Outreach will not tell you what Outreach costs. Its pricing page carries seventeen buttons reading Request pricing and no currency symbol anywhere, read on 11 August 2026.
So you do the sensible thing. You search for a per-seat figure, you find one on a comparison page, you put it in a spreadsheet, and you take that spreadsheet into a meeting. The figure is a resale listing or somebody's 2022 contract, and the model built on it is wrong in a direction you cannot see.
The reason it is wrong is structural, not incidental. Outreach describes its own model as a combination of seat-based pricing and consumption-based pricing powered by AI credits. That is two meters. A single per-seat number cannot express a bill with two meters even if the number were current, and the consumption meter is the one that moves without your headcount moving.
The page is a menu with every dish named and every portion size printed, and the prices left to the waiter.
What Outreach publishes is structure. What it withholds is arithmetic. Those are two separate decisions and most vendors make both in the same direction, which is why the distinction is worth the page.
What follows: exactly what the pricing page does and does not contain, the credit ladder read as ratios, the two cheap tiers that are a fork rather than a rung, what this directory can compute without a price and what it cannot, the redirect that dates every page ranking above this one, and the five questions to get answered in writing before anyone signs.
What does Outreach cost?
Not published, in any currency, at any tier. There is no free tier and no self-serve signup, so the earliest moment a figure exists is after a discovery call. That is the complete honest answer and everything after it is context.
What you can establish before the call is the shape. Four packages exist: Amplify Essentials, Amplify Core, Amplify Plus and Amplify Pro. What separates them is an AI credit allowance running 10,000, 25,000, 50,000, 100,000. The header above them reads Per user pricing. No platform fees.
A seat charge common to all four, a consumption allowance that quadruples across them, and no rate on either. Read 11 August 2026.
What the pricing page publishes, and what it does not
Worth setting out as two columns, because a reader who has only been told they hide their pricing will assume the page is empty. It is not. It is specific about everything except money.
| Published on the pricing page | Not published |
|---|---|
| Four tier names, Essentials through Pro | Any dollar figure at all |
| AI credit allowance per tier, 10k to 100k | What an AI credit costs |
| Per user pricing. No platform fees. | The per-user rate |
| Credit packs sold on top of the allowance | What actions consume a credit |
| Which agents ship on which package | Seat minimums or contract term |
The right-hand column is not a list of oversights. Each row is a thing a buyer needs and each one is absent, which is what makes the pattern legible rather than accidental. Checked 11 August 2026.
What is an AI credit, and what spends one?
The second question has no published answer, and it is the one that decides your bill.
A credit is the unit that separates the four packages. Outreach sells additional credit packs to teams that exhaust the allowance, so the allowance is a floor on consumption rather than a cap on it. Everything else about the meter is unstated: what a Research Agent run costs, whether a failed action still burns credit, whether credits expire at the end of a month or a term.
So the per-user rate you eventually get quoted is not the unit cost of the product. It is the entry fee.
This matters most for exactly the team that looks cheapest on a seat model. Twenty reps who lean hard on the agents will move up tiers on consumption while the headcount sits still, and the seat-based budget approved in January will be wrong by March. Our note on what happens when credits run out covers the three ways vendors handle the wall, and Outreach's is not among the published ones.
The credit ladder, read as ratios
You cannot compute a price from allowances. You can compute the shape of the ladder, and the shape tells you something the four raw numbers do not.
| Tier | AI credits | Step up from the tier below | Ratio to Essentials |
|---|---|---|---|
| Amplify Essentials | 10,000 | n/a | 1x |
| Amplify Core | 25,000 | 2.5x [computed] | 2.5x [computed] |
| Amplify Plus | 50,000 | 2x [computed] | 5x [computed] |
| Amplify Pro | 100,000 | 2x [computed] | 10x [computed] |
Three multiplications, none of them a price. The span end to end is tenfold on consumption, which is a wide ladder for a product whose header advertises per-user pricing. A vendor that intended the seat to be the story would not build a ten-times consumption range on top of it.
The two cheap tiers are a fork, not a rung
The agent list changes between Essentials and Core, and neither is a subset of the other. That is not how a price ladder normally behaves, and it is the finding this site has that the comparison pages do not.
| Amplify Essentials | Amplify Core | |
|---|---|---|
| Sold as | Conversation intelligence and coaching | Pipeline creation and account management |
| AI credits | 10,000 | 25,000 |
| Deal Agent | Yes | No |
| Meeting Prep Agent | Yes | No |
| Revenue Agent | No | Yes |
| Personalization Agent | No | Yes |
Amplify Plus is described on the page as everything in Essentials + Core. So a team that wants deal inspection and pipeline creation cannot buy the cheap version of both, and lands on the third package by requirement rather than by growth. Read 11 August 2026.
