What are the best marketing channels for an indie SaaS?

For a solo founder with no audience and about ninety minutes a morning: warm outreach to your own network first, then referrals from the people already using it, then cold email, then niche communities such as the right subreddit or Slack group, then cold DMs. Those five are the only ones that can report back inside a month at zero cost. SEO, directories and posting in public run underneath them at a low cadence, because their clocks only start when you publish the first thing. Paid channels come last, since they spend money to amplify a message none of the free channels have proved yet.

Which marketing channel should I start with if I have no audience?

Warm outreach, if you can name 30 people who would open a message from you and remember why. It converts at roughly ten times the rate of cold email and it pays back the same afternoon. If that list is honestly under 30 names, start with cold email instead, because it is the only channel that works on day one with no audience, no list and no money. Do not start with posting, ads or a launch, because all three assume something you do not have yet.

Does SEO work for a new SaaS website?

Yes, and not inside your first quarter. Ahrefs, sampling newly published pages, found that 1.74% reach the top ten within a year for even one keyword, revised down from 5.7% in an earlier run of the same study. Fifteen pages published today should therefore be expected to produce nothing measurable for six to twelve months. That is an argument for starting SEO early at one page a week and an argument against making it your main channel, which are different things and often get confused.

Are paid ads worth it for a bootstrapped SaaS?

Not as your first channel, and the reason is arithmetic rather than taste. Dreamdata's benchmark of non-branded B2B Google search ads, covering August 2024 to July 2025, puts the average cost per click at 5.34 dollars, up 29% from 4.13 the year before. At a 2% click-to-trial rate and a 25% trial-to-paid rate, that is 200 clicks and roughly a thousand dollars per paying customer. If your product costs 20 euros a month, no amount of copy fixes that division. Ads amplify a message that already converts, so earn the message somewhere free first.

Is Product Hunt still worth launching on?

It is worth one day of your year and not one month of it. The spike lasts about 48 hours and decays inside a fortnight, so the honest way to value a launch is as a single event with a fixed cost, not as a channel you can run again next Tuesday. Twenty to forty hours of preparation for a one-off spike is a poor trade against the same hours spent on 800 cold emails. Launch when the product is ready, take the backlink and the badge, then go back to the channels that repeat.

Is Reddit good for marketing a SaaS?

It is the best free channel on this list that does not require an audience, and it is fourth rather than first because it takes two to four weeks and one badly judged link to lose the room. The rule that works is boring: answer questions in your area for three weeks without mentioning your product, then mention it only when it is the literal answer to what somebody asked. Three good subreddits or Slack groups is the entire supply, so treat them as a finite resource rather than a scalable one.

How many marketing channels should a solo founder run at once?

One per session, two or three across a week. Splitting a ninety-minute morning across four channels produces four efforts too small to read, so the week ends with a feeling of busyness and no information. Four mornings on the top-ranked channel you can actually stomach, and one morning on something that compounds, is the shape that survives a hundred repetitions.

← All posts

Method

The best marketing channels for indie SaaS: 17 ranked for a founder with no audience

9 August 2026 · 17 min read

Every ranked list of marketing channels is written for somebody. Most of them are written for a company with a growth hire, a quarter to spend and a budget that survives being wrong. This one is written for a person who ships the product, answers the support email, and has about ninety minutes before the day gets loud. Change that person and the order changes, which is why the scope goes first and the table goes second.

Who this ranking is for

One person. No audience worth the name, so under about 500 people who would notice if you published something. Ninety minutes a morning, five mornings a week, which is roughly a hundred hours a quarter and the only real currency in this whole exercise. Under a hundred euros a month available for tools and spend. A self-serve product somewhere between 15 and 100 euros a month, or a one-off price in the same range.

If you have 20,000 followers, the seventh row on this list is your first row and most of the top five are a waste of your afternoon. If you sell a 40,000 euro enterprise contract, cold outreach goes to number one and stays there. If you have a marketing hire, ignore the whole page, because the constraint the ranking is built around is not one you have. The order below is only defensible because the reader is fixed.

The four tests behind the order

Rankings hedge when they have no scoring rule. Here is the one used below, in the order it was applied. How many days until the channel tells you something true? A channel that answers in a week beats a better channel that answers in eight months, because you get to act on the answer seven months earlier. Does the unit fit inside ninety minutes? A YouTube video is four sessions and a cold email is three minutes, and the smaller unit gets finished on the bad days. Does money leave before information arrives? Every paid row does that, and that single property puts all four of them at the bottom. Does it keep producing after you stop? This is the tiebreaker rather than the first test, which is the part most channel advice gets backwards.

