How to market a SaaS with no audience and no budget

Zero followers. Zero euros. A product that works and nobody who knows it exists. That is the most common starting position there is, and almost every marketing guide written for it assumes a list, a logo or a card on file. This one assumes none of them. Here is what happens on day one, and what happens for the twenty-nine days after it.

What no budget actually removes

There are four ways to get a customer, and there is no fifth. You reach out to people who already know you. You post something in public. You reach out to strangers. You pay to get in front of people. Alex Hormozi named these four the Core 4 in his book $100M Leads, and the name stuck because it is short. If the framing is new to you, the Core 4 explained is the fifteen-minute version.

No budget closes exactly one of the four. That is the whole of the damage. Paid goes off the table until something else is producing revenue, which is a smaller loss than it feels like, because paid ads cost a solo founder around €150 per customer before they know whether those customers stay. Three lanes remain and they are all free at the door.

Free does not mean cheap. It means the price is denominated in mornings. So rank the three by hours per customer, not by how modern they feel, and the order comes out the same every time.

Lane	One unit of work	Minutes per unit	Units per customer	Hours per customer	When it pays out
Warm outreach	One personal message to someone who already knows your name	4 minutes	10 asks	1 to 2 hours	Days. Some people reply the same morning
Cold outreach	One hand-picked name plus one three-line email	3 to send, 2 to find	160 sends	13 to 15 hours	Weeks three to eight, once follow-ups land
Posting and community answers	One post or one long answer, plus twenty minutes of replies	45 minutes	20 to 40 posts	15 to 30 hours	Months two to four, and then it keeps going

Fifteen hours against one. That gap is why the order matters more than the tactics, and it is the single most expensive thing a founder gets wrong in week one. If you want a second opinion on the ranking for your specific product, the marketing channel quiz asks seven questions and gives you a split, and the channel comparison covers the cases where this order is wrong.

Why warm outreach goes first, even though everyone skips it

Everyone skips it for the same reason: messaging people who know you feels like spending social capital, and cold email feels anonymous and therefore safe. That instinct is the expensive one. Here is the arithmetic that settles it, run at volumes a founder can actually do in a week.

• 30 warm asks at 10% per ask gives you about 3 customers. Building the list of 200 takes 45 minutes and the messages take four minutes each, so the whole thing costs roughly two and a half hours.
• 300 cold sends at 0.6% per send gives you about 1.8 customers. Hand-building 300 good names costs about eleven hours at forty names per ninety minutes, and sending costs another fifteen, so call it twenty-six hours.
• Ten times the work for two thirds of the result. And the warm replies arrive within 48 hours while the cold ones need two follow-ups spread over a fortnight before you can read them.
• The catch is that warm runs out. Roughly 300 asks and your list is spent, which caps the lane at about 24 customers. That ceiling is the reason cold and posting start in weeks three and two rather than never.

Conversion rates here are our own working figures for a founder with no brand and no case study, used consistently across this site so the sums stay comparable. They are arithmetic inputs, not findings. Swap in your own the moment you have fifty sends of your own data.

The second objection is that your network is not your market. Often true, and mostly irrelevant in month one. Your list of 200 is not a customer list, it is a referral surface. A friend who does not need the product still knows the person who does, and a named ask to somebody who likes you converts far better than any first line you could write to a stranger. The messages themselves are the part people get wrong, so use the warm outreach templates rather than writing them from scratch at seven in the morning. The first one asks for nothing at all, which is the whole trick.

The 30-day sequence

One lane per week at full intensity, adding the next one once the first is running. Splitting a ninety-minute morning across three lanes gives you three half-efforts, none at a volume that tells you anything, and after a month you know nothing about any of them. The volumes in the third column are the definition of a finished week. A week where you did some of it is a week that produced no data.

Week	Lane	What you do	Done means	Hours	By Friday
Week 1	Warm only	Build the list of 200 names. Send 30 messages that ask for nothing. Install analytics on the site while you wait for replies.	200 names written down, 30 messages sent, analytics live	6 to 7	12 to 18 replies, 4 to 6 real conversations
Week 2	Warm plus first posts	Send the specific fifteen-minute ask to everyone who replied. Join three places where your buyers answer each other. Publish three posts about what happened in week one.	15 second asks, 3 posts, 15 community replies	6 to 7	4 to 8 people inside the product, possibly your first customer
Week 3	Community plus list building	Answer 20 questions properly with no link in any of them. Build a cold list of 40 companies using one hard filter. Write the cold email and both follow-ups.	40 names with a reason attached, 20 answers, 3 posts	7 to 8	A list you would defend and a message you have not sent yet
Week 4	Cold plus referrals	Send 50 cold emails across two sittings, changing nothing in the middle. Diary the two follow-ups. Ask your first customers for one named referral each.	50 sends, 2 follow-ups scheduled, 3 named referral asks, 3 posts	7 to 8	A reply rate on 50 sends, which is the smallest reading worth having
Month	Three lanes	Warm, then community, then cold, with posts running from week two	30 warm asks, 50 cold sends, 9 posts, 35 answers	26 to 30	One to four customers and a reply rate

Twenty-six hours across a month is about ninety minutes a day, five days a week. That is the real cost of this plan and it is less than one bad week spent building a feature nobody asked for. If ninety minutes is not available, the daily routine compresses the same four weeks into a shorter block, and the 100 in 100 calculator will tell you whether the target you have in mind survives the division.

