Seven of twelve sanctions-screening vendors publish a price. We read every rate card and show the arithmetic

Seven of the twelve sanctions-screening vendors we checked on 4 September 2026 publish a price on their own website, and five do not. Where a number exists it usually meters something other than a screening decision (an API call, a search, a screen, a query, a monitored entity), so two published prices are rarely comparable until you restate both in your own unit. Across the vendors that publish, one screening unit runs from €0.01 per API call at 120,000 calls a month to €0.45 per check at the top of one published range. Every figure below was read off the vendor's own page that day. None comes from a directory listing, a review site or a sales call.

Quoted as written, from the vendor's own page. All checked 4 September 2026.

Two rows need a caveat. Ondato's figures are real, but they sit in an identity-verification rate card as add-ons, and the volume that moves a buyer from €0.45 to €0.04 is not stated. dilisense's two cheapest and two dearest tiers sit behind an interactive volume selector on the page; the full ladder is enumerated in the page's own embedded pricing data, which is what we quote. The same data says the 100 free Starter calls are per month (100 free API calls per month), which we confirmed on 1 October 2026 when correcting the arithmetic below.

Annual cost for a book of 500, 5,000 and 50,000 checks, using each vendor's cheapest published route to that volume. Currencies are not converted: dollars stay dollars, euros stay euros.

Read the units before the totals. These rows are not five prices for one product. A dilisense or OpenSanctions unit is a lookup against a dataset; a SanctScan or Verifex unit is a screen inside a hosted workflow that also meters monitored entities separately; the last row's unit is a three-domain check that produces a point-in-time evidence PDF. A cheap row and a dear row can both be correct answers to different questions. We put ScreenVeritAI in the table on the same terms as everyone else, and it is the most expensive line in the 50,000-check column. We would rather you saw that here than in a procurement meeting.

Two vendors that publish figures are left out of this table, for stated reasons. ComplyAdvantage meters monitored entities, not screens: "from $99 per month" for 100 entities cannot be divided by a number of checks without inventing a relationship between the two. Ondato publishes a range whose volume bands are not published, so 500 checks resolves to anywhere between €20 and €225, an eleven-fold spread that is not decision-grade.

Footnote — the arithmetic. A book of 500 a year is ~42 a month; 5,000 is ~417; 50,000 is ~4,167.

dilisense: Starter's 100 free calls renew each month, and these figures assume checks spread evenly across the year; a month above 100 bills the excess at €0.10. At 42 a month the book sits inside the free calls, so €0. At 5,000, 5,000 − 1,200 free = 3,800 × €0.10 = €380 (the €150/mo Basic tier would cost €1,800). At 50,000, Basic's 3,000 calls a month fall short, and the page publishes no rate for calls above Basic's allowance. If the extra 14,000 bill at Basic's own €0.05, Basic costs €150 × 12 + €700 = €2,500; that rate is our assumption. On published terms alone the cheapest route is Professional, €300 × 12 = €3,600 for 10,000 calls a month.

(Corrected 1 October 2026. The 4 September version showed €50 and €500, charging every call at €0.10, and gave €2,500 at 50,000 without flagging the assumed rate.)

SanctScan: the free plan caps at 25 searches a month, so 42 a month needs Solo, $19 × 12 = $228; 417 a month exceeds Solo's 250, so Starter, $39 × 12 = $468; 4,167 a month exceeds Starter's 1,500, so Growth, $149 × 12 = $1,788.

Verifex: 42 a month fits the 50-screen free plan, $0; 417 needs Starter, $49 × 12 = $588; 4,167 exceeds Starter's 2,500, so Growth, $99 × 12 = $1,188.

OpenSanctions: 500 / 5,000 / 50,000 × €0.10.

ScreenVeritAI: 500 / 5,000 / 50,000 × €0.39; the rate does not change with volume, and monthly plans on /en/pricing price the same checks differently.

