weekly on all four tiers, while the Brand Radar AI prompt line says Daily on all four. The AI half of this product refreshes faster than the search half.

Where Ahrefs beats us, and where the two answer different questions

The criterion here, applied to every vendor named on this page including ourselves, is what the company publishes on its own live page that a buyer can check without a sales call. On that test Ahrefs wins three lines outright. It publishes an annual saving in the served page, where our annual figure sits behind a button a crawler will not press. It publishes a daily AI cadence on every tier, where we check weekly on your priority questions and at 30 and 90 days on the rest. And it sells seats at all: every one of our plans includes one login with full access, and we publish no single sign on, where Ahrefs prices a second user at 40 dollars.

The products also answer different questions, which matters more than the scoreboard. Brand Radar is built on an index of 519 million or more prompts that Ahrefs ran and stored, which is the right instrument for seeing a whole market at once. Our own readings are of a question set written for one business, asked on ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews, and our pricing page states the unit in words: one credit is one buyer question, asked on one AI app, once, with a Claude check costing 3. Both of those are legitimate measurements and they have different denominators, so do not average them.

Which plan to ask for, by what you are actually buying

  • Links, keywords and audits for one person. Lite at 129 dollars. The free route inside Ahrefs covers a domain you can verify, and the free keyword tool against the paid Keywords Explorer sets out where that stops.
  • A team of three. Lite plus two seats is 209 dollars against Standard plus two at 369, so buy the limits you need rather than the tier that sounds right.
  • Market level AI visibility. The 199 dollar index, because the plan quotas of 150 to 600 checks were never meant to carry it.
  • Your own buyer questions tracked on Claude. A custom prompt package, sized with the eight times Claude weight in the arithmetic, or a tool that prices Claude in its subscription.
  • A switch away from Ahrefs entirely. That is a question about replacing an index rather than an interface, and it is answered in Ahrefs alternatives and the index you are actually replacing.

Questions buyers ask before paying for Ahrefs

Is Ahrefs worth it if I only want to know whether assistants name me?

On its own the subscription is a thin instrument for that job, because the included quota is 5 prompts on Lite rising to 20 on Advanced. The add-ons are the real product there, and they are priced in checks, so compare them against a dedicated tool on cost per recorded answer rather than on monthly sticker.

Does the 129 dollar plan include a second seat?

No. Lite, Standard and Advanced each include one user, and extra users are 40, 60 and 80 dollars a month respectively, capped at 2, 5 and 10. Enterprise starts from 3 users with unlimited additions at 100 dollars each.

Why does Claude cost eight times more to check?

Ahrefs publishes the weight rather than the reason: Claude consumes 8 checks per update where the other platforms consume one. Our own pricing page prices the same engine at 3 credits against 1 for the same stated reason, that Claude costs more to run, so a buyer comparing the two should compare the multiplier and not the headline.

Can I predict the AI bill before I sign?

Yes, and that is the strongest thing about the way Ahrefs publishes this. Multiply prompts by platforms by refreshes a month, count Claude eight times, then read the result against the package quotas and the per check overage rates of 0.020, 0.015 and 0.010 dollars.

Is the annual discount worth taking?

The page states up to 17 per cent, and Enterprise requires an annual commitment regardless. On the lower tiers the saving is real money against a product whose limits you can read in advance, so the question is only whether your prompt set will still be the right one in twelve months.

The ladder nobody quotes you

The thing to take away from a morning spent in these two documents is that Ahrefs has split one product into a plan and eight purchases, and published every figure for all of it. That is a good deal more honest than a demo, and it also means a quote built only from the plan row is wrong by design. Write the eight lines down, tick the ones you will switch on, and put the total next to what any single purpose tool asks for the same job. When the job is specifically whether ChatGPT, Gemini, Perplexity, Claude or Google AI Overviews put your name in the answer, ours is the measurement built around that one question and it prices Claude inside the plan. Ahrefs is the better buy on everything the link index does, and the two bills are not the same bill.

Read next

Carpet cleaning advertising, the four channels that produce calls and the Google product that bills you per leadCarpet cleaning services is a named Local Services Ads category, read from Google's own help page on 31 October 2026, which changes the order a cleaning business should spend in.Marketing certifications compared on what each issuer publishes before you pay, with the prices read on 31 October 2026Four issuers, one criterion that anybody can check, and the two that publish no exam price on their own pages at all.Ecommerce consulting firms, the four engagements they sell and how advisory differs from hiring an agencyWhat a consultancy is for when the problem is the decision rather than the execution, the four shapes the work takes, how each is priced, and the deliverable to insist on.Promoting a YouTube Short, the one feed money cannot put you in and the test that decides whether to pay at allPaid distribution and feed distribution are two different products for a Short, so this is what each actually delivers, and the retention reading to take before a budget is committed.A Google Ads charge on your credit card, why several arrive in one month and how to identify one you do not recogniseGoogle's own billing documentation quoted on 31 October 2026, including the threshold mechanic that produces multiple charges, the first of the month charge, and the order to work through an unexpected one.SEO ROI calculators, the six inputs that decide the answer and the two that most of them quietly assumeThe arithmetic written out so you can build it in a spreadsheet in ten minutes, plus the assumption that makes a published calculator flattering and the honest way to replace it.

Or browse the whole archive.

See what AI says about you.

The first scan is free, it needs no account, and it asks the questions your buyers ask.

Type your website address to continue.

Free. No card. We ask 5 real buyer questions on 2 AI apps.

Already sure? Start free. No card needed, and the free plan stays free. Pay only when you want the work. Or talk to a person.

Or read how we fix what it finds.