There is no universal Google Ads budget, but there is a formula: leads needed × target cost per lead = ad budget. In practice in 2026, a test in a local, low-competition niche starts at $1,500–$3,000 a month, competitive metro markets and e-commerce run $3,000–$8,000, and niches with expensive leads (legal, medical, real estate, B2B) start at $10,000. On top of that, add setup (from $500–$1,500 one-off) and management ($500–$2,000 a month or 10–20% of spend).
Most articles answer this question with "it depends on your niche" — and then ask you to book a call. We will take the other route: show you how to derive your own number from your own economics, give honest benchmarks to sanity-check it, and walk through the cases where you should not launch Google Ads at all.
What Google Ads costs are made of
When people ask "what does Google Ads cost per month", three different payments get lumped into one number. Let's separate them — no calculation makes sense otherwise.
| Cost item | Who you pay | How often | 2026 reference |
|---|---|---|---|
| Ad budget | Google — for clicks or conversions | as it spends | from $1,500/mo in local niches to $10,000+ in expensive ones |
| Campaign setup | a specialist or agency | one-off at the start | from $500 with an independent specialist, $1,500–$5,000 with agencies |
| Management | a specialist or agency | monthly | $500–$2,000/mo or 10–20% of ad spend |
Three important clarifications. First: budgets in Google Ads are set per day, not per month. You define an average daily budget per campaign; your monthly maximum equals that figure × 30.4. On any single day Google may spend up to twice the daily budget chasing conversions, but never more than the monthly cap — keep that in mind when your accountant asks why Tuesday cost double.
Second: depending on your country, Google may add taxes or regulatory surcharges to your invoices (VAT, DST and similar). The percentages differ by jurisdiction; check the billing settings for your country before planning an annual budget — on large budgets the difference is thousands of dollars.
Third: the ad budget is the only one of the three items that turns directly into leads. Setup and management are what you pay so the budget does not burn in vain. The smart place to economize is not on them but on the cost per lead itself — through correct campaign structure and a strong website.
The budget formula from unit economics: working back from cost per lead
"The average budget in your niche" is a bad anchor: within one niche, two businesses with different margins and different sales teams can afford budgets that differ five-fold. The right order is from your economics to the budget, not from other people's numbers. Four steps.
Step 1. How many leads you need
Start from the sales plan. Say you sell custom kitchens and want +10 deals a month from Google Ads. Your sales team closes 20% of leads. So you need: 10 ÷ 20% = 50 leads a month.
Step 2. How much you can pay per lead
Average order value $25,000, margin 25% — profit per deal $6,250. You are willing to spend up to 20% of profit on acquisition: $6,250 × 20% = $1,250 per deal. One deal takes five leads, so your target cost per lead (CPL): $1,250 ÷ 5 = $250.
Step 3. Calculate the budget
50 leads × $250 = $12,500 a month of ad budget — about $410 a day. Plus setup and management. That is your number — not "the niche average" but one derived from your profit.
Step 4. Reality-check it against the market
The formula works on paper — now test it against the auction. Convert the target CPL into a target cost per click (CPC): at a 2% site conversion rate the allowed CPC = $250 × 2% = $5. Open Google Keyword Planner and look at the real bid ranges for your keywords in your geography:
- Real CPC at or below $5 — the economics work; you can launch.
- Real CPC at $12 — at your current conversion rate a lead will cost $600 instead of the allowed $250. The gap has to be closed: raise site conversion (from 2% to 5% — and the math works), increase the average order value and margin, factor repeat purchases into LTV — or honestly admit that Search is too expensive for you and test Display, YouTube or Performance Max, where clicks are cheaper.
Site conversion is the most underrated multiplier in this formula. Raising conversion from 2% to 4% cuts your cost per lead in half at the same bids — often cheaper than fighting for clicks for years. If your site is old and does not convert, start there rather than with the budget: see web development.
Test budget: how much you need to find out whether Google Ads works
The first campaign is always a hypothesis. Keyword Planner and our calculations give a range, but only live traffic reveals the real cost per lead. So the first budget is a test budget: its job is not to pay off but to buy reliable data.
What to set aside for a test in 2026:
| Situation | Test budget (click spend) | Test length |
|---|---|---|
| Local services, low-competition niches | $1,500–$3,000/mo | 3–4 weeks |
| Competitive metro markets, e-commerce, mid-ticket services | $3,000–$8,000/mo | 3–4 weeks |
| National campaigns, legal, medical, real estate, B2B | from $10,000/mo | 4–6 weeks |
Where the floor comes from: Smart Bidding needs roughly 30–50 conversions a month per campaign to exit the learning phase and stabilize the cost per lead. If your expected CPL is $50, the minimum for one campaign is 30 × $50 = $1,500 a month. Budget less than that and the strategy stays permanently "learning", leads arrive in expensive bursts, and you draw the false conclusion that "Google Ads doesn't work". This is the single most common reason small-budget tests fail.
The test's success criterion is not "did we like the leads" but numbers: actual CPL versus the allowed CPL from step 2 of the formula. If it matches, or there is a clear plan to close the gap (negative keywords, cutting weak placements, landing page fixes) — scale. If the gap is several-fold with no ideas — stop, having lost a test budget instead of a year's budget.
