Best outbound sales agencies

The ten at a glance

Read the middle column first. These ten aren't selling the same thing, and sorting by price before sorting by model is how buyers end up with a phone team when they wanted email.

Everything in that table was read on the company's own site, its Terms page or its Clutch profile on 9 November 2026. "Not published" always means somebody went looking and there was nothing there.

"Outbound sales agency" describes three different businesses

This is the reason the category is so hard to shop. Ten firms answer to the same two words and three of them are selling you something structurally different from the other seven. Sort that first and the shortlist writes itself.

Understory puts the distinction more bluntly than anyone else on this page. Their own FAQ contrasts their model with "a traditional SDR agency" that "rents you reps." That's a fair description of four companies here, and none of the four would phrase it that way themselves.

The channel each one actually runs

Worth checking before anything else, because several of these firms don't run the channel buyers assume they do.

• memoryBlue is a phone company. Their own site says "85% of the meetings we book come directly from the phone" and explicitly contrasts that with "not email or LinkedIn"
• SalesHive is phone plus email, with no LinkedIn at all. Their entry tier is phone only, with no email included
• SalesCaptain runs email, LinkedIn and text, and never offers cold calling
• OutreachBloom sells LinkedIn separately at $1,000 a month, sent from your own profile, with you answering the replies. That's a different product from a managed inbox
• Grow Surely runs cold email only, deliberately, and doesn't sell the other three

If you came looking for cold calling and hired an email shop, no amount of optimisation fixes that. It's the single most common mismatch in this category.

Six of these ten won't tell you the price

Four publish a number you can read before you talk to a salesperson. In this category that counts as unusual.

The four-week month, which isn't a month

SalesRoads is the most transparent pricing page in this group and it contains one detail worth slowing down for. The price is per 4-week period, not per calendar month.

There are 13 four-week periods in a year, not 12. So one SDR at $11,950 is roughly $155,350 a year, not $143,400. That's a real $11,950 gap, and it is the kind of thing that turns up in the thirteenth invoice rather than the proposal.

Nothing dishonest is happening. It's stated on the page. But every comparison article that converts it to "about $143,000 a year" has quietly dropped a month, and buyers reading those articles are budgeting for twelve.

The only one that publishes its contract is the one whose contract disagrees with it

There is an irony at the centre of this page and it is worth stating plainly.

SalesHive is the single agency here that publishes a real client agreement rather than website boilerplate. That deserves credit. It also means SalesHive is the only firm on this page whose marketing claims can be checked against its own contract, and when you check them, they don't match.

There is a second trap in how it is published. SalesHive has two terms URLs. The one called "terms of service" sets no commercial terms at all and points elsewhere; the actual Client Agreement lives at a different address. A buyer who checks the obvious one concludes nothing is published and stops looking.

One more line in that agreement matters more than any of the above: the Proposal Agreement controls over the Client Agreement wherever the two conflict, and the Proposal isn't public. So even the published contract isn't the final word on your own deal.

Nine of the ten publish no service contract at all, Grow Surely included

Having been hard on SalesHive for publishing, here is the thing that puts it in perspective.

• Martal Group, Understory and memoryBlue have no terms page whatsoever. Martal's footer link labelled "Privacy & Terms" goes to a privacy policy containing no contract language, and the terms URL returns a 404
• SalesRoads' "Terms of Service" sits inside its privacy policy and governs use of the website. There is no term in it, no notice period and no refund clause. Its "cancel anytime" promise has nothing behind it either way
• OutreachBloom's terms were last modified in June 2023, govern the website rather than the service, and the only refund language in them is "All sales are final" - sitting underneath a marketing page promising "month-to-month, no long-term contract"
• Grow Surely's own terms page covers website use and text messaging. It isn't a services agreement either. The price and the absence of a minimum term are published; the contract isn't

So the honest instruction isn't "read their terms page", because for most of this category there is nothing on it. It's ask for the actual agreement before the proposal, then search it for "initial term", "auto", "notice", "refund" and "cap". Four minutes, and it is the highest-value four minutes in the whole process.

