“How much should we spend on Google Ads?” is the wrong first question. The right one is “how much does it cost to find out whether Google Ads works for us?” This article gives you a calculator you can run in a spreadsheet in ten minutes, the benchmarks to plug into it, and a way to set a budget that reaches a decision instead of just spending money.
Why 50 leads is the target
Fifty is not magic, but it is the smallest number that lets you draw conclusions. With 50 leads you can see which keywords produce them, whether your landing page converts at a stable rate, and roughly what share of leads sales will accept. With 10 leads, one lucky week or one bad campaign dominates everything. With 200 you have spent more than you needed to before deciding. Fifty is the point where the data starts telling you something and the bill is still survivable.
If you already know your lead-to-customer rate, you can also translate 50 leads into expected revenue and decide in advance what result would make you continue.
The formula
Budget is a chain of four numbers:
Budget = Target leads ÷ Landing page conversion rate × Cost per click
Written out for the example of 50 leads, a 5 percent conversion rate and a $6 click:
50 ÷ 0.05 = 1,000 clicks needed. 1,000 × $6 = $6,000.
Add a buffer for the learning period, when Google’s bidding is still calibrating and costs run high, and you get a test budget. The buffer is usually 20 to 30 percent for a new account. So the realistic number in this example is $7,200 to $7,800 spread over four to eight weeks.
The spreadsheet version
| Cell | Input | Example | Where to get it |
|---|---|---|---|
| B2 | Target leads | 50 | Your decision |
| B3 | Landing page conversion rate | 5% | Your data, or the benchmark table below |
| B4 | Average cost per click | $6.00 | Google Keyword Planner top-of-page bid, low range |
| B5 | Learning buffer | 25% | 20 to 30 percent for new accounts, 10 percent for existing |
| B6 | Test length in weeks | 6 | Long enough for two bidding cycles |
| B8 | Clicks needed | =B2/B3 | 1,000 |
| B9 | Media cost | =B8*B4 | $6,000 |
| B10 | Test budget | =B9*(1+B5) | $7,500 |
| B11 | Daily budget | =B10/(B6*7) | $179 |
| B12 | Cost per lead at target | =B10/B2 | $150 |
Cell B12 is the number to sanity-check against your economics before you spend anything. If a lead is worth $80 to you and the calculator says $150, no amount of optimization inside Google Ads will rescue the plan. Change the offer, the landing page, or the channel.
Benchmarks for the inputs
Use your own numbers when you have them. When you do not, these ranges reflect what B2B and professional services advertisers typically see in 2026. Treat the midpoint as a planning figure and the low end as a stretch goal.
| Industry | Typical CPC (search) | Landing page conversion to lead | Resulting cost per lead |
|---|---|---|---|
| B2B software, mid-market | $5 to $12 | 3% to 6% | $120 to $300 |
| Marketing and creative agencies | $4 to $9 | 4% to 8% | $70 to $180 |
| Legal services | $15 to $60 | 5% to 10% | $200 to $700 |
| Financial and accounting services | $8 to $25 | 4% to 8% | $130 to $400 |
| Home services and local trades | $5 to $20 | 8% to 15% | $50 to $150 |
| E-learning and courses | $2 to $6 | 5% to 12% | $25 to $90 |
| Consulting and coaching | $4 to $10 | 3% to 7% | $80 to $250 |
Two cautions. First, the conversion rate is for a dedicated landing page with one form and one offer. Sending ads to your homepage typically halves it. Second, CPC varies more by keyword than by industry: “crm software” and “crm software for real estate agents” can differ by a factor of three. Pull actual bid estimates from Keyword Planner for your specific list rather than relying on the table.
How to set the daily budget
Google spends up to twice your daily budget on any given day and evens it out across the month, so daily budget is really a monthly cap divided by 30.4. The calculator gives you a test budget and a test length, and the daily figure follows. Three practical rules:
- Do not go below about 10 clicks a day. Under that, Smart Bidding never leaves the learning phase and results are noise. With a $6 CPC that means a $60 daily floor.
- Do not change the budget more than 20 percent at a time. Large jumps reset learning. If you need to scale, step up weekly.
- Start with one campaign. Splitting a $180 daily budget across five campaigns gives each one too little data to optimize. Concentrate, learn, then split.
The two-phase test plan
Phase 1: Discovery (weeks 1 to 2)
Goal: find out what real clicks cost and which search terms actually trigger your ads. Run a single search campaign with tightly themed ad groups, Maximize Clicks bidding with a CPC cap at your planning figure, and all conversions tracked. Do not judge lead cost yet. Spend the time reading the search terms report daily and adding negatives. It is common to cut 30 to 40 percent of wasted spend in the first two weeks this way.
Phase 2: Optimization (weeks 3 to 6)
Goal: reach 50 leads at or below the cost per lead in cell B12. Once you have about 15 conversions, switch to Maximize Conversions, then set a target CPA once you pass 30. Pause keywords with 100 or more clicks and zero leads. Test a second landing page variant if conversion rate is below the benchmark. Keep tagging clean so you can compare Google against other channels in GA4; if your team does not have a naming standard yet, our UTM naming convention takes an hour to set up, and the Google Ads specific setup is covered in how to add UTM parameters to Google Ads.
Deciding at the end
Before the test starts, write down three outcomes:
| Outcome | Condition | Decision |
|---|---|---|
| Clear win | 50 leads reached, cost per lead at or below target, sales accepts at least a third | Scale budget 20 percent a week, add campaigns |
| Inconclusive | Cost per lead within 30 percent of target, or conversion rate improving week over week | Extend four weeks with a fixed change list, no new budget |
| Clear loss | Fewer than 25 leads, or cost per lead more than double the target after week four | Stop. Fix the offer or landing page before re-testing, or move the budget to another channel |
The “sales accepts a third” condition matters. Fifty leads that sales rejects are not a win. Track qualified leads, not just leads, using the method in CPL vs CPQL, and put the accepted rate into the decision from the start.
Three mistakes that inflate the number
- Broad match on day one. Google will happily spend your discovery budget on loosely related searches. Start with phrase and exact match and widen only after you understand the search terms.
- Counting the wrong conversion. If a “conversion” is a page view or a button click rather than a submitted form, the calculator’s conversion rate is fiction. Import CRM-confirmed leads if you can.
- Judging at week two. The first two weeks are the most expensive weeks you will ever run. Decide on the full test window, not the opening.
The short version
Divide the leads you want by your landing page conversion rate to get clicks, multiply by the cost per click, add a quarter for learning, and check that the resulting cost per lead is a number your business can live with. Run it as a two-phase test with the decision rules written down in advance. Most first campaigns fail not because Google Ads does not work but because nobody defined what “working” would look like before the money went out.