How to Size Your Addressable Market Before You Prospect

Every serious outbound conversation arrives at the same question, and it is almost never the first one asked. It comes after the pricing, after the guarantee, after the mechanics have been explained twice. Then there is a pause, and some version of: will you even find enough of my people?

It is the right question, asked at the right time, and it is the one the internet is worst at answering. Search for how to size your addressable market and you will get a formula that produces a number in dollars. That number is genuinely useful — to someone raising money. It cannot tell you whether your campaign runs out of businesses to contact in month three, which is the thing you were actually asking.

This article is about the other number: how many businesses in your niche you can realistically identify, reach, and message, in the geography you are willing to work. It is a smaller number than most people expect, it shrinks in stages that nobody publishes, and you can estimate it yourself in an afternoon before you commit a dollar to a campaign.

The two answers the internet gives you, and why neither one fits

Search the question and you land in one of two bodies of content, both competent, both aimed at somebody else.

The first is the TAM, SAM and SOM literature. It is written for founders assembling a pitch deck, and its method is consistent everywhere you find it: count the potential customers, multiply by what an average customer spends per year, and report the product. Then carve that down to a serviceable slice, and again to the slice you think you can win.

The mechanics are fine. The output is a currency figure describing the size of a prize, and what it does not describe is the effort required to reach the prize — the two are not proportional. Some of the better outbound agencies publishing this framework acknowledge as much in their own articles: a market-size calculation does not tell you how many of those companies genuinely fit your profile, or whether you can get to the decision-maker at all. They name the gap accurately, and then move on without filling it.

The second body of content is list-building advice for cold outreach. It is more practical and much closer to home, but it answers a different question again: how big should any one campaign's list be, and how much should you personalize it. The consensus advice — that a small, tightly relevant list beats a large generic one — is correct. It is advice about the quality of a batch. You can follow it perfectly and still have no idea how many batches your niche contains before you have spoken to everyone in it.

One literature gives you a prize, the other gives you a serving size. Neither gives you the inventory count, and the inventory count is what outbound actually consumes.

The unit that matters: contactable records

Outbound is not funded by dollars of market. It is spent in records. Every message you send uses up one business — not forever, but for a long interval — and when the records run out the campaign stops working regardless of how large the dollar figure on your market was.

So the unit to count in is the contactable record: a business that fits your criteria, that you can identify from outside, that has a working contact point of the type your channel needs, and that you have not already burned. Each of those four conditions removes businesses from the count, and they compound.

This is why a market can be enormous in dollars and nearly empty in records. A category of large firms with high contract values might have a magnificent market size and be almost impossible to reach cold, because every listed number routes to a switchboard and the person you need has no public contact point. Meanwhile a category of owner-operators with modest deal sizes can be trivially reachable, because the number on the listing rings the phone in the owner's pocket. The dollar figure ranks those two markets in exactly the wrong order for someone about to run outbound.

It is also why "how many businesses are in my niche" has two honest answers that differ by a large multiple, and why vendors and buyers routinely talk past each other on it. One of them is quoting listings. The other is asking about messages.

The shrink chain: from a category to a list you can message

Between the category and the campaign sits a sequence of reductions. Each one is a multiplier on the count, each one is measurable, and skipping any of them produces an estimate wrong by enough to change your decision.

Two of these deserve expanding, because they are where most estimates go wrong.

Category accuracy cuts both ways

People assume the category filter is roughly right and worry only about the count. In practice it leaks in both directions. Some businesses sitting in your category are not remotely your buyer — a national franchise location, a defunct listing nobody removed, a business filed under your category because it was the closest option available when someone set the listing up. And a meaningful share of your actual buyers are filed somewhere else entirely, for exactly the same reason.

The consequence is that widening your category set often adds more real prospects than widening your geography does, and it is cheaper. Which categories to include is a judgment call about your offer rather than a data problem, and it is worth making deliberately instead of accepting whatever label you happened to search first.

Line type is the reduction nobody counts

A phone number in a business listing tells you almost nothing about whether a text will arrive. It might be a mobile line, a landline that silently discards messages, or a VOIP number that behaves unpredictably. The listing looks identical in every case. This is where an optimistic estimate and a real one diverge most sharply, because the reduction happens after every other filter has already been applied — it applies to the survivors, so it hits the number you were counting on.

It also varies enormously by category, which is why a rate borrowed from someone else's niche is worse than useless. Categories dominated by solo operators and mobile trades tend to publish the owner's own cell. Categories built around a physical premises with a front desk tend to publish the desk. You cannot reason your way to the number; you have to check a sample. It is one reason where your data comes from changes the arithmetic — a list sold by the record has no incentive to tell you what fraction of it is unreachable. TaskBlink validates every number as a real working cell before it enters a campaign, which is less a feature than an admission that the raw count is never the count.

