On this page
If you are reading this, there is a reasonable chance you already know how to get a vending location. You have done it. You have walked into a building, found the right person, and left with a machine placed. The problem is that whatever you did to get the last ten placements has stopped producing the next ten.
That is a different problem from not knowing where to start, and it has different causes. This article covers both, in that order: first what location sourcing actually is and the channels operators genuinely use, then why an approach that used to work stops working, and what changes the numbers when it does.
What a "location" actually is
A vending location is not a building. It is an agreement with a specific person who has the authority to let a machine sit in a space they control, and who expects something in return — usually convenience for their staff or customers, sometimes a commission, always the absence of hassle.
That framing matters more than it sounds, because it explains most of the failure modes further down this page. When operators say they "can't find locations," they usually mean one of four different things:
- They cannot find buildings that would suit a machine. This is almost never the real problem. Most metros have far more qualifying sites than any single operator can service.
- They cannot reach the person who decides. This is very often the real problem.
- They reach the person, but the person says no, or says yes and then goes quiet.
- They get a yes, place the machine, and the location underperforms badly enough that it should never have been placed.
Those are four separate problems with four separate fixes. Treating them as one problem — "I need more locations" — is why a stall is so often hard to diagnose: the symptom is identical in all four cases, and the remedy is not.
Who actually decides
In a small independent business, the decision-maker is usually the owner, and they are usually on site. This is the easiest case and it is the one most cold-walking advice is implicitly written for.
Everywhere else, the picture is different. In a manufacturing facility it may be the plant manager, the HR lead, or whoever owns "employee amenities" as a line item. In a healthcare setting it is frequently facilities management, and there may be a procurement process attached. In a multi-site operator — a gym chain, a franchise group — the person on site may have no authority at all, and the real decision sits at a regional office that has never heard of you.
The practical consequence is that the person who greets you is very often not the person who can say yes, and is sometimes structurally incapable of passing you to them. A receptionist who takes your card is not a lead. That interaction feels like progress and produces none, which is exactly why it is so easy to spend eight weeks on it.
The four ways operators source locations
Placements tend to arrive through four channels. Each has a real cost, and the costs are not comparable in the way they are usually compared — one is paid in hours, one in money, and the accounting rarely puts them side by side.
Cold approach — walking or calling
You go to the building, or you phone it. This is the default for new operators because it costs nothing but your time, and it works.
Its real cost is your hours, and its real limitation is that it does not scale with effort in a straight line: early approaches in a territory draw on the whole pool of prospects, later ones draw on what is left. The mechanism is in the plateau section below.
Referrals and expansion within existing accounts
The cheapest placements you will ever get. An existing host who is happy will introduce you to their neighbour, their other site, or the business park manager. Operators consistently under-invest here because it does not feel like prospecting.
Its limitation is that it is bounded by your current footprint, and it produces placements in bursts rather than on a schedule. You cannot plan a growth quarter around it.
Buying lead lists
Someone sells you a file of businesses. The price is usually low, and so is the value: a list of addresses is not a list of interested parties, and a list sold to you has very often been sold to others.
The specific thing to check is whether what you are buying is information or interest. A list of gyms within twenty miles is information you could assemble yourself. A location that has said yes to a conversation is something else entirely, and it is priced differently for a reason.
Disclosure: we sell location leads too, on our leads board, so treat this section as interested rather than neutral. The test above is the one we would want applied to us — and the standard we hold our own board to is that a lead is listed only once our callers have actually spoken to the location, and that it is sold once and then retired.
Locator and appointment-setting services
Someone else does the outreach and hands you either a lead or a booked meeting. The quality range here is wider than in any other channel, and the reputation is correspondingly poor — much of it earned. We are in this channel, so read what follows with that in mind. It is also the only channel that buys back the prospecting hours rather than spending them, which is why operators keep returning to it despite the category's history. It does not buy back all of them: you still attend the meetings, and any service worth using will require that you can.
If you are considering this route, our scam-avoidance and due-diligence guide is the companion piece, and it is worth reading before you take a sales call rather than after.
What is actually different in 2026
Most sourcing advice online is undated, and a good deal of it describes a market that has moved. Three shifts matter for where you point your effort.
Occupancy patterns changed, and in many markets have not returned to where they were. The reliable daily headcount that made a given office worth a machine is, in those markets, no longer the number on the lease. Sites that were obvious in 2019 are now marginal, and — more usefully — the reverse is also true. Facilities that hold their headcount through the week because the work is physically on site (manufacturing, logistics, healthcare, trades, education, anything with a shift pattern) have become comparatively better prospects than the office parks that used to be the default target.
The practical instruction is to stop qualifying on building size and start qualifying on present headcount and its consistency. Ninety people on site five days a week is a better prospect than a three-hundred-person office running at a quarter occupancy — fewer people in the building on an average day, and a far less predictable number of them. The office looks better from the parking lot, which is the problem.