Six agents are named across the packages: Omni, Research, Deal, Revenue, Personalization and Meeting Prep. Not one carries a sentence on the pricing page saying what it does, what it may touch, or whether a human approves its output. Omni heads the list on every package and is never explained anywhere on it.
What this directory can compute without a price
Two things, and it is worth being precise about which, because the useful move here is to price the substitute rather than to guess at the original.
First, the category context. Of the ten sales-platform and CRM entities in this directory, three publish no monthly, per-seat or per-minute rate at all: Outreach, Salesloft and Momentum. Four require a sales call. Across all 264 tools here, 114 publish no rate on any of those three fields, 108 have no self-serve checkout, and 96 sit in both sets. That census is here, and a companion page counts 126 on a wider test that also catches vendors printing a rate nobody can buy at.
Second, and more useful, the substitution floor. Outreach's sequencing is the part most buyers under fifty reps actually want. That part is sold separately by twelve dedicated tools in this directory, ten of which publish a monthly rate and none of which requires a sales call.
| Sending tool | Published monthly entry | Self-serve | Free tier | Rate read |
|---|---|---|---|---|
| MailReach | $25 | Yes | No | 2026-08-07 |
| Mailshake | $29 | Yes | No | 2026-08-07 |
| Woodpecker | $35 | Yes | Yes | 2026-08-06 |
| Saleshandy | $36 | Yes | No | 2026-08-07 |
| Smartlead | $39 | Yes | No | 2026-08-11 |
| Instantly | $47 | Yes | Yes | 2026-08-06 |
| Salesforge | $48 | Yes | No | 2026-08-11 |
| QuickMail | $49 | Yes | No | 2026-08-07 |
| Reply.io | $59 | Yes | Yes | 2026-08-07 |
| lemlist | $69 | Yes | Yes | 2026-08-06 |
Median $43 a month, range $25 to $69 [computed from the ten rates above]. The other two sequencing tools in the directory, Klenty and Meetz, publish nothing.
That median is not an estimate of what Outreach charges and must not be used as one. It is the price of the job Outreach is most often bought to do, sold by vendors who will let you start this afternoon, and it is the number that makes the platform's premium a real question rather than a vague one.
Why does outreach.io redirect to outreach.ai?
Because the company moved its apex domain. Verified 5 August 2026 by direct request: HTTP 301, site-wide including the pricing path, with a one-year cache directive attached.
A company does not move its apex domain casually. This one repositioned away from sales engagement, a category it largely defined, onto ground where a credit meter is the expected way to bill. The product strategy and the pricing structure are one decision.
It is also a free dating tool for the pages ranking above this one. Any comparison still linking outreach.io and quoting a flat per-seat figure was written before Amplify existed, and most of them are.
How to price Outreach yourself before the call
The line the platform does not own
Everything above prices the software. This section is the part that runs on your own account, costs about ninety minutes spread over two weeks, and produces a number no quote will contain. Run it before the pilot, because a platform rollout onto filtered domains looks exactly like a platform that does not work.
Google Postmaster Tools, and the three thresholds
Google Postmaster Tools reports a spam rate per sending domain, and the three thresholds are pass under 0.10%, warning between 0.10% and 0.29%, and fail at 0.30% or above, which is the line Google tells senders never to reach. On a 10,000-send month, 0.30% is 30 complaints. Add every domain your reps send from rather than the corporate one alone, since the thresholds are computed per domain and the rollout is what multiplies them.
The seed test, at about a dollar a run
A seed test sends the live campaign copy to addresses across Gmail, Outlook.com, Yahoo and corporate Microsoft 365 and reports which folder each one landed in. This site has priced the standalone tester at 200 credits for $210 a month, which is $1.05 a test, and the pass line is inbox placement above 90% on Gmail and above 80% on Microsoft. A Microsoft-only failure with a clean Gmail result points at authentication or reputation rather than at content, because content would fail both. This is the one number in an Outreach evaluation that costs a dollar rather than a discovery call.
Thirty minutes, run on your own numbers, and it produces a threshold you can hold a quote against rather than a feeling about whether the quote is high.
- Count the reps who will actually be enrolled in month one. Not licensed, enrolled. Multiply by the $43 median above to get the sequencing-only baseline. Ten reps is $430 a month [computed]. Write it down.
- Price what you would otherwise buy separately. Conversation intelligence and forecasting are the platform's real argument. If you already pay for either, that line moves from the Outreach column into the saved column and the comparison changes shape.
- Set your pass threshold before the call. A number, written down, in the form we proceed below $X per rep per month all-in including credits. Deciding this after hearing a quote is deciding it with an anchor already in your head.
- Ask for the seat rate with credits stripped out. The gap between that and the all-in figure is your consumption exposure, and it is the only way to compare against tools priced on seats alone.