Notice the fourth test comes last. Compounding is the strongest property a channel can have, and it only matters to a founder who is still working in nine months. Roughly nothing here compounds inside your first quarter, so ordering the list by compounding first would produce a beautiful table for a founder who has already quit. The reason founders quit marketing is almost never that they picked the wrong channel. It is that they picked one with a nine-month feedback loop and got no evidence they were not wasting their mornings.

Every row also carries a lane, because seventeen channels is a lot to hold in your head and four is not. The lanes are warm outreach, posting in public, cold outreach and paid ads. Alex Hormozi calls them the Core 4 in his book $100M Leads, and the full explanation of the four lanes covers why every marketing tactic ever invented sits inside one of them. A cold DM and a cold email are the same thing with different plumbing. A subreddit answer and a YouTube video are both posting. The lane tells you how a channel behaves; the rank tells you when to run it.

The ranking

Seventeen channels, ranked for the person described above. Read the last column first, because how a channel fails for one person is more useful than how it works for a team.

Channel	Lane	First honest signal	Money	Effort per unit	Compounds	Ceiling	How it fails for you
1 Warm outreach to your own network	Warm	Same day. Ten messages usually return two or three replies before lunch.	None.	Four to six minutes a message, thinking included.	No. A spent name stays spent.	40 to 300 names. About 20 hours of work in total, ever.	You skip it. Emailing 40 strangers costs you nothing socially, and messaging someone whose opinion you care about does.
2 Referrals from people already using it	Warm	Same week. Twenty asks produce three or four names.	None.	Three minutes a message, or one line in the product if you automate the ask.	Weakly. Every new customer adds a name to the list.	One or two asks per user per year before it grates.	You never ask. Founders assume happy users refer on their own, and almost none of them do.
3 Cold email	Cold	Two weeks and 50 sends of one unchanged message. Below 50 the rate is noise.	Nothing, up to about 80 euros a month for a sending domain, data and a sequencer.	Two to four minutes a message, including the one researched line.	No. A message is spent the moment it lands.	Around 800 a month by hand, before the researched line falls apart.	You send 15, get one polite no, rewrite the whole thing, and restart the count from zero forever.
4 Niche communities and subreddits	Content	Two to four weeks of answering questions properly.	None.	30 to 60 minutes to read the room and leave three useful answers.	Yes, when the thread is still searchable a year later.	The size of the room, and it caps low. Three good rooms is the whole supply.	You drop a link in week one, get removed, and decide the channel hates founders.
5 Cold DMs on X and LinkedIn	Cold	About a week. Faster than email because people answer inside the app.	None, or 30 euros a month for the tier that raises the caps.	Three to five minutes a message, plus reading their last few posts.	No.	Platform limits. Roughly 20 to 30 a day before the system decides you are a bot.	Automation. The day you buy a tool that sends them for you, the reply rate and the account both go.
6 Directories and listicles	Content	Three to six weeks, once the listing pages have been crawled.	Nothing, up to about 100 euros in total for a handful of paid slots.	15 to 25 minutes a listing once your copy and screenshots exist.	Yes. The page sits there and the link keeps pointing at you.	Low and finite. The first 30 listings do nearly all of it, so about 12 hours total.	Nothing dramatic. It is small, boring and finite, which is exactly why it stays undone.
7 SEO and long-form content	Content	Six to twelve months on a new domain. Nothing readable before that.	None, unless you buy a keyword tool you do not need yet.	Five to eight hours a page done properly, so three or four sessions each.	Yes, harder than anything else on this list.	Nothing you will hit in year one or year two.	You quit in month four, which is before the first page could plausibly have ranked.
8 Build in public on X	Content	Six to twelve weeks, or about 30 posts. The first ten are rehearsal.	None.	20 to 45 minutes a post, with replies counted as part of the job.	Weakly. The feed buries a post in 48 hours, so the following it leaves behind is the asset.	High for attention, low for revenue, unless founders are your buyer.	You start counting followers. Follower count and revenue are not the same variable.
9 App Store optimisation	Content	Three to six weeks, and only once you have enough installs for the store to rank you at all.	None, plus a screenshot tool if you want one.	A one-off day for the listing, then an hour a month.	Yes, for as long as the ranking holds.	Only exists if you ship an app. If you do, move this row into the top three.	You write the subtitle for yourself instead of for the words a stranger types into store search.
10 Indie Hackers and founder forums	Content	Two to six weeks, and quicker than most rooms because posting is encouraged.	None.	30 to 60 minutes a post or thread.	Mildly. Old threads get found, mostly by other founders.	The audience is founders. That is either a ceiling or a bullseye, and only you know which.	Everybody in the room came to be read, not to buy. Encouragement is not demand.
11 Product Hunt and launch platforms	Content	One day. The day is the entire result.	None, plus 20 to 40 hours of preparation.	A week of your mornings spent on a single event.	No. The spike decays inside a fortnight and the page goes quiet.	One real launch per product.	Three weeks of preparation, a fifth place finish, 400 signups and nobody who opens it twice.
12 Partnerships and integrations	Warm	Six to twelve weeks, because it moves at the other side's speed rather than yours.	None, plus engineering time you were going to spend on the product.	Days, not minutes. One integration is a sprint.	Yes, once it ships. The integration keeps sending people while you sleep.	High, and gated on having something worth partnering with.	Four friendly calls, nobody says no, and nothing ever ships.
13 YouTube	Content	Six to twelve months and roughly 20 videos.	None to start. A decent microphone eventually.	Three to eight hours a video, so two or three sessions each.	Yes. A video that ranks keeps playing for years.	Very high, and further away than anything else here.	Four videos in a fortnight, then none for two months, because the unit is too big for a 90-minute morning.
14 Affiliate deals	Paid	Two to three months, because affiliates only promote what already converts.	20% to 30% of revenue, and only after a sale.	Hours per partner to recruit, then close to zero.	Yes, when a partner sticks.	Bounded by how many people already hold your buyer's attention.	You recruit 30 affiliates and 28 of them never send a single click.
15 Newsletter sponsorships	Paid	One send. You know inside 72 hours.	200 to 1,500 euros for a slot in a small, well-targeted list.	An hour to write the copy, once you already know what converts.	No. The issue is read and then it is over.	The list size. Small lists convert better than big ones and cost a tenth as much.	You buy the slot before you own a message that converts, and the slot amplifies nothing.
16 Paid search	Paid	Seven to fourteen days and 300 to 500 euros of spend.	Real money before any result. Dreamdata puts the average B2B non-branded click at 5.34 dollars.	Minutes a day once it runs. Cheapest row for your hours, dearest for your bank.	No. Stop paying and it stops the same hour.	Your budget, which is a constraint none of the top ten rows have.	The arithmetic rather than the copy. Two hundred clicks per customer at that price is about a thousand dollars each.
17 Paid social	Paid	Seven to fourteen days and 250 to 500 euros of spend.	The same shape as paid search with worse intent, because nobody on the feed asked.	Hours to make the creative, minutes to run it.	No.	Your budget again.	You optimise toward installs and buy 21 signups who never pay. We spent 258 euros learning that.