The reply rate every one of these agencies quotes was measured against the wrong denominator

Sit through enough pitches and the same number arrives every time: a 5 to 8% cold email reply rate, offered as the industry benchmark. It's worth knowing where that figure comes from, because it traces almost entirely to one agency, and that agency has since stopped using it.

Belkins publishes the largest cold email dataset in this category. Here is what their own pages say, in order:

Three different averages for the same period, and two different sample sizes for the same year. Then the latest figure lands 13 times lower than the one everybody quotes.

What changed was the denominator, and Belkins say so themselves

The earlier studies divided replies by unique openers. The 2026 study divides replies by emails sent, because open tracking was turned off. Belkins state on the page that the figures "are not directly comparable to prior reports."

So the famous 5 to 8% was never a reply rate per email sent. It was replies divided by people who appeared to open, on a denominator that Apple's Mail Privacy Protection and image blocking had already broken. Measured properly, the same agency's same clients reply at 0.45%, and at 0.51% in the US.

Independent support for that order of magnitude: Gong's analysis of 28 million cold emails puts the average at 344 emails per meeting booked, which is roughly 0.3% of sends producing a meeting. Nowhere near 5.8%.

What this means on a sales call

Grow Surely publishes its own figures across 6,074,430 emails: 1.6% human replies per prospect contacted and 0.8% per email sent, with auto-replies separated out rather than counted. On the same per-send basis, Belkins measures 0.45%.

Those two numbers are comparable. The 5.8% isn't comparable to either, because it is answering a different question.

So ask one question and the whole conversation changes: what is the denominator, and does it count auto-replies? In Grow Surely's data auto-replies are 51% of everything that lands in the reply folder. An agency quoting a reply rate that includes them is quoting roughly double the real number, before you even get to the opener problem.

36% of B2B software companies cut their SDR team last year

Relevant here because four of these ten agencies sell you SDR headcount, and the market for SDR headcount is contracting.

Emergence Capital surveyed 560+ venture-backed B2B software companies in April 2025:

SDR is the only role on that list with both the highest cut rate and the lowest growth rate. Every other function is closer to balanced.

A caution that matters, because almost every article citing this gets it wrong: the survey shows headcount falling. It doesn't establish why. The widespread attribution to AI is commentary rather than a finding, and the survey didn't ask the question.

The part that is genuinely measured

Two numbers from datasets with published methods are more useful than the speculation:

• Outbound is mid-pack, not dead. Ebsta and Pavilion measured 655,000 opportunities worth $48 billion across 387 companies, scoring every channel on the same formula of win rate times deal size divided by cycle length. Partner referral scored 1.3, organic inbound 1.2, outbound 1.05, events 0.78 and paid 0.68. Outbound beats both of the channels most companies spend more on
• The race is usually over before a buyer makes contact. 6sense studied 3,596 completed B2B purchases of $25,000 or more. Buyers first contact a seller 59.2% of the way through the cycle, the eventual winner was on the Day 1 shortlist 91.7% of the time, and the vendor the buyer already favoured before any contact won 81.5% of the time

Put those together and the argument for outbound isn't that it converts better than inbound, because it doesn't. It's that the shortlist forms months before anyone fills in a form, and outbound is the only channel that reaches a buyer while that is still happening.

Which is an argument an outbound agency is obviously happy to make. The difference is that this version comes with sample sizes.

Where the reps actually sit, and who won't say

CIENCE's published rate card settles an argument the rest of the category prefers to leave open. Offshore SDRs start at $1,750 a month. US and Europe SDRs start at $3,500. Same role, same platform, roughly double the price for geography.

Which makes "where do your reps sit" a pricing question, not a preference question. Here is who answers it.

Martal's is the one worth looking at twice, because it isn't a refusal to answer. It's an answer that its own next paragraph contradicts. A team that is 20% outside North America, 10% of it in LATAM, isn't what most buyers understand by "onshore."

None of this makes offshore delivery bad. Plenty of excellent outbound runs from Manila, Guadalajara and Krakow. It makes an unstated location a pricing problem: if the quote doesn't say, you can't tell whether you're paying the $1,750 rate or the $3,500 one.

Named clients or named numbers, almost never both

Every agency on this page publishes proof. Read a few dozen of those case studies in a row and a pattern appears that is more useful than any individual number: most firms will give you the client's name or a hard figure, and quietly withhold the other.