Two things in that table are easy to skip and expensive to skip. The first is week three, where you build a cold list and write the message and send nothing. Founders hate that week because it produces no visible output, and it is the week that decides whether week four produces a 2% reply rate or an 11% one. The second is the posting column, which starts small on purpose. Three posts a week from week two, drawn from what actually happened in the product, and the supply problem is handled in what to post when you have no audience. That page is the one to open on the Tuesday morning when the box is blank.

The free-tool inventory, with the limits

Everything below has a free tier a founder can genuinely run a month on. The limits are the interesting column, because a free tier with a ceiling you hit in week two is a paid tool with a delay. Two of these got smaller in the last twelve months, which is the reason the numbers are here at all rather than a list of names.

Job	For	Tool	Free tier, August 2026	What that means for you
Sending	The first 30 warm messages	Gmail, free account	500 recipients a day, counted across To, Cc and Bcc together	Enough for a year of hand-sent outreach at ten a morning. This is your whole sending stack in month one.
Sending	An update to a list of readers	EmailOctopus, Starter	2,500 subscribers and 10,000 emails a month, with EmailOctopus branding in the footer	The largest free sending tier here by a distance, and the one to pick if you only pick one.
Sending	The same job, smaller ceiling	MailerLite, free plan	250 subscribers and 2,500 emails a month since 1 July 2026, and the logo cannot be removed	Sending locks the moment you cross 250 subscribers. This tier held 1,000 a year ago, which is the whole warning.
Sending	Transactional and drip sends	Brevo, free plan	300 emails a day with unlimited contacts, and a Brevo logo on every email until you upgrade	A daily cap rather than a monthly one, so it suits a slow trickle better than one big send.
Finding	An address for a name you already have	Hunter, free plan	50 credits a month. Finding an address costs one credit and verifying it costs another	About 25 verified addresses a month. Spend them on the names you would be annoyed to get wrong.
Finding	Companies and the people inside them	Apollo, free plan	100 email credits a month if you sign up with a Gmail address, 10,000 on a company domain	Sign up with your own domain. The gap between those two numbers is a hundredfold and nobody tells you at signup.
Finding	People by job title	LinkedIn, free account	A commercial use limit that LinkedIn deliberately does not publish a number for. It resets at midnight PST on the 1st of each month	You hit it after a few hundred people searches, then every query returns three results until the 1st.
Publishing	A blog	Bear Blog	Free on a yourname.bearblog.dev subdomain. Custom domain, media uploads and analytics need the paid plan at 5 dollars a month	The subdomain is fine for a year. Almost nobody arrives by typing your domain in.
Publishing	A newsletter	Substack	Free to publish, with no subscriber cap and no send cap. 10% of revenue, only if you switch on paid subscriptions	Costs nothing at all until somebody pays you, which makes it the cheapest place to start writing.
Tracking	Whether anyone read it	Cloudflare Web Analytics	Free for any site via a JavaScript snippet, whether or not the site sits behind Cloudflare. No cookies, so no consent banner	Five minutes to install. Do it in week one so week four has something to look at.
Tracking	What people do inside the product	PostHog, free tier	1 million events, 5,000 session replays, 1.5k survey responses and 1 million feature flag requests a month	You will not go near a million events in year one, so treat this as free rather than freemium.
Tracking	Letting a warm reply book time	Calendly, free plan	One active event type, one connected calendar, unlimited bookings on it	One live link is all a founder needs. Make it fifteen minutes so the ask stays small.
Tracking	The list, the sends and every sum on this page	Google Sheets	Free with a Google account, 10 million cells per spreadsheet	This is your CRM until customer 100. Do not buy one before then.

Notice what is not on that list. No cold-email sending platform, no scraper, no CRM, no scheduling tool for posts. At 50 sends a month you send from Gmail by hand, which lands better anyway, and the templates in the cold email guide for SaaS founders assume exactly that. Tooling becomes worth paying for somewhere around 200 sends a week, and you will not be there this month.

Where you can post for free, and what gets you removed

Community answering is the highest-yield free thing available to a founder at zero, and it is also the fastest way to get banned from the exact place you needed. The rules are not hidden. They are written down, they differ per community, and reading them takes four minutes.

The pattern across all six is the same. Every community with a moderator wants participation before promotion, and every one without a moderator has no readers waiting for you. Pick two moderated places where your buyers already answer each other, spend a fortnight being useful in them with no link in anything you write, and the third week is when you get to mention what you build. That order is slower than it sounds and faster than getting removed.

