Paid IQ tests and the billing model behind them

0 Quick Answer

A paid IQ test charges you in one of four ways, and the way it charges you matters more than the number on the button. The four are a single one time payment, a recurring subscription, a low cost trial that converts into a subscription, and a free score with the report behind a paywall. The first is bounded. The other three are not, and the twelve month cost of two products that look alike at checkout can differ by more than a factor of thirty.

Prices below were read on the providers' own pages on September 5, 2026, and they change. Cerebrum IQ's published terms describe a seven day trial for a one time charge of one dollar, followed on day eight by a plan billed weekly, monthly, or quarterly, with a weekly Premium rate of 14.99 dollars. IQScore publishes a one time pack at 6.99 dollars and states that it has no subscriptions and no recurring billing. BRGHT advertises unlimited testing starting at 19.95 dollars. ACIS charges 15, 30, or 50 dollars once, with no renewal.

What follows is the mechanics: what money actually buys in a cognitive assessment, why a real normative study is the honest reason a serious test is not free, how each billing model prices out over a year, what refund and cancellation policies really say when you read them, and the questions worth asking before a card number goes into a form. This page does not rank tests and does not price a psychologist. Those live elsewhere and are linked below.

1 What Money Actually Buys on an IQ Test

Payment does not buy accuracy, and any page that implies otherwise is selling the wrong thing. A cognitive assessment is a measuring instrument. What separates one from a quiz is a chain of specific, expensive work: writing items, trying them out on people, discarding the ones that do not discriminate, assembling a form, administering that form to a reference sample, converting raw performance into a standard score, and estimating how much the resulting number would move if the same person sat the test again. Money can fund that chain. It does not substitute for it.

That is why the useful question is never how much a test costs but which parts of the chain the price paid for. A twenty dollar test that ran an item analysis and publishes reliability coefficients is buying you something. A twenty dollar test that licensed a matrix puzzle set and wrote a scoring formula is charging you for delivery, not for measurement. Both look identical at checkout, and the checkout page is exactly where the distinction is invisible.

There is a second thing money buys, and it is worth naming because it is real. A funded product has a support address, a named legal entity, a privacy policy that somebody wrote on purpose, and a report that gets updated. Free instruments frequently have none of that, which is not a moral failing but is a practical limit. When a charge appears on a statement, having a counterparty to contact is worth something.

The third thing, and the one buyers usually want, is the report. A bare number is close to useless. What makes a result interpretable is the surrounding apparatus: a percentile, a confidence interval, a domain profile, and a statement of what reference group the comparison was made against. Whether a given paid test supplies that apparatus is checkable before you buy, because responsible providers publish an example report. The distinction between a score generator and an interpretable result is worked through on the comparison of free and validated instruments, and the components of a real report are itemized on the page on reading a score.

2 Why a Real Normative Study Is Expensive

The honest reason a serious test is not free is that its reference sample had to be recruited, screened, and individually tested by a person. This is not an abstraction, and the scale is documented. Gary L. Canivez, reviewing the WAIS-IV for the Buros Center's Mental Measurements Yearbook in 2010, records a standardization sample of 2,200 people drawn by stratified proportional sampling across age, sex, race and ethnicity, education level, and geographic region, matched against the October 2005 United States Census and divided into 13 age groups.

Now attach the administration time. The same Buros entry lists the WAIS-IV administration at 59 to 100 minutes, one examiner to one examinee. Multiply that by 2,200 and the standardization phase alone represents roughly 2,200 to 3,700 hours of individual professional time, before item writing, before the tryout rounds that produced the item set, and before a single analysis was run. That arithmetic is ours, from the two published figures, and it is the cheapest possible reading of the cost, since it counts only the hours in the room.

The retail price reflects it. Canivez's entry lists 2008 price data of 1,079 dollars for the complete WAIS-IV kit. As of September 5, 2026, the Pearson store lists the WAIS-5 Complete Kit at 1,784.00 dollars, with a softcase version at 1,889.00 dollars. Nobody pays that for a single test. It is the price of the apparatus, amortized across the examinees a clinician will see with it.