Martal's is the one to understand, because it is invisible at a glance

Martal publishes 20 named clients. Martal publishes large, impressive numbers. They're never the same clients.

• The named studies: 20 clients, no hard numbers attached to any of them
• The big numbers: industry dashboards labelled "27 months" or "over 3 years", with no company attached

Open the DeepHow study, which is named: 20,000 prospects a month, 15 qualified leads a month, and no campaign duration at all. Then look at where the big figures live - industry dashboards labelled "27 months" or "over 3 years" with no company attached to any of them.

One of those dashboards reports 1,219 sales-qualified leads against 1,048 marketing-qualified ones. More SQLs than MQLs is backwards, since the first is meant to be a subset of the second. That isn't fraud, it is a sign the dashboards are assembled loosely, and it is a reason to treat them as illustration rather than evidence.

SalesCaptain's is the one most likely to mislead a skim-reader

Their headline metric is positive replies, and 26 of 27 case studies are named, which is genuinely good practice. But a reader moving quickly reads "521" and thinks meetings.

A positive reply is someone writing back without saying no. A meeting is a calendar invite accepted. The gap between the two is the entire middle of the funnel, and only 5 of their 27 studies cross it.

And one of their own numbers disagrees with itself

Three traps, then, and on the page they look nothing alike:

1. Named client, no number. You can check who, never how much
2. Big number, no client. You can check how much, never who
3. Real number, wrong unit. Positive replies read like meetings to anyone skimming

SalesRoads reports its Bid Retriever result as 1,614 meetings on one page and 1,164 on another. Both are their own pages. The digits are transposed somewhere and there is no way to tell which is right from outside, which is the practical lesson: a figure appearing once hasn't been checked by anybody, including the people who published it.

The four standards every agency here is judged on

Same four questions asked of all ten, so the comparison means something.

1. Is the price published? Written down, where a buyer can read it before talking to anyone
2. Is there a service agreement published, and does it match the marketing? Two separate questions. Nine of these ten fail the first one
3. Is the proof both named and measured? A named client with a number and a date range. Most firms give you two of those three
4. Does independent review evidence exist? Not self-published testimonials. A profile somebody else controls

Nobody here scores four out of four. Grow Surely doesn't either, and the scorecard further down says exactly where it falls short.

The review record, and the two companies with no record at all

Read these as sample sizes first and averages second. A 5.0 from three reviews tells you less than a 4.2 from a hundred and forty.

SalesHive's Trustpilot is a barbell, not an average

3.5 out of 5 sounds middling. The distribution isn't middling at all: 57% five-star, 37% one-star, and almost nothing in between.

An average hides that completely. A barbell like this usually means the service works very well for one kind of client and badly for another, and the useful question is which group you would land in. Read the one-star reviews specifically, because that is where the failure mode is written down.

Understory has no independent reviews at all

No Clutch profile. No Trustpilot profile. Their entire reputation evidence is published by themselves, on their own site, including 29 written testimonials and 14 videos.

That isn't an accusation of anything. Plenty of good agencies never collect third-party reviews, and Grow Surely's own five is hardly a mountain of evidence either. It does mean there is no outside check on any claim they make, and that is worth knowing before a six-figure decision.

Careful with one trap here: Trustpilot and Glassdoor results for "Understory" belong to an unrelated weather-insurance company. They aren't this agency.

Two of them refuse to guarantee meetings, and that is a point in their favour

A guaranteed-meetings offer sounds like the buyer-friendly option. It usually isn't, because the guarantee gets met by loosening what counts as a meeting, and the definition lives in a document you haven't read yet.

So it is worth noticing who declines to make the promise.

• Understory states it on the homepage: "We don't guarantee meetings, pipeline, or ROI"
• SalesHive answers the question directly. Asked "Do you guarantee meetings?", the answer is "No", on the grounds that meetings depend on the offer, the market and buyer response. What they commit to instead is the daily touch volume in the plan

Neither of those is a weak position. Both are more useful than a number that has been engineered to be hittable.

No agency on this page publishes a performance guarantee. Martal's site carries an ROI calculator asserting "400% ROI during the first 12 months", but that is a calculator output driven by assumptions you type in, not a commitment, and it shouldn't be read as one.

