# What $200 actually buys you on Google Ads > At the 2026 average of $5.42 a click, a $200 Google Ads budget buys 36 clicks and roughly three conversions. The arithmetic the case studies never show you. Published 2026-07-28 · 9 min read · google-ads-budget, advertising, ppc, cost Canonical: https://bidsurvivor.space/blog/what-200-dollars-buys-you-on-google-ads --- A founder I know put a $200 Google Ads budget into one search campaign, watched it disappear over nine days, and got 38 clicks. Two of them filled in the form. Neither replied to the follow-up email. His conclusion was that he'd done it wrong. He hadn't. He'd done it exactly right, and $200 is simply not very much money in that auction. The arithmetic was always going to end there, and nobody had shown it to him before he spent it. So here it is. ## The number everything else follows from WordStream publishes an annual benchmark study built from real accounts — the 2026 edition covers **more than 13,000 search campaigns across 23 industries**, running from April 2025 to March 2026. The headline: | Metric | All-industry average | | --- | --- | | Cost per click | **$5.42** | | Conversion rate | 8.18% | | Cheapest vertical (Arts & Entertainment) | $1.63 | | Most expensive vertical (Legal) | $9.87 | That $5.42 is up about 3% on the prior year's $5.26. It has gone in one direction for as long as anyone has been measuring. Now do the division. **$200 ÷ $5.42 = 36 clicks.** Thirty-six. Not thirty-six thousand impressions, not thirty-six hundred sessions. Thirty-six human beings arriving on your site, before you have written a headline, tested a landing page, or excluded a single irrelevant search term. Apply the 8.18% average conversion rate and you get **2.9 conversions**. Round down, because you cannot have three tenths of a signup: two, maybe three. If you're a legal tech startup bidding in a $9.87 vertical, $200 is **20 clicks**, and the expected conversions round to one. ## The part the case studies leave out Every ads case study you have ever read has the same shape. Spend, then revenue, then a multiple. What it never shows is the first $200, because the first $200 is always terrible. Learning periods, no negative keywords, no conversion history for the algorithm to optimise against. You are paying market rate for clicks while the system works out what you are. The case study starts at month four. You are at day one. There is a second thing missing. Juniper Research put digital ad fraud losses at **$84 billion in 2023**, forecasting **$100.2 billion for 2026** and $172 billion by 2028. Independent measurement puts invalid traffic at roughly **18–20% of programmatic ad traffic**. Search is cleaner than programmatic display — meaningfully cleaner — but "cleaner than the worst channel on the internet" is not the same as clean. Take a conservative 10% off your 36 clicks for invalid traffic and you're at 32. ## What changed on 1 September Since this was first written, Google has moved the default underneath small accounts. From 1 September 2026, Search campaigns using broad match were migrated automatically to AI Max, which hands query expansion, creative variation and landing-page choice to the system rather than to the settings you typed ([Search Engine Land](https://searchengineland.com/google-sets-ai-max-migration-timeline-for-search-campaigns-485006)). For an account spending five figures a month, that is a productivity feature. For a $200 account it is the opposite: the learning period now covers not just your bids but the system's guesses about which searches you meant, and a budget that buys 36 clicks gives it almost nothing to learn from. The one-keyword test I describe below still works. You now have to open the campaign settings and confirm you are actually running it, rather than a broadened version of it. What carried over in the migration and what control did not is in [what the AI Max migration cost small advertisers](/blog/google-ai-max-migration-what-you-lost). ## What I am not going to claim I run an advertising board. You should read the next part with that in mind, so let me be direct about where Google wins, because it wins on the thing that matters most. **Google has intent.** Somebody typing "crm for plumbers" wants a CRM for plumbers. That is the single most valuable signal in all of advertising, and no display network, newsletter, billboard or ad board has anything like it. If your $200 finds four people at the exact moment they were going to buy, $5.42 a click is cheap. **Google has scale and targeting.** You can reach a specific city, on a specific device, at a specific hour, and keep spending until you run out of money. Nothing here matches that. If your product solves a problem people actively search for, and you can afford to spend past the learning period — genuinely past it, several thousand dollars, not two hundred — search advertising is usually the right answer. I would tell you that even if it cost me a customer. The problem is narrower than "Google Ads is bad". It is this: **$200 is too small a number to learn anything from.** It buys 36 clicks, which is not a sample. It is an anecdote you paid for. ## The comparison I'd actually make The honest question isn't "Google or something else". It's: *what is the most useful thing $200 can do when $200 is all there is?