Online instruments do not carry that cost structure, and pretending otherwise would be dishonest. Nothing sold for twenty dollars over the internet funds a census matched sample of 2,200 individually tested adults. What a well built online test can do is document what it did instead, publish its reliability estimates and its reference frame, and let the buyer judge. How a reference sample turns raw performance into a comparable number is set out on the norming page, and why the same score carries different precision on different instruments is on the reliability and validity page.

3 The Four Billing Models Priced Over Twelve Months

Put the models side by side at their published rates and the range is not subtle. Every figure in the table below was read on September 5, 2026 from the provider's own page, and the twelve month column is the stated rate multiplied by the number of billing cycles a year contains. It assumes nobody cancels, which is the assumption the model is built on.

Read the last column, not the third. The weekly plan and the one time payment differ by about five dollars at checkout and by more than 750 dollars a year, and nothing on a checkout screen makes that gap visible. The gap is the product of the cadence, and cadence is a design decision the seller makes and the buyer inherits.

Notice also what the third row does not say. BRGHT's homepage advertises a price. Its published terms and conditions, last updated May 2023, contain no price, no renewal clause, no cancellation mechanism, and no refund policy. The words refund, cancel, subscription, and trial do not appear in the document. That silence is itself a fact about the product, and it is examined in the full BRGHT review.

One clarification about what this table is for. It compares billing structures, not test quality. A subscription product can be well built and a one time purchase can be worthless. If you want tests ordered by evidence rather than by payment model, that job belongs to the ranking page, and choosing by situation belongs to the decision guide.

4 What a One Time Payment Actually Buys

A one time payment matches the shape of the thing being sold, which is why it produces so few complaints. An IQ assessment is an event. You sit it, you get a result, and the result does not decay next month. Almost nobody wants to buy twelve of them a year. A charge that happens once and then stops is the billing model that corresponds to that reality, and every other model is trying to convert an event into a relationship.

The trade-off is real and worth stating. A one time seller has to make its money on the first transaction, so it has a stronger incentive to price higher at checkout and a weaker incentive to keep improving the product afterward. A subscription seller has the opposite incentives. Neither is virtuous by construction. The difference for a buyer is that the one time model has a known worst case, and the subscription model does not until you read the cancellation clause.

Two current examples make the shape concrete. IQScore states on its homepage that the score, percentile, and a four domain breakdown are free, with the IQScore Pack at 6.99 dollars covering a certificate, a written report, and worked explanations for all 36 questions, and it says directly that there are no subscriptions and no recurring billing anywhere. ACIS charges 15 dollars for a six subtest form, 30 dollars for a thirteen subtest form, and 50 dollars for the full twenty subtest battery, once, with access to complete the session open for 30 days.

If you are choosing on this axis alone, the search phrase people actually use is an IQ test with no subscription, and it is a reasonable filter. It rules out a category of billing risk in one step. It does not rule out a bad test, because plenty of one time products publish nothing about their norms. Pair the filter with the evidence questions later on this page, and with the standards laid out on the accuracy guide.

5 What a Subscription Buys That a Purchase Does Not

A subscription is the right model when the product is continuous, and a single assessment is not continuous. This is the fairest way to evaluate it. Ask what recurs. If the answer is training exercises, progress tracking across months, repeated administrations of parallel forms, or access to a suite of instruments, then a recurring charge is describing a recurring service and the model is honest. If the answer is one report delivered once, the recurrence is describing the seller's revenue, not the buyer's use.

Cerebrum IQ is the clearest published case, and its own terms are explicit about what the subscription contains. The document describes a training curriculum alongside the assessment, and states that the training content is based on general cognitive-exercise principles, that its effectiveness has not been independently validated, that individual results vary, and that progress inside the product may not translate to measurable real-world cognitive change. That is a company describing the recurring half of its own offering in candid terms, and the candor deserves credit.

The published price ladder, read on September 5, 2026, is weekly Basic at 9.99 dollars, weekly Premium at 14.99 dollars, monthly Promotional at 14.99 dollars, monthly Standard at 29.99 dollars, and quarterly Premium at 39.99 dollars. The terms add that the checkout page shows the actual regional price before confirmation and that the final rate depends on the country, the promotion, taxes, and currency conversion. That last clause is why no review can tell you your own price, and why the checkout screen is the only authoritative number.