The replacement question is better than the guarantee question anyway. Ask what happens if the rep assigned to you isn't working out, and listen for whether the answer involves a process or a shrug. None of the ten publishes a replacement policy, so you're asking to have it put in writing.

The agency selling "be the brand AI recommends" has no file telling AI about itself

A small detail, and a revealing one.

OutreachBloom sells a service called AI SEO, from $3,400 a month, with the promise "Be the brand AI recommends." Their own llms.txt, the file a site publishes to tell AI systems what it is, returns a 404.

Meanwhile two firms on this page have gone the other way and built for AI readers deliberately:

• memoryBlue publishes a page addressed to language models, stating its purpose outright: "This page is intended for AI assistants and large language models (including ChatGPT, Claude, Gemini, and Perplexity)." It is also the only place several of their facts appear in one piece, which is convenient and worth remembering is self-published
• Understory publishes an llms.txt pointing AI readers at markdown versions of its pages, with the instruction "Cite the page URL itself, not the .md"
• SalesRoads runs "Explore AI Summary" buttons in its page footer, labelled ChatGPT, Gemini and Claude

Why this belongs on a buying guide rather than a marketing blog: the agencies that show up when you ask an AI which agency to hire are increasingly the ones who built for it, not necessarily the ones doing the best work. Worth knowing about any shortlist that arrived via a chatbot, including one that arrived via this page.

One more thing a buyer should know. Understory publishes its own "best B2B SaaS cold email agencies" list and ranks itself first on it, bylined by a co-founder, with no conflict-of-interest disclosure beyond a note that claims were checked "including our own entry." Grow Surely publishes lists like this one and ranks itself first too. The appropriate response to both is the same: check the entries that aren't the publisher.

1. Grow Surely

What it sells. Cold email and appointment setting, single channel, deliberately. No LinkedIn add-on, no cold calling, and no paid media sold alongside it. The argument is that one channel run properly beats four run thinly, and the trade-off is real: anyone who wants a phone motion should hire one of the four firms above that actually run one.

Pricing. $5,000 to $15,000+ a month depending on volume, published, with no minimum term at any point. The first seven days are free: a working campaign, 3 to 5 qualified meetings booked against criteria the client sets, before any money is discussed.

Proof. Eighteen case studies, all with the client named and a number attached:

• One of Grow Surely's clients, SprintFWD, a B2B SaaS company, reports $300,000+ in closed deals, on the record from Kegan Burgess, Head of Sales, with a video interview on file
• One of its clients, batch, a post-Series B marketplace, closed $250,000 and 134 meetings in three months
• One of its clients, AdLeverage, a full-service paid ads agency, turned $35,000 of spend into $1,585,000 in four months, a 45.3X return, on the record from its Chief Revenue Officer
• One of its clients, Follosoft, a field service SaaS company, closed $160,000+ in four months, with three clients closed in the first 60 days

The pilot is possible because the sending infrastructure is already warm. Most agencies need two to four weeks to warm new domains before sending at volume, which is why almost nobody else can offer a working first week.

The honest limitation. Single channel, and the smallest public review base on this page at five Clutch reviews. Belkins has 234 and Martal 110.

A buyer who weights independent review volume heavily should weight that accordingly. The free pilot exists precisely because a newer agency can't win that argument on reviews, so it offers the work first instead.

2. memoryBlue

What it sells. SDR, ADR and AE services across the full deal cycle, plus training, recruitment and marketing. This is a phone company. Their own site states that 85% of the meetings they book come directly from the phone, and contrasts that explicitly with "not email or LinkedIn."

The model is rented headcount: 650+ dedicated reps, with an audition-to-hire option that lets a client convert a rep onto their own payroll, and direct hiring from an alumni network of 1,200+.

Pricing. Not published. No rate card exists on the site. Competing agencies circulate figures in the region of $7,000 to $11,500 per dedicated SDR per month, but every one of those estimates comes from a direct competitor, so treat them as marketing rather than data.

Proof. The deepest named case study library here. Trend Micro: 9 deals in active forecast within the first three months, 16 longer-term opportunities, $2.25 million in pipeline. Accolite $30M+ pipeline, Couchbase 1,307 qualified meetings, Motorola Solutions 988 sales-qualified leads over 20 months.