* Three real options. **Spend it on search anyway, but on one keyword.** Not a campaign. One long-tail phrase with obvious buying intent, exact match, one landing page. Thirty-six clicks split across twelve keywords tells you nothing. Thirty-six clicks on one phrase tells you whether that phrase converts. It is a small experiment, but it is an experiment. **Buy attention where you can see it before you pay.** This is what I built, so weigh it accordingly: [BidSurvivor](/) auctions advertising twelve hours at a time — two slots a day — and every brand's click-through count is public on the board *before* you bid. You are not buying an estimate. You can look at what a slot sent the last brand that held it, decide it isn't worth it, and keep your money. The first brand into an empty slot pays nothing at all, and every visitor starts with $100 of house credit to bid with. Where it's plainly worse than search: no intent, no targeting, and an audience whose actual size is printed on the front page rather than measured in billions. **Don't spend it on ads at all.** With $200 and no distribution, writing something worth linking to usually beats renting attention. It is slower and less certain, and it compounds, which paid traffic never does. If you have more time than money, this is probably the right answer and I'd rather say so than sell you a slot. There are cheaper paid channels than search, and some of them suit a $200 budget better; the trade in each case is that you give up the intent. I went through them one by one in [Google Ads alternatives for small budgets](/blog/google-ads-alternatives-for-small-budgets). ## If you do spend it on search: the one-question plan The failure mode is not spending $200 on Google. It is spending $200 and learning nothing. Here is the version that at least produces an answer. 1. **Pick one phrase.** Something a buyer types, not a browser: "invoice software for electricians", not "invoicing". Exact match only. If you cannot name a phrase like that, you are not ready for search, and that itself is worth knowing for free. 2. **One ad, one landing page.** The page should answer the phrase in its first line. Do not send the click to your homepage; the difference is the subject of [homepage vs landing page](/blog/homepage-vs-landing-page), and at 36 clicks you cannot afford to lose half of them to navigation. 3. **Cap the daily budget at $20.** Ten days of data beats two days of data, and a campaign that spends everything in an afternoon has told you about one afternoon. 4. **Count form fills, not clicks.** Clicks are what you paid for; they are not what you wanted. Write down before you start what number of conversions would make you buy more. Two? Five? If you do not decide in advance, you will decide afterwards, and afterwards you will be arguing with yourself. 5. **Stop at $200.** Do not top it up "to give it a chance". The whole point was to find out what the phrase does at market price, and you now know. At the end you have one of three results: the phrase converts (buy more of it, carefully), the phrase does not convert (change the page before you change the channel), or the phrase barely got any impressions (there is no search demand for the words you thought there was, which is the cheapest lesson on this list). ## Putting the result in a unit you can compare The reason $200 disappears without teaching anyone anything is that every channel prices in a different unit. Google sells clicks, display sells thousands of impressions, newsletters sell sends, our board sells hours. Before you compare the next quote to the $200 you just spent, convert both to the same thing: cost per visitor who arrived, and cost per conversion if you had any. The conversion is a couple of lines of arithmetic and it is written out in [CPM vs CPC vs flat fee](/blog/cpm-vs-cpc-vs-flat-fee). Once you have one number per channel, the decision usually makes itself. ## The arithmetic, one more time - $200 at the all-industry average is **36 clicks** - At the average conversion rate, that is **2.9 conversions** - In an expensive vertical, it's 20 clicks and about **1 conversion** - Take off invalid traffic and you have roughly **32 real visitors** None of that is an argument against advertising. It's an argument against expecting $200 to answer a question that costs $2,000 to answer. Decide what you want that money to teach you, then pick the channel that can teach it. Sometimes that's Google. Sometimes it's twelve hours on a board where you can watch the clicks land. --- **Sources.** Google Ads benchmarks: [WordStream, Google Ads Benchmarks 2026](https://www.wordstream.com/blog/2026-google-ads-benchmarks). AI Max migration timeline: [Search Engine Land](https://searchengineland.com/google-sets-ai-max-migration-timeline-for-search-campaigns-485006). Ad fraud losses: [Juniper Research](https://www.juniperresearch.com/press/digital-advertising-spend-lost-to-fraud-68-billion/). Invalid traffic share: [Fraud Blocker ad fraud data](https://fraudblocker.com/data/ad-fraud-data-facts). If it helps, the [ad budget calculator](/tools/ad-calculator) runs this arithmetic for any budget, and tells you whether the budget is big enough to trust the answer. It is free and needs no account.