There is one more question worth asking of any subscription attached to a cognitive test, and it is a measurement question rather than a commercial one. Repeated administration of the same instrument produces practice effects, so an improving score across a subscription is partly the instrument being learned rather than the person changing. Whether measured ability moves at all under training is a contested empirical question covered on the page on improving an IQ score, and the reason retest gains are not simple is on the preparation page.

6 The One Dollar Trial Is a Billing Model, Not a Discount

A trial priced at one dollar exists to capture a card number, and the terms usually say so plainly if anyone reads them. Cerebrum IQ's published terms describe the mechanism without euphemism. When you start a Trial, the company authorizes your card for one dollar. The trial gives seven days of full premium access. On day eight, the plan you selected at sign-up starts at its standard rate and bills on its regular cycle, weekly, monthly, or quarterly. Only one trial per user is available.

Read the phrase the plan you selected at sign-up. The subscription is chosen at the moment the one dollar is paid, not afterward. Nothing further has to happen for the larger charge to arrive. Whether that constitutes a clear and conspicuous disclosure is a legal question, and the answer in the United States turns on the Restore Online Shoppers' Confidence Act rather than on any test specific rule, which is covered two sections down.

The pattern shows up on customer complaint records with dates and amounts attached. The Better Business Bureau profile for Cerebrum IQ, Inc., an unaccredited business listed at 1034 E Brandon Blvd in Brandon, Florida, records 85 complaints closed in the last three years and 59 in the last twelve months as displayed on September 5, 2026. One complainant, writing in May 2026, describes buying a test on November 14, 2025 for one dollar and being charged 29.99 dollars repeatedly through April 2026 for a total of 179.94 dollars. Another, in July 2026, describes cancelling a free trial and being billed 24.99 dollars monthly afterward.

BRGHT draws the same reports. Its Trustpilot page showed a TrustScore of 2.8 across 164 reviews on September 5, 2026, with 15 percent at one star. A United States reviewer on October 8, 2025 wrote of a one dollar trial followed by a 24.99 dollar charge. A reviewer in Great Britain on September 27, 2025 described one euro becoming a 49.99 euro charge days later. These are user reports rather than findings we verified, and reviewers sometimes confuse similar looking providers, which is why the provider's own terms are the better evidence. In BRGHT's case those terms are silent on billing entirely.

7 The Score Is Free and the Report Is Paywalled

The fourth model gives away the number and charges for the meaning, and it is the one most likely to be misunderstood as generosity. You take the test, a figure appears, and the interpretation, the domain breakdown, the percentile context, or the certificate sits behind a payment step. The buyer at that moment has already invested twenty to forty minutes and has a number they cannot read. That is a well designed moment to ask for money, and it is not by itself dishonest.

Whether it is fair depends on one thing: what was disclosed before the time was spent. A site that says on its landing page that the score is free and the report costs 6.99 dollars has told you the deal. A site that presents itself as a free IQ test and reveals a charge only after the last item is a different product. IQScore, read on September 5, 2026, is in the first category, publishing the score, percentile, and domain breakdown for nothing and pricing the pack up front.

There is a measurement point buried in this model that buyers rarely notice. IQScore states that a result is plotted against 46,951 completed assessments. That is a real and disclosed reference frame, and disclosing it is better practice than most of the market. It is also a frame made of people who chose to take an online IQ test, which is not the general population, and a percentile against that frame answers a narrower question than a percentile against a census matched sample. The distinction between a score and its reference group is worked through on the score versus percentile page and quantified on the tool that converts a score to a percentile.

• Disclosed before the test. The price, the payment step, and what the free portion contains all appear on the landing page.
• Disclosed after the test. The number appears, then a paywall does. The disclosure came after your time was spent.
• Disclosed as a subscription. The report unlocks with a plan rather than a purchase, and the plan renews.
• Not disclosed. A charge descriptor appears on the statement that does not match anything you remember agreeing to.

Certificates deserve a separate warning and have their own page. A printable document with a serial number is a design asset, not a credential, and no institution treats one as evidence. What documentation is actually accepted, and by whom, is set out on the certificate page.

8 What Refund and Cancellation Policies Really Say

Read the exclusion list, not the headline, because the exclusions are where the policy lives. Refund pages are written to sound accommodating in the first sentence and to enumerate the cases that do not qualify in the fourth. The exclusions are the operative text, and in a digital product they usually cover the exact situation a disappointed buyer is in.

