TL;DR
Apollo.io is the most-quoted price in B2B outbound and the least-understood one. Almost every ranking article on the topic tells you Basic is $49 and Professional is $99 and leaves it there. Both of those numbers are real, and putting them in the same sentence is already wrong — $49 is an annual rate and $99 is a monthly one.
The bigger problem is that the tier price stopped being the interesting number. Apollo has migrated to a single unified credit, and in that model your bill is set by what you reveal, not by which box you tick. Two teams on identical Professional seats can differ by 8x in data cost depending on whether they pull mobile numbers. That's the post: the tiers, then the credit arithmetic that actually decides what you pay.
All Apollo prices, credit costs, tier gates and contract terms below were read directly off apollo.io/pricing, docs.apollo.io and dated knowledge.apollo.io articles on 7 September 2026. Figures we derived ourselves (effective cost per credit, the stack model) are labelled as ours. Third-party accuracy numbers are labelled directional, and several widely-circulated ones we deliberately refuse to repeat — there's a section on why.
Apollo.io costs $49, $79 or $119 per seat per month on annual billing (Basic, Professional, Organization), or $65 and $99 monthly for the first two — Organization is annual-only with a 3-seat minimum. There's a free tier and an unpriced Custom tier.
Each seat comes with a credit allowance — 30,000, 48,000 and 72,000 per year respectively — and every action draws on it. All three paid tiers work out to about the same $0.020 per credit, so the tier decision is not a price decision. It's a feature-gate decision, and the gate that matters most is that Basic caps you at one mailbox and 250 emails per day. If you're sending cold email at volume, Professional is your starting point, not Basic.
| Plan | Annual ($/seat/mo) | Monthly ($/seat/mo) | Credits/seat/yr | Min seats | Effective $/credit |
|---|---|---|---|---|---|
| Free | $0 | $0 | 900 (granted monthly, 75/mo) | — | — |
| Basic | $49 ($588/yr) | $65 | 30,000 (upfront) | None | ~$0.0196 |
| Professional | $79 ($948/yr) | $99 | 48,000 (upfront) | None | ~$0.0198 |
| Organization | $119 ($1,428/yr) | Annual only | 72,000 (upfront) | 3 | ~$0.0198 |
| Custom | Sales-only | Sales-only | Negotiated | — | — |
Prices exclude tax. The effective cost-per-credit column is ours — annual price divided by included credits — and it's the single most useful thing on this page, because it shows the tiers are priced almost perfectly linearly. There is no volume discount hiding in the upgrade.
Apollo's pricing page tags every plan with a 24% annual saving. Do the arithmetic on Apollo's own numbers and only Basic clears it: $65 to $49 is 24.6%, but $99 to $79 is 20.2% and $149 to $119 is 20.1%. It's a small thing, but if you're building a business case off the annual discount, use 20% for the tiers you're actually likely to buy.
This is the part that changed, and the part that decides your bill. Apollo used to meter email credits, mobile credits and export credits separately. Now there's one credit, and each action has a conversion rate:
| Action | Credits | Cost at ~$0.020/credit | Max per Professional seat/year |
|---|---|---|---|
| Verified email address | 1 | ~$0.02 | 48,000 |
| Mobile / personal phone number | 8 | ~$0.16 | 6,000 |
| HQ / corporate phone number | 0 | Free | — |
| AI power-up / AI research run | 1 | ~$0.02 | 48,000 |
| Full record enrichment (email + mobile) | up to ~9 | ~$0.18 | ~5,300 |
| Organization search | 1 per page (up to 100 results) | ~$0.02 | — |
| Waterfall email enrichment | 1–4 typical, 20+ possible | $0.02–0.40+ | Varies |
| Waterfall phone enrichment | 8–25 typical, 45+ possible | $0.16–0.90+ | Varies |
Read the mobile row twice. Eight credits per mobile number means Professional's entire annual allowance converts to 6,000 numbers per seat — about 500 a month — with nothing left for email. If your motion is phone-led, Apollo's headline price is not your price and the credit line will dominate your spend. If your motion is email-led, 48,000 emails a seat per year is generous and you will likely never touch the ceiling.
Waterfall enrichment is where budgets actually die. Apollo's own docs warn that some waterfall vendors "consume credits per lookup even when no data is found", and that phone waterfalls can reach 45+ credits — roughly $0.90 for one record that may return nothing. Tellingly, Apollo's own pricing calculator budgets 2 credits per lead when waterfall is enabled, versus 1 without: Apollo itself assumes turning waterfall on doubles your per-lead cost. Turn it on deliberately, per list, not as a global default.