Ranked for a solo founder with no audience, ninety minutes a morning, under a hundred euros a month, and a product priced between 15 and 100 euros a month. Durations and costs are Distronaut's own estimates from running these lanes, not survey findings. The two figures taken from published sources are named where they appear.

Where the four lanes landed

Collapse the seventeen back into four and the shape of the argument shows up immediately. Warm holds the top of the table and runs out fastest. Paid holds the bottom of it for a single structural reason.

The seventeen, grouped by lane

Warm 1, 2, 12 Two of the top three, and the fastest lane to exhaust. Twenty hours of work and it closes until you make more warmth somewhere else.

Cold 3, 5 The only lane that works on day one with no audience, no list and no money. Supply is effectively unlimited and every message costs you minutes.

Content 4, 6, 7, 8, 9, 10, 11, 13 Eight of the seventeen, and the widest spread of quality on the list. Rank 4 pays inside a month and rank 13 pays inside two years.

Paid 14, 15, 16, 17 The bottom four, for one reason. Money leaves before you know whether the message works, and you are the only person who can tell it what to say.

Why warm outreach is first and not third

The top two rows are the ones founders argue with, and the argument is never about the numbers. Warm outreach converts at roughly 10% per ask. Cold email converts at around 0.6% of sends. Same product, same sentence, sixteen times the result, purely because of who is on the other end. A list of 150 names is around fifteen hours of messaging and produces something like twelve to fifteen customers. Nothing else in the table gets near an hour and a half per customer.

The objection is always that the list is small, and the objection is correct. Most solo founders have somewhere between 40 and 300 names worth contacting, which is 20 hours of work and then the channel closes. That is a ceiling, not a flaw. It is also an argument for doing it in week one rather than week twenty, because the names do not get warmer while you avoid them. The warm outreach scripts exist because the real blocker is never the strategy. It is writing the first line to someone who knew you before this was your job.

Second place goes to the group nearly everyone forgets: the people already using the thing. They are warm, they have proved intent by turning up, and most of them have never been asked for anything at all. Forty users asked one direct question is about six hours and reliably produces three or four names. The reason this sits at rank two rather than rank one is only that it depends on having users, which the reader of rank one may not.

Why cold email is third and not first

Cold email is the best channel on this list by every measure except one. Unlimited supply, works on day one, no audience required, and a readable answer in about a fortnight. It is also the only channel that hands you vocabulary: twenty replies show you the words strangers use for your problem, and those words then fix your landing page, your ads and everything you post afterwards.