There are also compliance reasons to know the line type of everything you send to, separate from whether the message arrives. Those are covered in our piece on what the rules actually say about B2B cold texting, and they are worth reading before you build a list rather than after.

Measure your own multipliers with one city

The temptation is to hunt for published rates for each stage above and multiply them together. Resist it. Those rates vary by category, by geography and by data source, and stacking six borrowed estimates produces a number with no relationship to your situation.

Do this instead. Pick one metro you would genuinely work. Pull the whole category for it. Then walk a sample through every stage of the chain and record what survives each one. You now have your own multipliers, measured on your own category, and you can extrapolate to the full geography you are considering. The exercise takes an afternoon, and it is the difference between a market size you can defend and one you inherited from a blog post.

Two cautions on the extrapolation. Density is not uniform — a category can be several times denser per capita in one region than another, so scaling a single metro by population is a rough instrument and the result deserves to be treated as a range rather than a figure. And your filters are not free: every criterion you add removes businesses, which is correct and desirable, but the cost has to be visible at the sizing stage rather than discovered in month two. Our guide to qualifying filters that keep junk off your calendar covers which filters genuinely exist in local business data and which ones you are imagining.

What to do with the number once you have it

A pool count on its own is trivia. It becomes a decision when you divide it by the rate at which you intend to contact people, which converts a count into a duration — how many months of campaign your niche contains at your planned pace. That is the number worth carrying into a vendor conversation, and it is the one that makes two otherwise identical proposals look completely different.

How to set that pace is genuinely its own problem: monthly volume is the lowest of several ceilings and market depth is only one of them. How many cold messages you should send per month works through the others. What matters here is that the two numbers are meant to be decided together. Sizing a pool without a pace, or setting a pace without a pool, leaves you with half a plan either way.

Broadly, the answer lands in one of three shapes.

A deep pool — years of runway at your intended pace — means depth is simply not your constraint. Stop thinking about it and go argue about targeting and message quality instead. This is the common case for national campaigns in broad categories, and it is why national outreach forgives a good deal of inefficiency that a local campaign would not.

A shallow pool — several months rather than years — is workable, but it changes the operating discipline. You cannot afford to burn the list with a message you have not tested, because there is no second copy of it. Testing on a deliberately small slice stops being a best practice and becomes the only responsible option.

A thin pool — weeks — means the honest conclusion may be that cold volume is the wrong instrument for this market, and that something closer to account-by-account work is what its shape is asking for. That is a real answer, and it is much cheaper to reach before you have spent money discovering it.

In all three cases the pool renews eventually, because a business you contacted and did not convert is not permanently spent. What the right re-approach interval is, and what to do at the moment a campaign genuinely exhausts its category, is a separate problem from sizing and we will treat it on its own.

How to ask a vendor for your pool size

The sizing question is also one of the better diagnostics available on a prospective vendor, because the shape of the answer tells you how they work. There are three answers and they rank cleanly.

The weakest is a large national category count delivered quickly. It means nobody has applied your filters, and the number describes a database rather than your market. The middle answer is a filtered count for your geography, which is a real answer and shows they have at least run the query. The strongest is a filtered, validated count plus a sample of the actual businesses — the sample being the part that cannot be faked, and the only way you can check category accuracy yourself.

Useful follow-ups, in rough order of how much they reveal:

• How many records survive after line-type validation, not before?
• What geography and what category set produced that number?
• Can I see fifty of them, so I can judge whether they are my buyer?
• How many of these has anyone contacted before, on any campaign?
• At the volume you are proposing, how many months does this pool last?

An honest vendor answers all five, and will sometimes tell you the number is too small — which is the response most worth paying for. On a TaskBlink demo the pool for your niche is pulled and shown live rather than described, along with the messages that would go to it, partly because it is the fastest way to have this conversation and partly because a category that will not support a campaign is better found on a call than in month two.

When the number comes back small

A small pool is information, not a verdict, and it has far more remedies at the sizing stage than it does later. Before treating it as a problem, check the three things most likely to have caused it artificially: an over-narrow category set, a geography chosen for comfort rather than economics, and a filter stack you never priced. Widening any of those is free at this point and expensive after launch.

If the pool is genuinely small after those checks, the useful reframe is that pool size and deal size trade against each other. A thin market with high per-client value can be entirely viable — it simply cannot be run as a volume campaign, and the economics have to work at a much lower contact count. That is a different calculation, and the affordability side of it is covered in which niches can actually afford your agency, while the channel side — where market depth decides whether text or email is the cheaper route to an appointment — is worked through in cold SMS versus cold email on cost per appointment.

What a small pool should never do is get silently corrected by loosening your criteria after launch. That is the failure mode this whole exercise exists to prevent: a campaign that appears to be running fine while quietly contacting businesses you would never have chosen, because the alternative was running out. Decide the pool honestly first, and the rest of the plan — the budget, the pace, the offer — has something real to sit on.

Frequently asked questions