Rarely mentioned anywhere and genuinely expensive: Apollo bills talk time in the same unified credits, at a rate that depends on where you're calling.
| Region dialed | Credits per minute | ~Cost per minute | ~Cost of a 10-min call |
|---|---|---|---|
| US / North America | 2 | ~$0.04 | ~$0.40 |
| Europe | 10 | ~$0.20 | ~$2.00 |
| APAC | 10 | ~$0.20 | ~$2.00 |
| Central / South America | 14 | ~$0.28 | ~$2.80 |
| Middle East | 28 | ~$0.56 | ~$5.60 |
| Africa | 36 | ~$0.72 | ~$7.20 |
Two billing mechanics multiply this. Connected calls carry a 30-second minimum and then round up to the next 30-second increment, so a 35-second call bills as 60 seconds. And calls that hit voicemail or an answering machine still bill — only ring-outs, busy signals, failed and cancelled calls are free. For an SDR team dialing lists with heavy voicemail rates, that is a real and invisible line item. Apollo's own rule of thumb from its knowledge base: roughly 10 minutes of talk time per dollar spent domestically, 3 minutes per dollar with international dialing.
Apollo states this three separate ways across its pricing page, its credits explainer and its knowledge base: unused credits don't roll over, they don't carry into the next billing cycle, and they don't survive renewal.
Pair that with how annual plans are granted — all credits unlocked upfront at the start of the annual cycle — and the failure mode is clear. An annual buyer who runs light in month two cannot bank that capacity for month eleven, and at renewal the counter resets regardless of what's left. Add-on credits bought mid-cycle expire the same way, are charged in full and never prorated, and re-bill next cycle.
This is the sharpest contrast with the alternatives. Clay accumulates unused credits up to 2x your monthly allowance on its self-serve tiers, and Lusha rolls over to a 2x cap on Pro and Premium. Cognism goes further still: once a contact is redeemed, your team keeps using that record for the whole contract, and a credit is only spent again if the person changes jobs. Apollo's hard expiry means your credit allowance is a use-it-or-lose-it rate limit, not a balance. Size your plan to your steady-state monthly volume, not your best month. We covered the mechanics of the Clay side of this in our Clay pricing breakdown.
Given all three paid tiers land within a fraction of a cent of $0.020 per credit, upgrading Apollo is never a data-cost decision. You're buying feature gates. These are the ones that matter for an outbound team:
| Gate | Free | Basic $49 | Professional $79 | Organization $119 |
|---|---|---|---|---|
| Daily email sends | 250 | 250 | Unlimited | Unlimited |
| Mailboxes per user | 1 | 1 | Unlimited + 5 SMTP | Unlimited + 15 SMTP |
| SendGrid & Mailgun | ✗ | ✗ | ✓ | ✓ |
| Sequences | 2 per team | Unlimited | Unlimited + A/Z testing | Unlimited + A/Z testing |
| Record selection limit | 25 | 1,000 | 2,500 | 10,000 |
| Buying intent topics | 1 | 6 + filters | 6 + filters | 12 + filters |
| Waterfall + job-change enrichment | ✗ | ✓ | ✓ | ✓ |
| Agentic Loops | ✗ | ✗ | ✓ | ✓ |
| Automated workflows | 2/team | 5/team | 50/team | 500/team |
| Call recording + AI insight minutes | 150 | 150 | 4,000 | 8,000 |
| SSO, custom permissions, territories | ✗ | ✗ | ✗ | ✓ |
The two bolded rows are the whole story. One mailbox per user and 250 sends per day is not a cold email configuration — it's a prospecting seat with a sequencer bolted on. Serious outbound runs many mailboxes across several domains precisely so no single inbox carries enough volume to burn, which is the architecture we walk through in how many domains and mailboxes you need. You cannot build that on one mailbox.
So if a comparison article tells you Apollo starts at $49 for cold email, it's quoting a plan that structurally can't do the job. Professional at $79 is the real entry price for outbound, and the gap from $49 to $79 buys unlimited mailboxes, unlimited sends, 5 SMTP connections and SendGrid/Mailgun support — not more credits per dollar.
Underrated gate. Basic lets you select 1,000 records in one action; Professional 2,500; Organization 10,000. Apollo spells out the consequence itself: at a 25-record limit, saving 10,000 leads means clicking select-and-add 2,000 times. If someone on your team is doing large list pulls weekly, this row is an hours-per-month decision.
Nothing about Organization makes data cheaper. What it adds is SSO, custom permission profiles, billing and seat-manager roles, custom territories, 12 intent topics and the option to bring your own LLM API key — plus a 3-seat annual-only minimum ($4,284/year committed at list price). That's an IT-and-governance purchase. If nobody is asking you for SSO or territory rules, Professional is the ceiling worth paying for.
