PPC in New York: The Most Expensive Auction in the Country, and Three Ways Budgets Leak Into It

Paid search management in New York is usually charged as a floor of $2,200 to $6,500 a month for a single business, or 8 to 14 percent of spend once media passes roughly $40,000 a month. A local account in most categories needs $6,000 to $20,000 a month in media before the data arrives quickly enough to make decisions on. Those are my own working figures from how this market quotes, offered as a benchmark for reading a proposal rather than as published rates.

New York is the most expensive search auction in the United States and that is a fact about your competitors, not about Google. It means two things in practice: a mistake costs more here than anywhere else, and the three or four percent of waste that a smaller market tolerates is the difference between profit and loss in this one.

What a click costs in the five boroughs

Category	Search cost per click	What that means for the budget
Personal injury and mass tort	$130 to $380	Below roughly $25,000 a month in media you cannot gather usable data
Addiction and mental health treatment	$90 to $300	Heavily contested and tightly policed, so certification is a prerequisite
Emergency plumbing and heating	$38 to $105	Works at modest spend if calls are answered live around the clock
Cosmetic dentistry and aesthetics	$28 to $80	Profitable at $4,000 to $9,000 a month in media with a strong landing page
Commercial real estate and office space	$20 to $70	Long cycle, so judge it on qualified enquiries rather than conversions
Business to business software	$22 to $65	Cheaper than the category average nationally, and lead quality is the whole problem

Leak one: one campaign for eight million people

The most expensive error in this market is treating the city as a single location. Somebody in Astoria will not travel to Park Slope for a dentist, a Riverdale household and a Flatbush household behave differently, and a Staten Island customer may genuinely be easier and cheaper to win than a Midtown one.

Build the account by borough and then by commercial cluster inside it: Midtown, the Financial District, Chelsea, the Upper East and West Sides and Harlem in Manhattan; Williamsburg, Park Slope, Downtown Brooklyn and Bay Ridge in Brooklyn; Long Island City, Astoria, Forest Hills and Flushing in Queens; Riverdale and Fordham in the Bronx; and Staten Island on its own. Each gets its own budget, bids and landing page.

Radius targeting is wrong in this city and it is wrong for a specific reason. People here measure journeys in subway stops and transfers, not miles. A location three miles away on a direct line is closer, in the only sense that matters, than one a mile away requiring two changes. Target by neighbourhood and by station adjacency, and say the station in the ad copy.

Then handle the overlay. Commuters from New Jersey, Westchester, Long Island and Connecticut are in the city on weekdays and at home at weekends, and visitors are here constantly. For some categories, dentistry and physiotherapy among them, the commuter who works in Midtown is the ideal customer. For others, the visitor click is pure waste. Decide which you are, then set location options to match, because the default behaviour will serve ads to people merely showing interest in New York from elsewhere.

Leak two: bidding on the category word

The broad category term in this city is bid up by national advertisers, aggregators and lead resellers with budgets and lifetime values that are not yours. Competing on it with a local budget hands money to companies who are buying the click to resell it to you.

What works instead is the specific job, the specific problem and the specific neighbourhood. Not the practice area but the situation. Not the service but the symptom, the brand of equipment, the building type, the price question. Those terms cost a fraction of the category word and the person typing them is further along. A New York account with two hundred tightly matched terms almost always outperforms one with twenty broad ones at the same spend.

Leak three: a phone nobody answers

At these click prices, an unanswered call is the most expensive thing in the account. A $90 click that produces a call that rings out has cost you the click and the customer. Before you raise a budget, fix the intake: live answering during every hour the ads run, a human on the first attempt, a callback inside five minutes when one is missed, and call recording so somebody actually listens to what callers ask for. Most New York accounts have more profit available in the intake than in the bidding.

The structure a New York account should have

• Separate campaigns by borough or cluster, each with its own budget, so you can see which part of the city pays.
• Brand terms in their own campaign, so branded performance never flatters the rest of the account.
• A negative keyword list maintained weekly, which at these prices is the highest return hour anybody spends on the account.
• Conversions defined as a qualified enquiry, meaning a call over a set duration or a form a salesperson actioned, not a thank you page view.
• Dayparting that matches your intake. Paying $90 for a click at eleven at night when nobody answers until nine is a choice, and usually the wrong one.
• Your own account, your own card. Create it yourself, grant access, and keep the conversion history that you paid to build.

The arithmetic before you start

Multiply your average sale by your close rate from an enquiry to get the value of one enquiry. Divide a click cost by your landing page conversion rate, and you have your cost per enquiry. If the first number is not comfortably larger than the second, paid search in New York is not your channel yet, and the money belongs in your business profile, your reviews, your organic pages and your intake until it is. That is an honest and often money saving answer, and a good agency will say it on the first call. The broader agency fee and contract picture for this market is laid out in the NYC digital advertising line items guide.

Fee shapes, and the floor below which the work cannot be done

A flat fee suits accounts under about $40,000 a month in media, because the work does not scale smoothly with spend at that level. A percentage suits larger accounts, where management genuinely grows with the budget. A hybrid with a floor and a declining percentage is the fairest structure for an account that is growing. Below roughly $2,000 a month in management fee, nobody senior is touching your account weekly in this city, whatever the proposal says, because the hours do not exist at that price.

PPC in New York: what advertisers ask before the first month of spend

What does PPC management cost in New York?

$2,200 to $6,500 a month as a floor, or 8 to 14 percent of spend on larger accounts, with a separate setup fee of $1,500 to $4,000 for a rebuild.

How much media do I need to run paid search here?

$6,000 a month is a realistic minimum in most local categories. In legal or treatment categories it is several times that, because one click can cost more than a day's budget elsewhere.

Should I target all five boroughs?

Only the ones you can serve and want to serve, each as its own campaign. Running all five in one campaign means the cheapest borough quietly eats the budget.

Is a New York agency worth a premium over a remote one?

Remote management can be excellent. Local knowledge pays for itself when the account depends on neighbourhood behaviour, building types or the commuter pattern, which in this city is more often than not.

Why is my cost per lead so much higher than the national figures I read?

Because you are bidding against the densest set of competitors in the country. Compare yourself against New York benchmarks and against your own trend, and judge the account on profit rather than on cost per lead.

What should the first sixty days produce?

A rebuilt account structure, clean conversion tracking, a negative list in active use, a reduction in wasted spend you can see in the report, and a cost per qualified enquiry you can compare month to month.

What a New York buyer reads before they click an ad

Here is a change worth testing against your own category. An office manager in Midtown or a homeowner in Forest Hills now frequently asks ChatGPT, Gemini or Perplexity which firms in New York do the thing they need, reads the three or four names it gives, and starts from that list rather than from the top of a results page. Those names are assembled from the indexed record: directory data, press and trade coverage, review text and pages that state in plain sentences what a company does and where. A business spending heavily in the auction can be entirely absent from that answer, because advertising does not feed it. Ask an assistant your own category question in a customer's words and read the list it returns. AI Knows Us (https://www.aiknowsus.com) was built to ask that question across the assistants on a schedule and to show which pages each named firm came from.