What counts as a placement has widened. Micro-markets and smaller unattended formats mean a site that cannot justify a full bank of machines can still justify something, and a host who says no to a machine may say yes to a different footprint. Operators who pitch one format lose sites that a different format would have won. This also cuts the other way: it is now easier for a competitor to serve a site you passed on.
Cashless changed the objection list. The host objections that used to be about mess, theft and coin handling have shifted toward reliability, restocking cadence and whether their people will actually be able to pay. Theft and vandalism have not gone away — machines still get broken into — but if the objection you rehearse hardest is one your last several hosts never raised, you are spending your best minutes on the wrong reassurance.
None of this changes the fundamentals below. It changes which sites are worth your hours, which is a sourcing question rather than an operating one.
The incumbent problem
"We already have a vendor" is the non-answer you will hear most, and it is worth handling deliberately, because it is usually treated as a hard no when it is closer to an unqualified maybe.
Three situations hide behind that sentence. The site is genuinely well served and happy — a real no, and the correct response is to note the date and leave. The site has a vendor and is not happy, which is extremely common and almost never volunteered to a stranger at the front desk. Or the site has a vendor for part of what they need — a drinks machine and no snacks, a machine in one building and nothing in the other three.
The second and third are real opportunities, and the thing that separates operators who find them from operators who do not is a single follow-up question asked without pressure. Not "are you happy with them" — which invites a polite yes — but something specific and low-stakes: how often it gets restocked, whether it covers the whole site, whether it has been out of service recently.
The other thing worth doing is keeping the date. An incumbent relationship you cannot displace today is a prospect with a clock on it. Logging a "has a vendor" with a date and revisiting a year or so later costs almost nothing and occasionally converts; writing them off permanently converts none of them by construction. This is one of the few places where a spreadsheet outperforms effort.
Why an approach that worked stops working
Here is the part most sourcing advice skips, because it is not a beginner problem. You had a method. It produced placements. It is now producing fewer, or none, and you have not changed anything. Four things are usually happening, often at once.
Territory exhaustion is not about running out of buildings
You have not run out of sites. You have run out of unworked sites within the radius you are willing to service. Those are different, and the second one arrives much sooner than operators expect.
Every approach you make marks that site as worked for some period. A "no, we already have a vendor" is a no for the length of that vendor's relationship, which you cannot see. As your worked set grows, the density of genuinely fresh prospects inside your radius falls — and your hit rate falls with it, even though your technique has not changed at all.
This is the failure we see misdiagnosed most often: the operator concludes the pitch has gone stale when the pitch is fine and the pool has thinned. The two feel identical from the inside and have opposite remedies.
The qualification gap
Cold approach qualifies late. You find out whether a site is viable — right size, no incumbent contract, a decision-maker who will engage, a location that will actually move product — at the end of the process, after you have spent the time.
That is tolerable at low volume and brutal at high volume, because the cost of disqualification is paid in your hours and it is paid after you have already spent them. Every hour spent on a site that was never going to work is an hour not spent on one that would.
Decision-maker mismatch
Related, and worth separating. As operators grow, they naturally move up-market: bigger sites, more machines per placement, better economics. Bigger sites have more layers between the front door and the person who decides.
The technique that worked on a forty-person auto shop — walk in, ask for the owner, make the case in four minutes — does not transfer to a facility with a procurement process. Operators often experience this as "my close rate dropped" when what actually happened is that they changed market segment without changing method.
The arithmetic stops working before it feels like it stops working
Do this with your own numbers rather than anyone else's, because the industry averages you will find online are mostly unsourced.
Take your last complete month. Count the approaches you made — every call, every walk-in, every follow-up. Count the conversations that reached a decision-maker. Count the meetings that happened. Count the placements. You now have a four-stage funnel with real numbers in it.
Two things usually fall out of that exercise. The first is that the drop-off is concentrated in one stage, not spread evenly — and it is very often the approach-to-decision-maker stage, not the meeting-to-close stage operators tend to blame. The second is that the hours per placement are higher than you thought, because the time spent on sites that went nowhere is invisible until you count it.
One caution before you compare anything. Some channels sell placements, some sell leads, and some — ours included — sell booked appointments. An appointment is a meeting with a decision-maker, not a placed machine, so comparing a per-appointment price against your own cost per placement is not like for like. Convert first: divide by the share of meetings you actually close. If you close half your meetings, a per-appointment price is twice that number per placement. Any service that lets you skip that arithmetic is not helping you.
With that conversion in hand, hours per placement is the number that lets you compare channels honestly. A placement that costs you fourteen hours of your own time is not free, and whether it is cheaper than paying someone else depends entirely on what your hours are worth in your business — which for most full-time operators is a real number, because those hours have an alternative use in servicing the route you already have.
What actually changes the numbers
Qualify before the meeting, not during it
The highest-leverage change available to most stalled operators is moving disqualification earlier. Anything that tells you a site is not viable before you spend an hour on it converts directly into capacity.