A per-check price answers one question, what a new lookup costs. Four other meters decide the invoice.

Ongoing monitoring of the book you already have. This is a stock, not a flow, and it grows every month you stay in business. dilisense prices it apart from screening entirely, at €0.15 per name per month or €150 a month for 1,000 names, so a 1,000-name book costs €1,800 a year before a single new check is run. SanctScan caps monitored entities per plan (25, 150, 1,000) and charges $0.30 per monitor overage, falling to $0.15 on Growth. ComplyAdvantage's Starter price is a monitoring price. Verifex includes monitoring on all plans. Four vendors, four different answers to the same line item.

Re-screens forced by list updates. When an authority adds, amends or deletes entries, a defensible programme re-runs the affected book, and a re-run is billed like any other screen on a per-call meter. Re-screen a 2,000-name book once a month and you add 24,000 calls a year that have nothing to do with new customers. At €0.10 that is €2,400; at €0.03 it is €720; on an unlimited-search plan it is nothing. Ask which of those your quote assumes.

Adverse media as its own meter. dilisense sells adverse media as a separate API with its own ladder (€0.10 per call at Starter, €300/mo with 10,000 calls included at Professional), and so do we: the Adverse Media Check is its own €0.99 line. Ondato folds it into its AML screening line. A quote that says "screening included" and stays silent on media is describing half a workload.

Seats. Nobody in this survey publishes a per-seat price. It is the line most likely to appear for the first time in a contract, and the one worth asking about earliest, particularly where a second reviewer is a control your policy already requires.

Publishing a number and being findable are different achievements.

When we searched dilisense pricing on the survey date, the vendor's own page was not the first result: SourceForge and Slashdot directory listings ranked above it, even though dilisense publishes a complete five-tier ladder. A buyer, or a model answering on a buyer's behalf, reads a third-party summary of a price the vendor states itself.

Descartes publishes A Pricing Guide to Denied Party Screening Software Solutions. It gives ranges by company size (up to $5,000, up to $20,000, $100,000+), attributes them to the category rather than to Descartes, prices no Descartes product, and closes by inviting the reader to "contact our export compliance experts today". The figures that circulate for Descartes come from software directories instead.

NameScan's price is presumably knowable; it never reached the HTML we were served. A crawler, a screen reader or a language model fetching that page on the survey date saw {{currencyPrice}}. Sanction Scanner's pricing URL was still indexed under a "Plans & Pricing" title but returned 404 to anyone who followed it.

For a buyer shortlisting on a Tuesday afternoon, all four come to the same thing. A price behind JavaScript, behind a form, below a directory listing or behind a broken URL might as well not be there.

For any vendor in the quote-only column, get these in writing before a demo, not after.

1. What is one billable unit? A name submitted, an API call, a matched entity, or a monitored counterparty, and does a call that returns twenty candidates cost the same as one that returns none?
2. What is the overage rate, in the same units, above every included allowance?
3. What does monitoring my existing book cost per year at my current customer count, listed separately from new screening?
4. Are re-screens after a list update billed? If yes, at what rate, and is there a cap?
5. Is adverse media on the same meter as sanctions and PEP, or a second contract?
6. How many seats are included, and what does the next one cost?
7. What is the minimum term and the minimum annual spend, the two numbers that turn a per-check price into a real one.

If a vendor will not answer 1, 2 and 4 in an email, the published price of anyone who does is worth more than the discount you are being offered.

Prices move, pages get rebuilt, and a survey with a date on it is only useful if the date keeps changing. We re-check this table quarterly: every URL is fetched again, the quoted figure is restated as it stands on the day, and the previous figure is kept with its date rather than silently overwritten. Where a vendor starts publishing, the row moves out of the "not published" column with the date it moved.

Next scheduled check: 4 December 2026. If a figure here is wrong on the day you read it, the vendor's page is the authority, not this post; every URL is in the table for exactly that reason.