Read the timeframes carefully. On the Trend Micro study the "within first 3 months" attaches to the nine deals, not to the $2.25 million, which carries no window at all.

The corporate history most lists get wrong. memoryBlue acquired Operatix in July 2023, not the other way around, and the combined business kept the memoryBlue name with Operatix's Aurelien Mottier as President. A private equity firm, Avesi Partners, made an investment in memoryBlue in June 2022 - terms were never disclosed, so anyone writing "acquired by" or "majority-owned" is guessing. The old Operatix Clutch URL now lands on the memoryBlue profile.

The honest limitation. Twenty-three Clutch reviews is a modest independent sample for a 650-person firm, and the 4.6 across 300+ G2 reviews they cite is self-reported on a page they wrote. No terms page exists anywhere on the site, and their own materials never say who manages the reps day to day.

3. CIENCE

What it sells. Outsourced SDRs as people, offshore or onshore, running on its own platform, with an in-house research and data operation behind them.

Pricing. Fully published, and the most transparent rate card in this category.

• Offshore SDRs, from $1,750 per SDR per month
• US and Europe SDRs, from $3,500 per SDR per month
• Campaign team, $2,000 a month flat, however many SDRs
• Platform, $499 a month
• One setup fee, amount not published

Stated terms: "Month to month, no long contract." A worked example of five offshore SDRs lands at $11,249 a month.

This changed in 2026 and most comparison pages haven't caught up. CIENCE previously ran a twelve-month initial term and published no price list. Any article still describing that is out of date, and an earlier version of Grow Surely's own CIENCE page was wrong about it too.

Proof. Named case studies with hard numbers: SendBird at 227 appointments in four months, Stage 1 Financial at 190+ appointments with CEOs and founders, Mandolin at 110 qualified meetings in six months.

The honest limitation. 4.2 across 142 Clutch reviews is the lowest average on this page, and it comes with the second-largest sample, which makes it harder to dismiss than a poor score on twelve reviews. The setup fee is the only unpriced item on an otherwise complete page. There is an argument that publishing everything attracts more honest reviews rather than better ones, and CIENCE is the evidence for it.

4. SalesRoads

What it sells. Dedicated SDR headcount, phone-led with email support. No LinkedIn outreach at any tier. Priced per rep, with volume discounts, which makes it the easiest firm here to budget.

Pricing. Published in full, which almost nobody else here manages: $11,950 for one SDR per four-week period, falling to $7,192 each at six. Each seat includes 500 prospect records a month, a dashboard and CRM integration. There is no setup fee and no per-appointment option, and no trial either.

Remember the four-week billing period. Thirteen periods a year, not twelve, so a single SDR is about $155,350 annually.

Proof. Thirty-three named studies, almost none anonymised: AchieveIt $27M total pipeline, Black River Group $28,000,000 pipeline, Optimas Solutions $13,900,000, Parker Hannifin $2.2m revenue. The strongest single entry is Bid Retriever, taken from pre-revenue to acquired in 36 months, which is a rare thing to be able to point at.

Almost none of them carry a date range. Layfield is the exception at "$1.25mm pipeline generated in 15 weeks", and it is the most useful study on the site for exactly that reason. Pipeline with no window attached could have taken one quarter or five years.

The honest limitation. Their Bid Retriever meeting count appears as 1,614 on one page and 1,164 on another, both theirs. "Cancel anytime" appears on five marketing pages and is backed by no published agreement - the Terms of Service is website-use language inside the privacy policy. And they never state where their reps sit, while contrasting themselves with "poorly trained offshore reps."

5. Belkins

What it sells. Appointment setting and outbound as a managed programme, multichannel, at the largest scale of any firm on this page by client count.

Pricing. Not published. A pay-per-appointment option is reported at $300 to $800 per meeting, which is a third-party figure rather than one Belkins publishes.

Proof. The review record is the argument. 4.9 across 234 Clutch reviews is the largest independent sample any agency in this category holds, and consistency at that volume is genuinely hard to fake.