Apollo includes a dialer from Basic upward, and that fact gets quoted as if Apollo replaces a sales phone system. It doesn't, at the tier price. The included dialer is US-only, single-line. Everything that makes a dialer worth having for an outbound team — power dialer, parallel dialer, international dialing and local presence — sits in an Advanced Dialer add-on at $119/month per team on annual billing ($149 monthly). There's a second add-on, Inbound, at the same price, which lifts website-visitor identification from 100 to 50,000 companies a month and form enrichment from 10 to 5,000.
Both are per-team rather than per-seat, which is unusually favourable at scale — and both are labelled introductory pricing by Apollo, so treat them as subject to change. Credit consumption still applies on top of the add-on fee. This is the same pattern we found comparing CRMs: the dialer is almost never in the price you're quoted, and it moves the total more than the seat price does. See Close CRM vs HubSpot for the CRM-side version of that trap.
Apollo's add-on credit rate card runs from $0.025 per credit at small volumes down to $0.015 at 250,000 credits, with $0.020 around the 2,500–5,000 mark and $0.016 from 20,000 up.
Notice what that means: top-ups cost roughly the same as in-plan credits (~$0.0198). There is no punitive overage multiplier — which is genuinely good design, and worth saying plainly because competitor pricing posts routinely claim Apollo charges $0.20 per overage credit. That's off by a factor of ten against Apollo's own live rate card. Nor will Apollo cut you off mid-cycle; it notifies you instead.
The penalty isn't the rate, it's the expiry. An extra 1,000 mobile numbers means 8,000 credits, roughly $128–144 — and whatever you don't spend by cycle end is gone. Buy top-ups reactively against a known list, never speculatively.
Here's the model that matters. Three SDR seats running email plus phone, at list price, annual billing:
| Line item | Why you need it | Annual cost |
|---|---|---|
| Apollo Professional × 3 seats | Basic can't send at volume | $2,844 |
| Advanced Dialer add-on | Power/parallel dialing, local presence | $1,428 |
| Dedicated sending tool | Inbox rotation and warmup Apollo doesn't do well | ~$940 |
| Third-party email verification | Re-verify before sending (see below) | $39–1,188 |
| Total | — | ~$5,250–6,400 |
This model is ours, not a quote, and it's list price with no negotiation. The sending-tool line uses a mid-tier plan from a dedicated sender; the verification range spans a low-volume bulk verifier to a bundled monthly plan. The point isn't the exact total — it's that the seat price is under half of it. Anyone budgeting Apollo at 3 × $79 is budgeting about 45% of the real number. If you want the full bottoms-up version of this exercise for an email programme, we built one in cold email cost per meeting.
Apollo's data quality is the most contested topic in this category, and almost every number you'll read comes from a company selling an alternative. We don't need those. Apollo's own knowledge base is enough:
Those four statements can't all be operationally true at once. Apollo's stated accuracy implies a bounce rate roughly four times its own published danger threshold. Apollo also maintains an official ZeroBounce integration whose stated purpose is removing invalid leads from your Apollo account — which is Apollo conceding the re-verification step exists.
None of this makes Apollo bad data; at $0.02 a verified email it's priced well below most single-purpose vendors. It means you must budget a verification pass between Apollo and your sender, and the plans start around $39 for bulk credits. Our email verification tool comparison covers the options, and SPF, DKIM and DMARC setup covers the authentication side that bounces will otherwise punish you for.
Numbers we refuse to repeat — including one of our own. You'll see confident claims that Apollo's accuracy is "~65%", that exported lists bounce at "15–25%" or "32–38%", and a widely-copied 900-lead test reporting "19% valid, 60% catch-all". We tried to trace these to a primary study and couldn't — no disclosed methodology, no sample size, and in every case a publisher selling a competing product. Our own cold email list building guide currently cites a "~65–70% real-world" figure against a "~97% vendor claim" from that same circulating pool — we're flagging both here rather than quietly leaving them, since Apollo's own published accuracy claim is 91%, not 97%. The best-documented third-party datapoint we found was a competitor's disclosed 100-contact test in which 27 Apollo-verified contacts were flagged by a second verifier and 6 of the remainder still hard-bounced — one mailbox, one author, n=100. Directional at best. Apollo's own 9% figure is both more citable and more damaging.
Apollo's self-serve billing is more restrictive than the checkout flow suggests. All of the following is published:
Agencies and anyone building on top of Apollo: read the licence first. Apollo states that its self-serve plans are permitted for internal business use only, and that using them to power external products, share data with customers, or resell Apollo data is not allowed under standard terms — those cases "require a separate agreement with custom pricing and terms". If you're a lead-gen agency planning to run client campaigns off your own Apollo seat, that is a conversation with Apollo's sales team, not a checkout decision. It's also a reason plenty of agencies price data through a different layer entirely; we cover how that shows up in fees in cold email agency pricing.
