Mid-Market Sales: Why It Is Not Small Enterprise or Big SMB

Before you argue about a mid-market playbook, try getting two people in your company to agree on which accounts are in the segment. One will say revenue, one will say headcount, and the numbers they pick will differ by a factor of five. That is not sloppiness on their part. The institutions that define the term do not agree either.

This piece sorts out the definitions first, because a quota set against a boundary nobody shares is a quota nobody can hit. Then it covers what genuinely changes in the deal, what the available data supports, and what to measure in your own book. It is written for an AE or a manager arriving from either direction, SMB or enterprise.

Nobody Agrees Where the Middle Starts

The number most sales content repeats comes from a single vendor page. Dock's guide to the segment pegs the segment at annual revenue between $50 million and $1 billion, headcount between 200 and 1,000, and a process running 2 to 6 months. Read the page closely and there is no citation attached to any of it. No survey, no sample, no outside body. It is one author's working rule, repeated until it acquired the texture of a fact.

Now compare the institution that actually owns the term. The National Center for the Middle Market, housed at Ohio State's Fisher College of Business, has run its Middle Market Indicator twice a year since 2012. Per its published methodology, the survey covers US companies with annual revenues between $10 million and $1 billion, polls roughly 1,000 CEOs, CFOs and other C-suite executives in each wave, and treats the segment as representing about a third of private sector GDP. Its own internal tiers are $10 to 50 million, $50 to 100 million, and $100 million to $1 billion.

That floor is five times lower than the one every sales blog quotes. On the NCMM's definition, the band most sellers call mid-market is actually the upper two thirds of it, and an entire tier of companies sits below the line sales teams have drawn.

Cross a border and it inverts completely. The European Commission's official categories, derived from Recommendation 2003/361, cap a medium-sized enterprise at fewer than 250 staff with turnover no greater than €50 million, per the Commission's SME definition page. So a company with 400 employees and $80 million in revenue is comfortably mid-market in a US sales org and a large enterprise in European statutory terms.

Who is defining it	Where the middle sits	What backs the number
Dock, a vendor guide	$50M to $1B revenue, 200 to 1,000 staff	No source stated on the page
National Center for the Middle Market	$10M to $1B revenue, with three internal tiers	Twice-yearly survey of about 1,000 executives since 2012
European Commission	Under 250 staff, turnover up to €50M	EU Recommendation 2003/361

Four Authorities, Four Different Middles

Most guidance on this segment opens by telling you what mid-market is. Our position is that the definition is the wrong place to start, and treating it as settled is the single most expensive habit in the segment.

Look at what the evidence actually shows. The figure the industry repeats has no source behind it. The body that named the category, per its own methodology, sets its floor an order of magnitude lower. European law puts the same company on the other side of the line. These are not three refinements of one truth, they are three different instruments measuring different things for different purposes, and only one of them was built to describe how a company buys.

What follows practically is that the boundary is your decision and should be made on buying behaviour rather than on revenue. The useful question is not how much a company earns. It is how many people have to agree, whether a security review exists, and whether procurement is a department or a person with a second job. Draw the line there, write it down with marketing and finance in the room, and accept that it will not match anybody's published range.

A study showing that revenue band predicts buying-group behaviour better than a direct measure of it does would overturn this. That would make revenue the efficient proxy and the argument above a waste of a meeting. Nothing in the sources above tests it, and the fact that three credible authorities disagree by a factor of five is reason enough to measure the thing you care about directly.

The Only Cycle Figure Worth Quoting, and Its Caveats

Everyone wants a cycle length for the segment. There is almost nothing honest to give you. The closest usable data point comes via SaaStr: in December 2024, Jason Lemkin reported from Gong's analysis that deals around $100,000 take roughly 70 days to close, drawn from Gong's own customer base.

Three caveats, all of which matter. It is a single data point, not a breakdown by segment, so it cannot tell you how a $40,000 deal or a $400,000 deal behaves. The page states no sample size. And it reaches you through SaaStr's reading of Gong's report rather than from Gong directly, so treat it as second-hand. Set against Dock's unsourced 2 to 6 months, the pair at least converge on a magnitude: this is a matter of weeks or a couple of months rather than a calendar year. For the far end of the spectrum, see our piece on long enterprise cycles.

Why Acquisition Gets Harder Exactly Here

Here is a pattern that explains more about this segment than any cycle figure. Benchmarkit's 2024 B2B SaaS performance report, built on data from roughly 1,000 participating companies with a stated sample of 936 for its participant profiles and covering calendar year 2023, segments its metrics by average contract value. Per that report, the customer acquisition cost ratio rises as contract value rises, and it notes the increase becomes dramatic beginning around $50,000 in average contract value.

