IP to Company API Pricing: Models, Costs and Hidden Charges

IP to company API pricing comes in three shapes: pay-per-call (cents per lookup, sometimes only when a company is found), monthly subscriptions with a call allowance, and enterprise contracts quoted by sales. The number that matters is the cost per correct company-level answer, which is always higher than the headline per-call rate.

The three pricing models

Model	How it charges	Typical fit	Watch for
Pay-per-call, prepaid	A fixed amount per lookup or per result, drawn from a balance	Variable or small volumes, experiments	Whether non-company answers are billed
Monthly tier	Flat fee with a call allowance	Steady volume	Overage rates, unused allowance
Enterprise licence	Annual contract, quoted	Large organisations, many data types	Minimums, seat requirements, auto-renewal

None is better in general. Pay-per-call suits a team that wants to test a day of traffic. A tier suits a team that already knows its volume. Contracts suit organisations that need support, security reviews and bundled datasets.

A concrete example: BusinessMCP

BusinessMCP’s IP to company API is a pay-per-call service. According to its product page:

• A single lookup costs $0.02, charged only when a result is found.
• The bulk endpoint takes 1 to 100 addresses per call at $0.02 per item with a result.
• New accounts get $5 of free credit for 7 days, prepaid, with no subscription.

Its comparison of company data APIs adds that prepaid packs start from $10, with no seat or contract. One of its guides describes the trial as covering 250 company answers, which matches $5 at $0.02 each.

There is a wrinkle worth checking. The product page says charges apply only when a result is found, while another guide says other verdicts are free and the FAQ appears to say every address is billed whatever the verdict. We cannot settle that from outside. Run a small batch with the trial credit, compare your balance before and after, and write down what you observe. Surprises in billing wording are a reason for a test, not a reason to avoid a vendor.

How the neighbouring tools charge

IP data vendors come at the problem from different angles, and their prices reflect that. Per the same BusinessMCP comparison page, which is vendor-published and should be verified on each vendor’s own site:

• IPinfo offers a free Lite tier and paid tiers from about $26 per month for its Core product. Its strength is geolocation and network data rather than company resolution; see IPinfo alternatives for company data.
• Hunter.io is credit-based for email finding and verification, a different job from IP resolution.
• Clearbit’s API is now bundled into paid HubSpot plans rather than sold as new standalone keys, according to that page.

If your actual need is company data by domain rather than IP, the same publisher’s company data APIs page lays out the field. Prices in this market move quickly, so treat any table, including ours, as a starting point.

Estimating your own bill

Use a short worksheet.

1. Count unique client IPs per day after removing bots and your own traffic. Call this U.
2. Estimate the share that resolves to a company. Use a conservative figure; BusinessMCP’s guide suggests roughly 20 to 35 percent on B2B traffic and lower for SMB or consumer audiences. Call it R.
3. Multiply: named companies per day is U × R.
4. Multiply by the per-result price for a pay-per-result vendor, or by U for a vendor that bills every lookup.
5. Divide spend by correct answers (see below) to get the effective cost per useful company.

Worked example with round numbers: 2,000 unique addresses a day, a 25 percent resolve rate and $0.02 per result gives 500 named companies and $10 a day if billing is per result. If billing were per lookup on all 2,000 addresses, the same day would cost $40. That fourfold difference is why the billing rule matters more than the sticker price.

Key takeaway: compare vendors on cost per correct named company, not cost per call. Precision and billing rules change the answer more than the headline rate.

The correction for wrong answers

Not every named company is right. A shared range or a stale record can produce a false positive. The method in how to test an IP to company API has you hand-check about 50 named results. If 40 are correct, your precision is 80 percent and your effective cost per correct answer is the unit price divided by 0.8.

Ways to cut cost

• Deduplicate addresses per day and cache answers for a few days, not weeks.
• Skip known infrastructure: filter cloud and bot traffic before the lookup. See filtering bot and cloud traffic.
• Resolve only sessions that viewed high-intent pages if budget is tight.
• Use the bulk endpoint for batch jobs to reduce overhead.
• Test with trial credit before committing to a plan.

Questions to ask before you pay

• Are non-company verdicts billed?
• Is there a minimum spend, expiry on prepaid balance or an overage penalty?
• Which fields come back, and is a confidence value included?
• How often is the data rebuilt?
• Is there an SLA or support channel, and does the price include it?
• Can you export or delete data on exit?

The pricing page of any vendor should answer these; if it does not, ask the vendor in writing.

When free is enough

If you need an occasional look at a handful of addresses, a free web tool is enough. BusinessMCP’s reverse IP lookup tool limits free use to 10 lookups an hour. Paying only makes sense once you are resolving addresses at a volume where manual work costs more than the API.

Reading a price page critically

Price pages are written to be skimmed, and the details that change your bill are usually small print. Scan for these items before comparing numbers.

Detail	Why it changes the bill
Unit of billing (lookup, result, record, credit)	A “credit” may cover different actions at different weights
Expiry of prepaid balance	Unused money can lapse
Free tier limits	May be hourly, monthly or one-time
Overage pricing	Can be far above the plan rate
Bulk versus single pricing	Often identical, sometimes discounted
Support and SLA	Included, extra or unavailable
Data retention and export	Matters for exit costs

A note on honesty in vendor material

Vendor pages, including those we cite, describe their own products favourably. The useful ones also state limits. BusinessMCP’s pages, for example, admit that no match rate is published, that home and mobile lines name no company and that a changed range can lag by a day. That candour is a good sign, but it does not replace testing. Read the vendor’s wording, then verify with your own sample and your own balance. If two pages from the same vendor disagree, as billing statements sometimes do, send a short written question and keep the reply.