When outbound is the wrong answer for your company

Outbound is the wrong answer more often than the market admits, and the market has a reason not to admit it. Twenty-two outbound and lead generation agencies were read on their own live sites, and three publish anything about who the service is not for. Gravion Scale bills only for attended meetings, at $150 to $700 quoted per ICP, which means an engagement that should never have started costs a client nothing. The cases where it should never start are worth knowing before anyone signs.

Three of twenty-two agencies say who they are not for

Agency	What it publishes	Where
Growth Engine X	Not a fit for “Small Target Markets (Under 40K Contacts)”, “Lifetime Gross Profit Under $3,000” and “No Sales System to Leverage the Leads We Generate”	Homepage
Sapper Consulting	Serves “Any industry that serves a B2B market with a TAM greater than 2000 companies”	Homepage FAQ
Beanstalk Consulting	”If we don’t think it’ll work, we’ll tell you and point you elsewhere”	Homepage FAQ

Growth Engine X is the only company in twenty-two attaching numbers to a refusal, and the numbers are specific enough to act on: a market under 40,000 contacts, a lifetime gross profit under $3,000, or no sales team to work the meetings. Sapper’s floor of 2,000 companies is phrased as an inclusion rather than a refusal, so read it as a hint rather than a commitment. Beanstalk commits to the behaviour without attaching a threshold, which is worth something and is not the same as publishing one.

The other nineteen publish nothing. Not a minimum deal size, not a minimum market, not a required sales capacity, not a declined industry. Kalungi is asked directly whether not being venture backed disqualifies a buyer and answers that it does not: “You can still work with Kalungi if you are not a venture backed company.” Operatix states that its “client portfolio consists 100% of B2B Software vendors”, which describes who buys rather than who is turned away.

A vendor that never publishes a floor has no published reason to decline anyone, and a buyer researching fit finds an entire category agreeing that the answer is yes.

In Germany, a cold email to a business needs prior consent

The legal geography is sharper than most outbound writing allows, and it decides whether a campaign is viable before any question of economics arises.

Jurisdiction	Position on unsolicited B2B email	Source
Germany	”the addressee’s prior express consent” required for advertising by electronic mail, with no business exception	UWG section 7(2) no. 2
France	Prior consent “n’est pas systématiquement requis” for B2B prospecting, subject to conditions	CNIL guidance
United Kingdom	”You can email or text any corporate body (a company, Scottish partnership, limited liability partnership or government body)“	ICO guide to PECR
Canada	Prohibited unless the recipient “has consented to receiving it”, and the sender “has the onus of proving it”	CASL sections 6(1) and 13
United States	No consent required; “The law makes no exception for business-to-business email” for the other rules	FTC CAN-SPAM guide

The German provision repays close reading, because the distinction it draws is deliberate. Section 7(2) no. 1 treats a telephone call to a consumer as needing “prior express consent” and a call to “another market participant” as needing only “at least the latter’s presumed consent”. So German law knows exactly how to write a business exception, and wrote one for cold calling. Number 2, covering electronic mail, requires “the addressee’s prior express consent” and draws no consumer or business distinction at all. The carve-out exists for the phone and was withheld from email.

France reaches the opposite conclusion under the same European directive, because Article 13 of the ePrivacy Directive left the treatment of legal persons to member states. The United Kingdom lands closer to France, with one trap worth knowing: the ICO’s “corporate body” excludes sole traders and most partnerships, so a list of small British businesses is not automatically a corporate subscriber list.

Canada is the quiet risk. The prohibition itself is unremarkable, but section 13 puts the burden of proving consent on the sender, so an agency running a Canadian campaign has to be able to evidence consent for every address rather than wait to be told to stop.

Any provider proposing a single playbook across those five markets is proposing something that is lawful in two of them.

Your market may be too small, and you can count it exactly

No regulator and no independent body publishes a threshold for when a market is too small to work. What official statistics do publish is the denominator, and that is the more useful thing.