The honest limitation. No published price and no published term, so the discovery call is the only way to find out either. At their client volume, which account manager a buyer is assigned varies more than at a boutique, and that is the most common theme in the reviews that aren't five stars.

6. Martal Group

What it sells. Outbound lead generation, appointment setting, sales outsourcing, LinkedIn, cold email and cold calling, offered as "fractional", "full-time" or "enterprise" sales teams. Multichannel is explicit: they go live across email, LinkedIn and phone within two weeks of onboarding.

Pricing. Not published. The pricing page lists structures with no figures: Tier 1A is a flat monthly fee, Tiers 2 and 3 are a flat fee plus sales commission, and the commission rate is never stated. Every tier reads "inquire about pricing."

One trap for anyone reading their site quickly: the figures "$30 to $500 per SaaS lead" and "$4,000 to $12,000 per month" appear on their SaaS page as market commentary, not as Martal's prices.

Proof, and the structural problem with it. Martal names 20 clients and publishes large numbers, and they're never the same clients. The named DeepHow study gives 20,000 prospects and 15 qualified leads a month with no duration at all. The big figures live in industry dashboards with no company attached, one of which reports more sales-qualified leads than marketing-qualified ones, which is backwards.

Reviews. 4.8 across 110 on Clutch and 4.6 across 135 on G2 is a large, consistent record and the strongest thing on their page. Trustpilot sits at 2.9, but across three reviews, which is too small to carry weight in either direction.

The honest limitation. They describe 200+ executives as "onshore" and then publish a split of US 60%, Canada 20%, EU 10%, LATAM 10%. Their own numbers put a fifth of the team outside North America. And no terms page exists - the footer link labelled "Privacy & Terms" goes to a privacy policy with no contract language, so the 3 and 4-month pilots are unbacked by anything published.

7. SalesHive

What it sells. Cold calling and managed email, US-based or offshore, sold as dedicated SDR headcount. No LinkedIn at all. The entry tier is phone only with no email included, which is unusual and worth checking against what a buyer actually wants.

Pricing. Not published - three tiers, all custom quoted. They do publish two things that matter: no setup fees, and "you owe $0 until you approve everything for launch."

They're the only agency here that publishes a real client agreement, and it is the reason this page can show a marketing-versus-contract gap at all. The marketing says month to month. The agreement's stated default is a 12-month initial term with remaining fees accelerated on early exit, plus a 6-month option the marketing never mentions.

Proof. Entirely anonymised by industry: a workplace injury management provider at 1,440 meetings over four years, a commercial insurance agency at 596 meetings, a fleet maintenance software company at 564. To their credit they qualify it themselves, stating plainly that these are recorded bookings rather than attendance or revenue.

They say where the reps sit, plainly: "our core SDR team is 100% US-based", with Philippines SDRs sold as a separate, labelled option. That's the clearest disclosure on this page and most of the category won't match it.

The honest limitation. The review record is the weakest here: 0.5 across a single Clutch review, and Trustpilot at 3.5 across 35 that is really a barbell - 57% five-star against 37% one-star, with almost nothing between. Read the one-star reviews before signing anything.

8. Understory Agency

What it sells. An "allbound" bundle: GTM engineering, paid media, LinkedIn content and RevOps. Outbound is one component rather than the product, built on buying signals in Clay and synced to the client's CRM.

They position against this entire category explicitly. Their FAQ contrasts their model with "a traditional SDR agency" that "rents you reps", and they don't sell reps at all.

Pricing. Not published. "Custom flat retainers for each service, never a percentage of spend", which is a meaningful commitment for an agency that runs paid media, just not a number.

Proof. Fourteen named studies, strongest on outbound being FullPAC at roughly 365 positive replies, 75 meetings and nearly $10M in pipeline on under $60K of spend in three months. PatientNow: 242,750 emails at a 0.8% lifetime bounce rate with 270 positive replies over about six months, which is a genuinely good deliverability number to publish.

Most of their proof is paid media rather than outbound, which matters if outbound is what a buyer is shopping for.

They refuse to guarantee outcomes, in writing, on the homepage: "We don't guarantee meetings, pipeline, or ROI." That is more honest than most of this page manages.

The honest limitation. No Clutch profile, no Trustpilot profile, no independent review record of any kind. Every piece of reputation evidence is self-published.