That threshold is roughly where mid-market deals start. It means the segment is structurally more expensive to sell into per dollar of new revenue than the band beneath it, and the expense arrives before the enterprise-scale contract values that justify it.

The compensating finding sits on the other side of the ledger. The same report states that gross revenue retention above 90% has historically been associated with contract values greater than $100,000, and that retention correlates most strongly with average contract value. So the middle is where acquisition stops being cheap and retention starts being good, which is exactly why a motion borrowed from either neighbour fails: the SMB playbook cannot absorb the acquisition cost, and the enterprise playbook assumes a contract value that is not there yet.

Rep Time Is the Lever, Not Process Weight

The instinct when deals get more complicated is to add process. The available evidence points somewhere cheaper. Gartner surveyed 645 B2B buyers across August and September 2025, and per the release carrying the findings, the buyers who gave supplier reps more of their time reported the least dysfunction inside their own group, and low-dysfunction groups were 13x likelier to call the resulting deal a good one.

Read that carefully before you borrow it. This is a broad B2B result with no segment breakout, and deal quality here is the buyer's self-assessment rather than a measured win rate. What it supports is modest and useful: time with a rep is associated with a buying group that argues less, and a group that argues less reports better outcomes. In a segment where three or four people must agree without a formal procurement function to arbitrate, helping them agree is the work. Mapping that group before you need it is where it starts, and why one ally cannot carry a decision is the companion argument.

Andy Sietsema, a senior account executive at Dock, frames the temperament the segment needs as requiring "a lot more patience" than SMB. That matches the structure: you cannot compress a group's decision by selling harder at one member of it.

SMB, Middle, Enterprise: What Actually Changes

	SMB	The middle	Enterprise
Who must agree	One person, sometimes two	A handful, informally	A formal committee
Who arbitrates	The buyer	Nobody	Procurement
What kills the deal	Lost attention	Unresolved disagreement	Review queues
Cheapest lever	Speed	A shared business case	Breadth of contact
Where the cost lands	Your hours	Acquisition cost per dollar	Cycle length

The second row is the one worth sitting with. In an enterprise deal, somebody whose job is to arbitrate will eventually force a decision. In this band nobody has that job, so an unresolved disagreement does not get escalated, it just goes quiet. That is the characteristic failure mode, and it looks nothing like the failure modes on either side.

Running a Deal in This Band

1. Qualify on consequence rather than seniority. Put the question as: suppose this quarter ends and nothing has been done. A shrug is your answer about the deal.
2. Surface voices two and three before call three. Not for an introduction yet. Just learn who exists and what each would object to.
3. Build the numbers with them, do not present them. A business case the buyer helped build is a case they will defend in a room you are not in, which is the whole method in our guide to business cases that close.
4. Put dates on everything. With nobody arbitrating, a calendar is the only thing resisting drift.
5. Keep the process light. Two or three clear commitments per call. A ten-stage pursuit plan on a $60,000 deal burns the acquisition cost the benchmarks already warned you about.

Build Your Own Numbers, Not Ours

Given how little the published data says about this segment specifically, your own closed deals are the better source. Four figures drawn from your trailing year are enough to operate on.

• Elapsed days between the opening meeting and the signature, bucketed into deal sizes you pick yourself.
• Count of distinct people involved, won deals against lost ones.
• Win rate by where the opportunity originated.
• Share of deals where a dated mutual plan existed at the proposal stage.

Compare the won column to the lost one. The gap in the second and fourth rows is usually where the change belongs, and nobody can tell you in advance what yours will show. That uncertainty is the argument for measuring rather than borrowing.

If the Segment Is Wearing You Down

Deals that go quiet are harder on a person than deals that lose cleanly, and this band produces more of them. Two practical things. Ask a manager to sit in on a middle call and a late call of the same deal, because a second listener spots an absent stakeholder faster than the person running it. And if the pressure is reaching your sleep or your health, that belongs with a licensed professional rather than with sales content. In the United States the 988 Suicide and Crisis Lifeline answers both calls and texts on 988. Nothing published here is a substitute for that.

One Paragraph on the Middle

This band is neither a scaled-down enterprise nor an inflated SMB. It is the place where a group has to reach agreement with nobody employed to make them, where acquisition cost has already climbed but contract values have not caught up, and where every published definition of the boundary contradicts the next one. Define it by buying behaviour, run a light process with hard dates, spend your time helping the group agree, and build your cycle numbers from your own closed deals rather than from anyone's range, ours included.