Apollo is mid-migration between credit systems, and it says so: some features are only available on the new unified system, all new customers are on it, and existing customers may still be on the legacy one. The site hasn't caught up, which produces contradictions you should resolve in your own portal rather than trust:
One honest caveat before the table: we're not publishing headline prices for Clay, ZoomInfo or Cognism. Clay's own pricing page currently contradicts itself between its plan cards and its FAQ, and ZoomInfo and Cognism don't publish list prices at all. Anyone quoting you a ZoomInfo price is quoting a third-party estimate.
| Vendor | Priced by | Do credits roll over? | List price published? |
|---|---|---|---|
| Apollo | Per seat + unified credits | No — hard expiry each cycle | Yes |
| Clay | Per workspace + actions/data credits (unlimited seats) | Yes, to 2x monthly on self-serve tiers | Yes, but self-contradictory |
| Lusha | Per account + credits (seats bundled) | Yes, to 2x monthly on Pro/Premium | Yes |
| ZoomInfo | Negotiated contract | Contract-specific | No |
| Cognism | Negotiated package | Redeemed contacts stay usable for the contract term | No |
Structurally: Apollo is the cheapest published per-record price in this group and the least forgiving about unused capacity. If your volume is steady, that trade is good. If it's lumpy — agency work, campaign bursts, seasonal pushes — a rollover model is worth paying more per record for. For the feature-level version of this comparison, see Clay vs Apollo vs ZoomInfo.
We build and run outbound systems on this stack, so here's the plain recommendation by motion:
And whichever you pick, put the credit conversion rates somewhere your team can see them. The single most expensive habit we see is reps revealing mobile numbers by reflex on lists that were only ever going to be emailed — an 8x cost multiplier applied to records that never needed it.
Apollo.io costs $49, $79 or $119 per seat per month on annual billing for Basic, Professional and Organization respectively, or $65 and $99 per seat monthly for Basic and Professional — Organization is annual-only with a 3-seat minimum. There's a free tier with 900 credits per seat per year and an unpriced Custom tier sold through sales. Every paid tier works out to roughly $0.020 per credit, so the tier you choose changes which features you can use, not how cheap your data is.
Not any more. All four current plans ship a finite credit allowance — 900, 30,000, 48,000 and 72,000 per seat per year — and one verified email costs one credit. Apollo's site still carries legacy "unlimited plan" fair-use language, but that describes the old credit system Apollo is migrating customers off; all new customers are on the unified credit model. Separately, note that Basic caps you at 250 sends per day regardless of credits.
There is no separate mobile credit any more. Apollo unified its currencies, and a mobile or personal phone number costs 8 credits — about $0.16 at Professional's annual rate. Your full 48,000-credit annual allowance therefore buys at most 6,000 numbers per seat with nothing left for email. Corporate and HQ numbers are free, but personal numbers are charged even when the contact is on a do-not-call registry.
No. Apollo states it three separate ways: unused credits don't roll over, they don't carry into the following billing cycle or upon renewal, and per its terms they expire at cycle end. Annual plans receive all credits upfront, which means an annual buyer who under-uses early cannot bank that capacity for later in the term. Add-on credits behave the same way and are charged in full without proration.
Yes, but not instantly and not always by yourself. Cancellation takes effect at the end of your current term rather than on request, after which the account reverts to the free plan and keeps only one seat after 10 days. Apollo states that only "select paid teams" have self-serve cancellation — other accounts, and any account opened through Apollo sales, must go through support or an account manager. Mid-term downgrades are applied immediately but never refunded.
Professional, at minimum. Basic gives you one mailbox per user and 250 sends per day, which is a data seat with a sequencer attached rather than an outbound engine. Professional lifts both to unlimited, adds 5 SMTP mailboxes per user and SendGrid/Mailgun support, and unlocks Agentic Loops. Bear in mind that most teams still pair Apollo's data with a dedicated sending tool for inbox rotation and warmup, and re-verify addresses before sending.
Pricing a full outbound stack, not just a data seat? Read cold email cost per meeting for the bottoms-up model, or talk to our team — we build and run this stack for B2B teams every week, and we'll tell you honestly which tier you actually need.
By the GenFlows GTM engineering team. All Apollo plan prices, credit conversion rates, tier gates and contract terms were read directly off apollo.io/pricing, docs.apollo.io and dated knowledge.apollo.io articles on 7 September 2026, and Apollo notes it is still migrating some customers between credit systems — confirm figures in your own portal before committing. Effective cost-per-credit figures and the 3-seat stack model are our own arithmetic on those published inputs, at list price with no negotiation, and are not quotes. Third-party data-accuracy claims are identified as directional in-text where cited and refused where untraceable. Last updated September 2026.