Cerebrum IQ's terms, read on September 5, 2026, are unusually explicit and therefore useful as a worked example. The one dollar trial fee is refundable within thirty days if it is a first trial, no prior trial refund exists on any account, and the account is in good standing. Outside that window, refunds are issued for verified service unavailability, a confirmed platform bug reported within thirty days, a billing error such as a duplicate or unauthorized charge, or where local law requires it. The cases listed as not eligible are time not used after cancelling, plan downgrades, brief maintenance windows, a change of mind, and a charge that resulted from not cancelling before the renewal date.

That last exclusion is the one that matters. The most common complaint about a subscription is precisely a charge that arrived because the buyer did not cancel in time, and it is named as ineligible. The policy is not hidden and it is not unusual. It is simply the opposite of what most buyers assume when they read the word refund.

The BMI group, which operates the BMI Certified IQ Test at test-iq.org, runs a dedicated domain for this problem. BMIPAY.ORG exists to answer the question What is this charge for, identifies itself as the merchant descriptor for test-iq.org, test-personality.org, and mybrainforce.com, and offers order lookup by email address, order ID, or certificate serial number along with a self service cancel button. Its published position on refunds is that once the free trial period has passed the company normally cannot offer a refund, while cancellation stops future billing. ACIS reviewed that product separately on the Test-IQ.org review.

9 What Happens When You Dispute the Charge

Your rights against an unwanted charge come from payments law, not from the seller's refund page, and they run on a clock. This is the part most buyers discover too late. A refund is something a merchant grants. A billing error dispute is something your card issuer is legally required to investigate, and the two have nothing to do with each other.

In the United States the governing text is the Fair Credit Billing Act, codified at 15 U.S.C. 1666. It gives the cardholder sixty days after the creditor transmitted the statement containing the disputed item to send written notice to the address the creditor discloses for billing inquiries. On receiving it, the creditor must send a written acknowledgment within thirty days and must resolve the matter within two complete billing cycles, and in no event later than ninety days. The Federal Trade Commission's consumer guidance sets out the same procedure in plain language.

Two limits on that right are worth knowing before you rely on it. The statute covers billing errors, which includes charges you did not authorize, amounts that are wrong, and goods not delivered as agreed. It does not cover dissatisfaction with a product that arrived as described. A report you found shallow is not a billing error. A renewal you never consented to may be one.

Separately, the negative option rules that were supposed to make cancellation simpler are not currently in force. The Federal Trade Commission's Negative Option Rule, widely called the click to cancel rule, was vacated by the Eighth Circuit on July 8, 2025 on procedural grounds, and the Commission restarted rulemaking with an advance notice in January 2026. What remains in force is the Restore Online Shoppers' Confidence Act, 15 U.S.C. 8401 to 8405, which requires an online seller using a negative option to disclose all material terms clearly, obtain express informed consent to the charge, and provide a simple mechanism to stop recurring charges.

In the European Union the position is different and catches many buyers by surprise. Under Directive 2011/83/EU on consumer rights, Article 9 gives a fourteen day right of withdrawal from a distance contract without giving any reason. Article 16(m) removes that right for the supply of digital content not on a tangible medium once performance has begun with the consumer's prior express consent and acknowledgment that the right is lost. A test report delivered instantly is digital content whose performance has begun. The checkbox you clicked to see your score is very often the checkbox that ended your withdrawal right, and BRGHT, operated by Assessly BV of Amsterdam under Dutch law, sits squarely inside that regime.

10 How to Tell a Normed Test From a Gated Quiz

The difference is visible before you pay, because the evidence a normed test produced is either published or it is not. You do not need psychometric training to run this check. You need to know which four documents exist, and to notice that a product which invested in measurement will be eager to show them while a product that gated a quiz will have nothing to show.

The reference frame row does the most work. A percentile is meaningless until you know what group it ranks you inside. BRGHT's method page states that a raw score is compared to the entire population of test-takers to produce an IQ with a median of 100 and a standard deviation of 15, which is a clear and honest description of a self-anchored scale. IQScore states 46,951 completed assessments as its frame. Both disclose. Many competitors disclose nothing, and an undisclosed frame is the single most common gap in the paid market.