What "no audience" actually blocks

Founders treat zero followers as a general disqualification and it is a specific one. It closes a short list of doors and leaves most of the building open, and knowing which is which saves you from spending month one trying to fix the wrong problem.

Read the right-hand column and the plan writes itself. Three of the four ways of getting customers work identically at zero followers and at ten thousand, because a message to one person does not query your profile first. The handicap is real in exactly one lane, and that lane was always going to be the slow one. There is a longer argument about which of the two to weight in cold outreach versus content, and the honest answer for month one is both, in that order, at small volumes.

What thirty days honestly produces

Run the numbers on the sequence above and you get a small answer. Thirty warm asks at 10% is three customers. Fifty cold sends at 0.6% is 0.3 of one, so probably none, though the reply rate is the point of that week rather than the revenue. Nine posts is almost certainly zero customers, because posting reports in months two to four and not before.

So: one to four paying customers, with three in the middle of the band. Somewhere between 30 and 60 people who now know what you are building and have your name attached to a specific thing. A reply rate on 50 cold sends that tells you whether the list or the offer is wrong. Nine posts in an archive that is worth nothing today and worth something in October. That is the whole yield, and it is worth saying plainly because the alternative is a founder reading a normal month as a failure.

The failure mode is not the volumes. It is week five. Warm has paid out, cold has produced a number you do not love, posting has produced nothing at all, and the obvious conclusion is that two of the three do not work. Both of those lanes are reporting on schedule. Quitting in week five is common enough that we wrote the post-mortem on why founders quit marketing, and the pattern in it is always the same: the founder judged a slow lane on a fast lane's timetable.

After thirty days the plan stops being about the constraint and starts being about volume. The same three lanes, weighted by what worked, run for another four to nine months to reach a hundred customers. Every play for that stretch is in the first 100 customers playbook, and none of them need a budget either. The budget only starts to matter once you know what a customer is worth, and you will not know that in August.

Questions founders actually ask

How do I market a SaaS with no audience and no budget?

Work the three free ways of getting customers in a fixed order, ranked by the hours they cost rather than the euros. Warm outreach first, because a message to someone who already knows you converts at roughly 10% and costs about four minutes. Community answering second, because a good answer where your buyers already gather is read by strangers for years. Cold outreach third, once you can name the job title you sell to. Paid ads are closed to you this month and that is fine, because they are the most expensive way to learn something the other three teach you for free.

How do you get your first users with no marketing budget?

Write down every person who would recognise your name. Most solo founders end up with 180 to 400 names once they have merged their phone contacts, their LinkedIn connections and every address in their sent mail from the last year. Send 30 of them a short message that asks for nothing, then send a specific fifteen-minute ask only to the people who reply. At a 10% conversion per ask that produces roughly three customers from 30 messages, and it costs you about two hours plus the 45 minutes it took to build the list.

Can you really market a startup for free?

Yes for the first hundred customers, and it stops being true somewhere after that. Three of the four ways of getting customers cost time rather than money, and the tooling underneath them has real free tiers: a free Gmail account sends to 500 recipients a day, EmailOctopus carries 2,500 subscribers and 10,000 emails a month, Substack costs nothing until you charge somebody. What free does not buy you is speed. A budget compresses months into weeks, and with no budget you are paying for the same result in mornings.

What is the fastest free way to get customers for a new SaaS?

Thirty warm messages sent by hand this week. It is the only method that can produce a paying customer within days rather than weeks, because the trust question is already settled and you are only asking for fifteen minutes of attention. The arithmetic is the argument: 30 warm asks at 10% beats 300 cold sends at 0.6%, and the warm version takes about two hours against roughly fourteen for the cold one.

How many people do I need to message before I get a customer?

About ten warm asks, or about 160 cold sends to a hand-built list, at rates a solo founder can hit with no brand behind them. Those two numbers are the reason warm goes first. The cold figure assumes a 12% reply rate on a list you built by hand, roughly a third of repliers looking at the product and about 15% of those paying, so a bought list of 4,000 addresses will not get you there faster than 40 good ones.

Do I need followers to sell software?

No, and founders reliably overestimate the handicap. Having no audience blocks launch-day volume from your own account, anything that compounds off reach, and social proof in your copy. It does not block one-to-one messages, community answering, search, referrals, or a launch held on somebody else's audience. Three of the four ways of getting customers work identically at zero followers and at ten thousand.

How long does it take to market a SaaS with no budget?

Thirty days at these volumes gets you one to four paying customers, 30 to 60 named contacts and one reply rate you can trust. A hundred customers takes most solo founders four to nine months of the same work. The trap is reading month one as a verdict. Warm outreach pays in days, cold pays in weeks, and posting pays in months two to four, so a founder who judges all three at day 30 concludes that two of them failed when neither has reported yet.