CPL benchmarks by industry
For a first sanity check — cost-per-lead ranges by industry type. An honest caveat up front: these are US-market reference points from published agency benchmark studies for 2025–2026 plus our own campaigns, not a guarantee. In a specific account the number depends on geography, website, offer and season and can differ from the range several-fold — in either direction. Use the table to sanity-check your formula result, then replace it with your own numbers after the first test.
| Industry type | Examples | CPL reference | Comment |
|---|---|---|---|
| Local services | appliance repair, cleaning, auto service | $25–$80 | clicks from $3–$8, moderate competition |
| E-commerce, mid ticket | apparel, home goods, pet supplies | $20–$60 per order | plan through ROAS rather than CPL; Shopping and PMax are cheaper than Search |
| Home improvement in metros | remodeling, custom furniture, HVAC | $60–$150 | Search is overheated; Local Services Ads and PMax often deliver cheaper leads |
| Healthcare | dental, clinics, aesthetics | $80–$250 | per new patient; strict policy review of medical topics |
| Legal | bankruptcy, personal injury, immigration | $100–$300 | Search clicks reach $50–$100+, but one client pays for the month |
| B2B and industrial | equipment, wholesale materials, IT services | $100–$300+ | a qualified lead in industrial equipment runs around $150–$250 |
| Real estate and automotive | new developments, brokerages, dealers | $100–$400+ | among the most expensive auctions; budgets from $10,000/mo |
One trend worth planning for: across published agency benchmarks, the cost per conversion in Google Ads has been rising roughly 10–15% a year with the same campaign structure — more advertisers, more automation, hotter auctions. When you set an annual budget, add that buffer to today's CPL.
What Google Ads setup and management cost
The second and third cost items are specialists' work. The 2026 ranges look like this.
Setup (one-off). Independent specialist — from $500; agencies — $1,500–$5,000 depending on scope: number of campaigns, keyword research depth, conversion tracking and GA4 goals, call tracking, landing page alignment. "We'll set it up for $100" offers are, as a rule, an automated Smart-campaign launch with no research: ads will technically run, but they will test your economics with your money.
Management (monthly). $500–$2,000 a month or 10–20% of ad spend — on larger budgets the percentage model usually favors the client. Management must include: bid and strategy work, search term and placement hygiene, negative keywords, A/B testing of ads and assets, and a report that goes all the way to leads and their cost — not "impressions grew". If you cannot read your CPL out of the report, you are paying for a retelling of the Google Ads interface.
2026 realities: PMax, Smart Bidding and auction inflation — what changed in budget planning
If you last budgeted for Google Ads a few years ago, part of the logic is outdated. Four shifts that directly affect the numbers.
- Performance Max is the default cross-channel format. PMax runs your ads across Search, Shopping, Display, YouTube, Gmail and Maps from a single campaign. The budget is set per day and steered by a conversion goal — Target CPA or Target ROAS. It is convenient and it scales, but control is limited: negative keywords and placement reporting are far narrower than in classic Search. The practical rule: keep a classic Search campaign on your money keywords and let PMax expand around it.
- Smart Bidding decides — and it needs data. Manual bidding has moved to the margins: Maximize Conversions, Target CPA and Target ROAS drive most spend. A strategy needs 1–2 weeks and roughly 30–50 conversions a month to stabilize. The budget consequence: one campaign with a sufficient budget beats five campaigns "on a drip" — a smeared budget teaches none of them. And do not touch the campaign for the first two weeks: every restart is learning paid for twice. Budget changes bigger than ~20–30% also reset learning — scale in steps.
- Conversion tracking is the budget's foundation. Smart Bidding optimizes toward whatever you count as a conversion. Broken or double-counted tracking means the algorithm literally optimizes toward garbage. Before scaling a budget: conversion actions audited, enhanced conversions enabled, offline deal imports from the CRM connected where sales close offline.
- Auctions keep getting more expensive. CPCs grow roughly 10–15% a year across most industries. If your economics only work at today's bids with no margin, next year the auction will eat the difference — build the buffer into the plan from the start.
When Google Ads will not work out: the honest list
This section costs us some leads, but it will save you thousands. There are situations where paid search will not pay off no matter who sets it up.
- Low ticket + expensive clicks + no repeat purchases. If profit per deal is $10 and a click in your niche costs $4 at a 2% site conversion, a lead costs $200 — the formula does not work at any bid. Google Ads in such niches only works through LTV: when the customer returns and buys again, acquisition pays off on the second or third purchase. No repeat purchases — no Google Ads.
- No search demand. A new product nobody searches for cannot be sold through Search: there is nothing to show against, and Display or PMax without formed demand delivers cold traffic. What works here is paid social and content — create the demand first, then harvest it with search ads.
- The site does not convert, or nobody handles the leads. At a 0.5% site conversion rate, every bid becomes unprofitable: you pay for clicks that die on the landing page. Same with a sales team that calls back a day later: up to 70% of leads go cold. Fix the funnel first, then buy traffic.
- A budget below the learning threshold. $500 a month in a competitive metro niche is not "a small but honest test" — it is a guaranteed write-off: Smart Bidding will not learn and the reach will not even produce statistics. If a proper test budget is out of reach, it is more honest to save up or pick a cheaper channel.
- You need leads cheaper than the auction sells them — permanently. If the economics only work at promotional bids, every year the auction will eat your margin (those same 10–15% CPC increases). Over the long run such niches migrate to organic: SEO delivers leads cheaper than paid search once traffic reaches plan — we broke down that calculation in the article on SEO budgets. The working combo for most businesses: Google Ads covers "now", SEO lowers the average cost per lead over time.
In every other case Google Ads is the fastest way to test demand and get leads in the first week. The key is to enter it with calculated economics — not with "let's try five hundred bucks and see".