They also publish their own "best cold email agencies" list ranking themselves first, written by a co-founder, with no conflict disclosure. And their FullPAC pipeline figure reads "nearly $10M" in one place and "potentially eight figures" in the testimonial beside it.

9. SalesCaptain

What it sells. Multichannel outbound across email, LinkedIn and text, with a dedicated team per account: an account manager, two business development reps, two researchers and a copywriter. No cold calling at any tier. They manage the day-to-day, with biweekly syncs and Slack access.

Pricing. Not published. No pricing page exists. Clutch lists a $5,000+ minimum project size and a $200 to $300 hourly band.

Proof, with one thing to watch. Twenty-six of twenty-seven case studies name the client, which is among the best disclosure on this page. But the headline metric is positive replies, not meetings: Palowise 521 positive replies in nine months, GlossGenius 253 in five months, Nestlé 323 with no timeframe. Only 5 of the 27 report booked appointments.

A positive reply is someone writing back without saying no. The distance between that and a held meeting is the whole middle of the funnel.

A name collision worth knowing about. There is an unrelated AI voice software company at a near-identical domain, priced at $159 to $300 a month. Several comparison articles blend the two, so any cheap monthly price attributed to this agency is almost certainly the software company.

The honest limitation. Their own site claims "50+ native US and UK business developers" and separately that they operate "across the USA, Canada, UK, Europe and Asia", and never reconciles the two.

Headcount reads 80+ on their site and 50-249 on Clutch. Headquarters is London on their own footer and New York on Clutch. None of those is damning on its own; together they're a reason to ask direct questions.

10. OutreachBloom

What it sells. Four separate motions, bought standalone or combined: cold email, AI SEO, Reddit marketing and LinkedIn outreach. The cold email product is a full managed operation, from domains and warmup through to reply management, requiring about an hour a month from the client.

Pricing. Published in full, which is rare in this category.

• Cold email from $2,400 a month, covering 10,000 to 100,000 contacts emailed
• AI SEO from $3,400, to $4,900 at the higher tier
• Reddit marketing $1,900
• LinkedIn outreach $1,000, sent from the client's own profile, with the client answering replies

Publishing real prices in a category this coy is a genuine differentiator, and the lowest entry point on this page by some distance.

Proof. Thinner than the price transparency suggests. RevenueBase reports 100% ROI within 30 days, though the interview text on the same page describes it as making their money back, which is breakeven rather than a double.

Labyrinth Learning reports 157% ROI within 30 days. Three of the six named case studies carry no metric at all, and four of the ten are anonymised. No meeting counts, reply rates or pipeline figures are published for any cold email client.

The honest limitation. The marketing page promises "month-to-month, no long-term contract" while the Terms of Use say "All sales are final", contain no service terms whatsoever, and were last modified in June 2023. Three reviews on Clutch and zero on G2 isn't a record. And a company selling "be the brand AI recommends" has no llms.txt of its own, which is a small thing that happens to be exactly the service being sold.

The ten, scored against the four standards

The two firms that publish the most about price, CIENCE and SalesRoads, also carry two of the larger review samples. The firm with no independent reviews at all is also the one with no terms page. Transparency tends to arrive in bundles, in both directions.

Six questions to ask on the first call

1. "What is the denominator on that reply rate, and does it count auto-replies?" The industry's famous 5 to 8% was measured against openers. Per email sent, the same publisher now reports 0.45%
2. "Send me the actual agreement, not the terms page." Nine of these ten publish no services agreement at all. Then search it for "initial term", "auto", "notice", "refund" and "cap"
3. "Where do the reps sit?" A published rate card in this category prices offshore at $1,750 and US at $3,500 for the same role. If the quote doesn't say, you can't tell which you're buying
4. "Is that price per month or per four-week period?" One agency here bills 13 times a year. On a single rep that is an extra $11,950 nobody budgeted for
5. "Does that number mean replies, meetings booked, or meetings held?" One firm's entire case study library headlines positive replies. They're three different things and only one of them is revenue-adjacent
6. "What happens if the rep assigned to me is not working out?" None of the ten publishes a replacement policy, so this one has to be put in writing