The confidence interval row is the fastest tell of all. Every professional score report prints one, because a score is an estimate with error attached. A product that reports 127.4 with no interval is telling you it either does not know its own measurement error or has decided not to mention it. What that interval represents, and why it widens at the top of the scale, is on the reliability page.

Finally, check who the seller is. A named company with a registration number in a jurisdiction with a public register gives you a counterparty for a dispute, a data request, or a refund. A contact form and a logo do not. This is not psychometrics, but it determines whether anything else on this page can help you.

11 Nine Questions to Ask Before You Enter a Card

Every one of these is answerable in under two minutes from the seller's own pages, and any that cannot be answered is itself the answer. Run them in order. The first four are commercial and the last five are about measurement, because a clean billing model attached to a worthless instrument is still a bad purchase.

• Is this one charge or a recurring one? If the word plan, pass, membership, or access appears anywhere near the price, assume recurring until the terms say otherwise.
• What is the renewal amount and the renewal date? Both must appear on the checkout screen. If they do not, that is a disclosure failure regardless of what the terms page says.
• What will appear on my statement? The merchant descriptor is frequently a different name from the website. Knowing it in advance prevents a dispute you did not need to file.
• How do I cancel, and by when? Find the mechanism and the cutoff before paying. A cutoff of twenty four hours before renewal means a calendar reminder at forty eight hours.
• What exactly is refundable? Read the list of cases that do not qualify, since that list is the policy.
• What is the reference group? If the site cannot say who your percentile compares you against, the percentile is decorative.
• What is the reliability, and in what units? A coefficient alone is not enough. Ask for the standard error of measurement expressed in IQ points.
• Can I see a full example report first? If the answer is no, you are buying a product you have not seen.
• What does the seller say the result cannot be used for? A provider that names its own limits has thought about them. A provider that names none has not.

One practical habit is worth more than all nine. Screenshot the checkout screen before confirming, capturing the initial amount, the renewal amount, the renewal date, and the currency in a single frame. If a charge is later disputed, that image is the evidence, and under the Fair Credit Billing Act procedure described above your written notice is stronger with it than without it.

12 Where ACIS Sits, and What a Buyer Should Do

We sell a competing paid assessment, so treat this section as a disclosure and check it against the rest of the page. ACIS charges once. Quick is 15 dollars and covers six subtests in about 45 minutes. Optimized is 30 dollars and covers thirteen subtests in about 110 minutes. Full Scale is 50 dollars and covers all twenty subtests in about 175 minutes, returning a Full Scale IQ with all six primary indices. There is no plan, no pass, and no renewal, and access to complete a purchased assessment stays open for 30 days.

The trial is free and requires no card. It runs five subtests covering working memory and processing speed, shows scaled scores and a provisional Full Scale estimate, and holds back the confirmed Full Scale IQ, the six domain profile, and the percentiles. That design exists because a card capture trial is the mechanism behind most of the complaints documented above, and the only way to not participate in it is to not take a card.

The published guarantee is five days. If the report does not deliver what the checkout page described, or a technical fault blocks access to the results, the purchase is refunded in full on request within that window. That is narrower than a general satisfaction guarantee and we would rather state the narrow version than imply the broad one.

On the measurement side, the numbers in the technical manual are these. The General Ability Index composite, built from fifteen subtests, has an omega reliability of .9885, a standard error of measurement of 1.61 points, and a g loading of .954. Verbal Comprehension reports omega .9745 with a standard error of 2.40. Fluid Reasoning reports .9727 and 2.48. Processing Speed, built from two subtests, is the weakest at .8790 with a standard error of 5.22, and we publish that rather than reporting only the strong composites. The full tables are in the technical manual.

The decision splits cleanly once the billing model is separated from the price, and most buyers never make that separation. A paid IQ test is worth buying when you want a structured estimate of your own cognitive profile, when you understand that an unsupervised result carries no institutional weight, and when the seller has published enough that you can judge the instrument before paying for it. It is not worth buying to obtain a credential, because none of these products produces one.

The billing question then resolves in a sentence. If the product is a single assessment, a single charge is the only model that matches it, and every other model requires a specific reason to accept. A subscription is defensible when a continuous service is genuinely attached and you intend to use it. A trial priced at one dollar is defensible when you have read what starts on day eight and have set the cancellation reminder before paying.

