How to Set an Ad Budget for a New SaaS (Working Backwards From Cost per Trial)
To set an ad budget for a new SaaS, decide your qualified conversion, estimate what a paying customer is worth, work backwards to the most you can afford to pay per trial or demo, size a fixed test that buys enough conversions to read, set daily budgets in the ad platform, and agree in advance what result means stop, fix or scale. Infinall AI prepares the research, copy and creative for review; it does not place media or manage your spend.
A new SaaS rarely has the data to know what ads will cost, so the useful question is not "how much should I spend" but "how much can I afford to lose to learn whether this channel works." That depends on what a customer is worth to you and how many conversions a test needs to produce a readable answer. This guide gives a method rather than a benchmark number: any figure quoted without your pricing, trial-to-paid rate and channel is a guess, and this guide does not invent one. Platform minimums and learning behaviour change, so confirm current rules in each ad manager. Infinall AI does not buy media, set budgets in your account or guarantee results; it prepares campaign research and creative so the budget you do spend goes into reviewed ads.
How to Set an Ad Budget for a New SaaS (Working Backwards From Cost per Trial)
Define the conversion you are paying for
Pick one qualified conversion, such as a free trial that reaches a key activation step, a demo request from a work email, or a paid signup, and make sure it is tracked with the ad platform's pixel, tag or server-side events. Budget math only works if the conversion is countable and meaningful: a click or a view is not a unit you can price against revenue. If your trial-to-paid rate is unknown, treat your first test as a way to measure it rather than assume one.
Estimate what a customer is worth and what you can pay for one
Take your monthly price, an honest estimate of how long customers stay, and your gross margin to get a rough lifetime gross profit per customer; for a brand-new product use a conservative guess and mark it as one. Divide by the number of trials you expect per paying customer (your trial-to-paid rate, or a low estimate until you have data) to find the most you could pay per trial and still break even over a payback period you can tolerate. Cash-constrained startups often require a short payback period, which lowers the allowable cost per trial.
Size the test by the number of conversions you need to read
Decide how many conversions the test must produce before you will trust the result, and multiply by a realistic cost per conversion. Because you do not yet know that cost, estimate it from the platform's own budget and reach tools, search keyword planners, or a small pilot, and treat it as a range [NEEDS VERIFICATION: category cost-per-trial range for the chosen platform]. The test budget is then the number of conversions you need times the cost you expect, plus a margin for being wrong. If that number is more than you can afford to lose, narrow the audience, pick a cheaper channel or test a smaller step such as a lead magnet.
Convert the total into daily budgets in the ad manager
Divide the test budget by the number of days you will run, typically at least one to two weeks so the platform can stabilise and you avoid reading day-to-day noise. Enter it as a daily or lifetime budget at the campaign or ad set level in Meta Ads Manager, Google Ads or LinkedIn Campaign Manager, and note any platform minimums, which vary by platform and bidding option. Keep the number of ad sets and ads small so each receives enough delivery; splitting a small budget across many variations leaves each too thin to learn from.
Put creative inside the budget, not outside it
Account for the cost of making ads as well as showing them: copy, images and video take time or money, and a test with only one creative cannot separate a weak offer from a weak ad. Plan a small set of distinct angles rather than many near-duplicates. Infinall AI can prepare the research, copy, image ads, carousels and video drafts from your product URL for review before export; the Free plan is $0 with 1,000 credits/month and no video generation, and Pro is "$30/month" with video generation. Anything it produces lands in an Approval Queue for you to edit; you still launch and fund the campaign in the platform.
Write the stop, fix and scale rules before you launch
Before spending, write down thresholds: if cost per qualified trial is under your affordable ceiling after the planned conversions, increase budget in steps rather than doubling overnight; if it is above the ceiling, check tracking, landing page conversion and offer before blaming the channel; if there are no conversions after the planned spend, stop and reassess rather than topping up. Review weekly, not hourly. Put a hard cap on total test spend and a date to decide, so a test cannot quietly become an open-ended expense.
Frequently asked questions
How much should a new SaaS spend on ads?
There is no honest universal figure, because the right amount depends on what a customer is worth, your trial-to-paid rate, the channel's cost per conversion and how much you can afford to lose while learning. The method in this guide sizes a bounded test from those inputs. If the arithmetic says an affordable test would produce too few conversions to read, that is a signal to pick a cheaper channel or a smaller conversion, not to raise the budget.
Does Infinall AI set or manage my ad budget?
No. Infinall AI prepares campaign research, strategy, ad copy and creative for your review, and a launch guide; you set budgets, publish campaigns and manage spending in the ad platform. Infinall does not buy media, manage live ad spend or guarantee any cost per conversion. Free is $0 with 1,000 credits/month; Pro is $30/month with 3,000 credits/month, and 1 credit equals $0.01 of AI compute, which is separate from your ad spend.
How long should I run a budget test before judging it?
Long enough to buy the number of conversions you decided you need and to pass the platform's initial learning period, which for many tests means at least one to two weeks. Judging after two days mostly measures noise. Write the review date and the stop and scale thresholds before launch so the decision is not shaped by how the first few days happened to feel.
Prepare reviewed ad creative for your test budget with Infinall AI
1,000 free credits. No card required.