In practice: know the site's approximate headcount before you approach. Know whether there is an incumbent machine. Know who the decision-maker role is likely to be for that kind of business. None of that requires special data — most of it is visible from the outside, from a website, or from one question to whoever answers the phone.
Radius discipline
A location forty minutes outside your service pattern is not a location. It is a future service problem that will cost you more than it earns, and it will still be costing you in eighteen months.
Operators under growth pressure relax radius first, because it instantly expands the prospect pool. It is the change most likely to be regretted. If you are going to widen, widen deliberately — pick a second cluster and work it as a cluster, rather than letting individual outliers accumulate across the map.
Target the role, not the building
Decide which decision-maker role you are trying to reach before you approach, and let that determine the approach. Reaching a plant manager is a different task from reaching an owner-operator, and the method that suits one is close to useless for the other.
This is also the fix for the up-market transition described above: the plateau is usually not a technique failure but a technique-to-segment mismatch.
Track the funnel you actually have
You cannot fix a stage you are not measuring. The four-stage count described earlier, kept for three months, will tell you more about your sourcing problem than any amount of advice — including this article — because it is about your territory, your segment and your approach rather than someone else's.
The placement that should not have happened
The fourth failure mode from the top of this article deserves its own treatment, because it is the one that quietly undoes the other three. You can source well, close well, and still end up worse off by placing a machine that should never have gone in.
An underperforming location is not neutral. It occupies a slot on a service route, and the cost is the drive time, the restocking labour, the capital in the machine, and the opportunity cost of the site you would otherwise have serviced in that window. A machine earning a third of what it should is not making you a third of the money — once servicing costs are counted it may be making you nothing, while consuming a real slice of your week.
The signals are visible before placement more often than operators admit. Headcount that is present but not captive — a site where everyone drives past three shops on the way in. A decision-maker who is indifferent rather than interested, which predicts poor internal communication about the machine and poor placement inside the building. A location within the building that is convenient for installation rather than convenient for the people who would buy. A host who wants a commission conversation before they want a service conversation.
The discipline that helps is unglamorous: decide your minimum before you walk in, in whatever terms you actually manage the route by, and hold to it under pressure. Operators are most likely to break their own minimum during exactly the plateau this article is about, because a marginal yes feels better than another week of no. That is the moment it costs the most, because you are also least able to afford the route capacity.
If you already have placements under your minimum, they are worth counting separately in the four-stage exercise above. Removing a bad location frees the same route capacity that finding a good one does, and it is faster — but it is not free. Check the host agreement's termination terms first, count the removal and transport cost, and remember that a host you exit badly is a referral you will not get. Exiting well is worth the extra conversation.
If you decide to buy help
If you have placed fewer than five machines, skip this section for now — the section after it is the one written for you, and buying outreach is not the constraint at your stage.
For everyone else: at some point the arithmetic favours paying someone else to do the outreach. That point arrives earlier than many operators expect, because they undercount their own hours.
The category has a genuine trust problem, and the burden is on any service to answer it concretely. Before paying anyone, the things worth establishing are what exactly you are buying — information, a lead, or a meeting that will happen — what happens when what you are sold does not meet the description, whether what you buy is sold to anyone else, and who bears the cost when a meeting does not hold.
Those questions are the subject of our scam-avoidance and due-diligence guide.
For our part, and so you can hold us to the same questions: what we sell is a booked appointment, not a placement. The standard it is measured against is published at the rubric — radius compliance, the target decision-maker persona, no-show defense, double-booking exclusion, the dispute timeline, and the post-meeting follow-up window. The charge fires when the appointment is booked, not after the meeting, and an appointment that misses the rubric is replaced at no charge rather than argued about. Leads we sell are sold once and then retired, never resold. There is no retainer required. The commercial terms are at pricing.
Where to go from here
If you have placed fewer than five machines, the constraint is almost certainly not sourcing technique — it is that you do not yet have enough placements to see your own pattern, and the fastest thing you can do is get the next few placed yourself. Our first five placements playbook is written for that stage. Buying outreach is not the lever yet, and we would not sell you a pilot at this point either.
If you are already placing and have plateaued, run the four-stage count on your last complete month before changing anything. If the drop-off is at the approach-to-decision-maker stage — which is where it usually is — the fix is earlier qualification and role targeting, and you can make both changes without spending anything.
If you have run that exercise and the answer is that you are out of hours rather than out of method, that is the point where paying for outreach starts to make arithmetic sense — with one condition worth knowing first. Booked appointments only help if you can attend them: we ask whether you can make a scheduled meeting within five business days, and if the honest answer is no, more meetings will not fix your sourcing problem and we will tell you so. Coverage is also scoped to the verticals we serve today.
To see what is on offer first, browse the marketplace: anonymised venues that want machines, and a place to list your machines so hosts can find you. Our team makes every introduction by phone. The commercial terms are published at pricing, and how it works walks the process end to end.