- Fail the check if you cannot get the credit consumption schedule in writing. Not a reassurance on a call. A schedule. Without it your annual cost is unmodellable and no amount of negotiating the seat rate fixes that.
Pass means you have a per-rep ceiling and a written consumption schedule. Fail means you have a quote and a promise, which is what most buyers sign.
The five questions to get answered in writing
- What consumes an AI credit, and at what rate per action? This is not on the pricing page and it decides the bill.
- What happens at zero credits mid-term, and what does a top-up pack cost? Negotiate this before signing, not in month four when you need it.
- What is the seat rate excluding the credit allowance? The answer tells you how much of the price is consumption.
- What is the contract term, the notice period and the renewal mechanism? None of the three is published.
- Is there a seat minimum? The page says per user pricing and never says from how many users.
What practitioners say once they are inside it
Vendors publish the rate. Buyers publish the surprise, and the surprises here are about integration rather than price.
On a Hacker News thread from 2 April 2024 (item?id=39908514, 51 points, 26 comments), an engineer building against HubSpot and comparing it to Outreach, Salesloft, Apollo and Salesforce listed the complaints in order, and the third was that none of their APIs support sorting, ordering or filtering. That is one developer's characterisation of a group of products, not a finding of fact about Outreach specifically, and it is worth exactly what a single dated comment is worth. Its use here is narrow: if your implementation plan assumes a clean bulk export or a nightly reconciliation job, ask about pagination and filtering in the technical call rather than discovering the answer in month three.
The failure mode on this platform is adoption, not capability. A fifty-seat contract signed against a five-seat pilot is the regret that recurs, and the structural reason is that there is no free tier and no trial, so the burn rate on both credits and enthusiasm is discovered after commitment. Budget six weeks from first call to first sequence at a mid-sized organisation, longer with a security review.
Outreach against the tools it is compared with
| Outreach | Salesloft | Apollo | lemlist | |
|---|---|---|---|---|
| Published entry price | Not stated publicly | Not stated publicly | $69/seat/mo monthly, $49 annual | $69/seat/mo |
| Rate read on | 2026-08-11 | 2026-08-11 | 2026-08-25 | 2026-08-06 |
| Tier names published | Yes, four | No, none | Yes | Yes |
| Self-serve checkout | No | No | Yes | Yes |
| Free tier | No | No | Yes | Yes |
| Billing model as the vendor states it | Seats plus AI credits | Per seat | Per seat with credits | Per seat |
| Second meter priced | No | n/a, not stated | Credit allowances published per tier | n/a |
| Contract term published | Not stated publicly | Not stated publicly | Not stated publicly | Not stated publicly |
| Conversation intelligence included | Yes | Yes | No | No |
| What you can model before a call | The tier shape only | Nothing | A full annual bill | A full annual bill |
The row that matters is the last one. Against Salesloft the difference is not transparency in any moral sense, since neither publishes a price; it is how much of the shape you can see before entering a sales process, and Outreach lets you see more. Against Apollo and lemlist the difference is not really price either. It is whether a model can be built at all before someone talks to you.
If the chat layer is what you want, note that Drift no longer exists as a standalone purchase and its domain now resolves into Salesloft. We wrote that up separately.
Why there is no observed contract figure here
Because this directory does not have one, and inventing one would cost more than the section is worth.
Twenty-one of the 264 entities here carry a structured contract record. Exactly two of those carry an observed annual contract value from a third-party procurement aggregate: 11x at a $45,000 median and Artisan at $21,000 across a $9,000 to $57,000 range. Outreach carries neither a contract record nor an observed figure.
So this page prints no annual number for Outreach, and any page that does should be asked where it came from. For gated vendors the number reliably surfaces in exactly one place, which is a competitor's comparison page, and that is the one place it must never be taken from.
Who should not book the demo
Under about twenty reps, almost nobody. What a platform sells is governance, forecasting rollup, permissioning and administration. Those answer problems that appear at a certain headcount and are invisible below it, and the sequencing underneath is available from ten published-price tools at a $43 median.
Also anybody who has to model the cost before entering the process. That is not a preference, it is a procurement policy in a lot of organisations, and this vendor's page cannot satisfy it.
Above fifty reps the calculation inverts, because the cost of having no governance starts appearing in the forecast, and at that size the seat rate stops being the interesting variable anyway.
So what does Outreach cost?
Nobody outside the company can tell you, and the pages that claim to are quoting a resale listing, a years-old contract or a rival's marketing.
What you can do is arrive at the call with three things: the $43 sequencing baseline multiplied by your enrolled reps, a written per-rep ceiling you decided before hearing a number, and the five questions above. That converts the demo from a presentation into a price discovery exercise, which is what it always was.