The United States Census Bureau’s Statistics of US Businesses gives firm counts at full six-digit industry detail, including a table broken out by enterprise receipts size, so a seller can count the firms in a narrow niche above a revenue threshold. The Office for National Statistics publishes the equivalent for the United Kingdom at four-digit industry detail by employment size band, drawn from the Inter-Departmental Business Register.

Counting first changes the conversation. A niche holding 1,800 firms is not an outbound market at any price, and no agency will tell you that, because no agency in twenty-two publishes a floor low enough to catch it except by implication. A niche holding 90,000 is a different proposition and the question becomes economics rather than arithmetic.

What the count tells you	Roughly how many companies	What outbound is worth
Too small for a programme	Under about 2,000	Nothing an agency should sell you; work it by hand
Too small for a per-meeting rate	Under about 10,000 reachable	A list a small team finishes without help
Workable, economics decide	10,000 to 100,000	Depends entirely on deal value and reachability
Comfortably a market	Over 100,000	Coverage and sending capacity become the constraint

Read those bands as judgements rather than published standards, because that is what they are. Growth Engine X’s published 40,000 floor sits inside the third row; Sapper’s 2,000 sits on the first boundary. Nobody in this market publishes anything finer, so a buyer counting their own industry against official statistics knows more about fit than any agency page will tell them.

Nobody has ever independently verified the economics

No statistics agency and no peer-reviewed source publishes B2B customer acquisition cost, sales cycle length by deal size, or outbound win rates. Every number in circulation on whether outbound pays comes from a company selling into the market, self-selected and unaudited.

The most methodologically serious of them is worth knowing anyway. The Bridge Group’s 2025 report, drawn from 351 B2B companies, puts the median monthly meeting quota at “10 (global median)”, notes it is down “40% since 2018”, and reports “60% Reps at quota Lowest on record”. A channel whose own practitioners have cut the expected output per head by 40% in seven years, and still miss it four times in ten, is a channel that deserves a fit test before a contract rather than after one.

What Gravion Scale declines

Markets under roughly 10,000 reachable companies are turned away, because a per-meeting rate is the wrong instrument for a list a person could work by hand. Annual contract values under about $10,000 are turned away for the mirror reason: the meeting costs more than the deal returns at any point in the band.

Campaigns into Germany, and into any market requiring prior consent that a cold list cannot supply, are not run as cold email regardless of what a client is willing to accept. A provider willing to run them is transferring a legal exposure onto the client’s own company, and no billing model makes that a good trade.

A company with no free capacity to take meetings is asked to fix that first. Attendance is the billing trigger here, so meetings that nobody follows up are the one failure mode the model does not protect against; the invoice is still correct and the money is still wasted.

What it costs, and what happens if the answer was wrong

The rate is $150 to $700 per attended meeting, fixed per ICP at quoting, with deal value and market difficulty setting the position in the band. Nothing is owed before a prospect attends, no-shows are not billed, and the term is open ended with 14 days’ notice.

Billing on attendance is the practical version of everything above. An engagement that should not have started produces no invoice, so the cost of a wrong fit falls on Gravion Scale rather than on the client. No rate card is published anywhere, so a figure found elsewhere is unverified until it arrives in writing.

How this was checked

Twenty-two agencies were read on their own live pages in September 2026, covering homepages, service and pricing pages, and FAQs. Four terms of service pages could not be retrieved, at Belkins, Upcall, SalesRoads and CIENCE, and a prohibited-industries clause could plausibly sit in any of them, so the finding for those four is that nothing is published where a buyer would look rather than that nothing exists. Legal provisions were read at source: the official English translation of the UWG published by the German Federal Ministry of Justice, the ePrivacy Directive on EUR-Lex, CASL on the Justice Laws website, and guidance published by the ICO, the CNIL and the FTC. Three passages could not be retrieved verbatim and are described rather than quoted. Nothing here is legal advice, and a company operating across several of these jurisdictions should take its own.