The measurement question resolves independently, and it should be asked second because it is slower. Look for a stated reference frame, a reliability coefficient with a standard error in score points, a downloadable example report, and a written statement of what the score cannot be used for. Four documents, checkable in a few minutes, and their absence is more informative than any marketing claim on the same page.

Three neighbouring pages carry the parts this one deliberately leaves alone. The ranking of online tests scores the market against a published rubric. The decision guide matches an instrument to a situation. The cost page covers the full price range including professional evaluation, which runs into the thousands and is a different purchase from anything discussed here. How a raw performance becomes a standard score in the first place is explained on the page on how IQ is calculated.

13 Sources Behind This Page

Prices and terms were read on the providers' own pages on September 5, 2026 and will change. Where a figure could not be verified from a primary document it is labeled as reported rather than confirmed.

• Cerebrum IQ, Terms and Conditions. Sections 4 and 5 set out the one dollar seven day trial, the weekly, monthly, and quarterly plan rates, the automatic renewal clause, and the list of cases excluded from refunds.
• Better Business Bureau, Cerebrum IQ, Inc. complaint record. Eighty five complaints closed in three years and fifty nine in twelve months as displayed on September 5, 2026, with dated amounts quoted by complainants.
• BRGHT, Mission, Vision, and Commitments. States 40 questions in 30 minutes, twenty difficulty levels calibrated from over 500,000 test-takers, and an IQ produced by comparing a raw score to the entire population of test-takers at a median of 100 and a standard deviation of 15.
• BRGHT, Terms and Conditions, last updated May 2023, operated by Assessly BV, Amsterdam, KVK 76121224. Contains no price, renewal, cancellation, or refund clause.
• Trustpilot, brght.org reviews. TrustScore 2.8 across 164 reviews with 15 percent at one star on September 5, 2026. User reports, not independently verified.
• BMIPAY.ORG. The merchant descriptor page for test-iq.org, test-personality.org, and mybrainforce.com, with order lookup, self service cancellation, and the stated position that a charge is normally not refundable once a free trial period has passed.
• IQScore. Free score, percentile, and four domain breakdown, with a 6.99 dollar one time pack, 36 questions in 25 minutes, results plotted against 46,951 completed assessments, and an explicit statement of no subscriptions and no recurring billing.
• Canivez, G. L. (2010). Review of the Wechsler Adult Intelligence Scale, Fourth Edition, in the Buros Center's Mental Measurements Yearbook. Standardization sample of 2,200 across 13 age groups, stratified to the October 2005 United States Census, administration 59 to 100 minutes, 2008 kit price 1,079 dollars.
• Pearson (2024). WAIS-5 store listing, complete kit at 1,784.00 dollars and 1,889.00 dollars with softcase as displayed on September 5, 2026.
• American Mensa, Testing. Supervised group testing at 60 dollars and private testing at 99 dollars as published on September 5, 2026, listed here because a proctored admission test is the closest priced comparison to a paid online product.
• 15 U.S.C. 1666, the Fair Credit Billing Act. Sixty days from the transmitted statement to send written notice, thirty days for the creditor to acknowledge, and two complete billing cycles not exceeding ninety days to resolve.
• Federal Trade Commission consumer guidance on billing for things you never received, which restates the dispute procedure in plain language.
• Restore Online Shoppers' Confidence Act, 15 U.S.C. 8401 to 8405. Requires clear disclosure of material terms, express informed consent to the charge, and a simple cancellation mechanism.
• Federal Trade Commission (2022). Bringing Dark Patterns to Light. The staff report on interface designs that obscure charges and renewals.
• Directive 2011/83/EU on consumer rights. Article 9 grants a fourteen day withdrawal right and Article 16(m) removes it for digital content once performance begins with prior express consent.
• Buros Center for Testing. The independent center whose Mental Measurements Yearbook has reviewed commercial tests since 1938.

The framework behind every judgment on this page is the Standards for Educational and Psychological Testing (2014), published jointly by the American Educational Research Association, the American Psychological Association, and the National Council on Measurement in Education. The Standards require that a test document its norms, its reliability, and the evidence supporting each intended use, and that score reports state the uncertainty attached to a score. The APA standards on test use place the matching obligation on whoever reports a result: say what the score supports, say what it does not, and do not let a price or a certificate do work the evidence cannot back.

14 Frequently Asked Questions