It sits third for one reason. At 40 sends a session and 0.6% of sends becoming a paying customer, a hundred hours of mornings is 2,400 messages and about fourteen customers, so seven hours each. Warm does the same job in an hour and a half. Spending your first fortnight on cold while 150 warm names sit unmessaged is choosing the five-times-more-expensive option because it is emotionally cheaper. Run warm until the list is genuinely empty, then run cold forever, because cold is the one that never runs out. The cold email templates give you a starting message and the outreach volume estimator tells you how many sends your actual target needs at your actual rates.

Why a subreddit beats your blog for a year

Rank four is the one most likely to be argued with by people who write about marketing for a living. Sitting in three good rooms and answering questions properly beats publishing on your own domain for roughly the first year, and the reason is distribution rather than quality. The subreddit already has the readers. Your blog has you.

The failure mode is well documented and entirely self-inflicted. You join, post a link in week one, get removed, and conclude that communities hate founders. What they hate is a stranger arriving with a URL. Three weeks of answering questions without mentioning your product buys you the right to mention it once, when it is the literal answer to what somebody asked. Three rooms is the whole supply, so treat it as a finite resource. The what to post when nobody is reading piece covers what goes in the answers.

Why SEO is seventh, with numbers

Here is the row where an honest ranking has to be uncomfortable. SEO is the single best channel on this list over a two-year horizon and it belongs at rank seven for the first quarter, which are not contradictory statements once you look at the clock.

Ahrefs, sampling newly published pages, found that 1.74% of them reach the top ten within a year for even one keyword. An earlier run of the same study put it at 5.7%, so the figure has moved in the wrong direction rather than the right one. Take the friendlier of the two numbers. A hundred hours buys you fifteen to twenty long pages at five to eight hours each. At 5.7%, one of those twenty pages reaches the top ten inside twelve months. At 1.74%, none of them do. Either way your first quarter produces zero readable signal, and you will spend every one of those mornings unable to tell "working, not surfaced yet" apart from "never going to work".

So publish one page a week from day one and do not make it the plan. The clock only starts when you publish the first one and there is no way to buy back the months you did not spend, which is the entire case for starting early at a low cadence. The full version of that argument, with the ninety-day arithmetic, is in cold outreach versus content marketing.

The first hundred hours, priced

Ninety minutes a morning, five mornings a week, thirteen weeks. About a hundred hours. Here is what each channel does with them, using rates a founder with no brand and no case study can actually hit.

Channel	What a hundred hours buys	Customers by day 90	Hours per customer
Warm outreach	You cannot spend 100. The list runs out at about 20 hours, then it closes.	12 to 15	About 1.5 hours
Referrals	40 existing users asked once, properly, plus follow-ups. Around 6 hours.	3 to 5	About 1.5 hours
Cold email	2,400 sends at 40 a session, one message frozen for the first 50.	About 14	About 7 hours
Cold DMs	About 1,500 messages, capped by the platforms rather than by your hours.	8 to 12	About 9 hours
Niche communities	Roughly 300 genuinely useful answers across three rooms.	4 to 8	About 15 hours
Directories	Every listing worth having, twice over. The work runs out at about 35 hours.	2 to 4	About 10 hours, falling afterwards
Build in public	About 65 posts and the replies underneath them.	2 to 3	About 35 hours
SEO	15 to 20 long pages, published and forgotten about.	0 to 1	Unmeasurable yet. Look again on day 300
Paid search	Four hours of setup. After that your hours stop being the constraint.	Whatever the budget buys	Half an hour and about 1,000 dollars

Distronaut's own arithmetic, not a survey. Rates used: warm asks convert at about 10%, cold email at 0.6% of sends, cold DMs at about 0.8% of sends against a lower daily cap, community answers and posts at inbound rates closer to 20% of the conversations they start. Your market will move these. It will not reorder the top two.

Two things fall out of that table. The first is that the top of the ranking is not close. Warm outreach is roughly five times cheaper per customer than cold email and twenty times cheaper than posting, and it is the channel most likely to still be untouched in month four. The second is that the bottom of the table is not really competing for hours at all. Paid search costs you four hours and then starts costing money, which is why comparing it on hours-per-customer misses the point and comparing it on euros does not.

Why every paid channel is in the bottom four

Dreamdata's benchmark of non-branded B2B Google search ads, covering August 2024 to July 2025, puts the average cost per click at 5.34 dollars, up 29% from 4.13 the year before. Run that forward. At a 2% click-to-trial rate and a 25% trial-to-paid rate, which are both generous for a landing page nobody has tested yet, one customer costs 200 clicks. That is about 1,070 dollars. Call it a thousand euros, since the two currencies are close enough that the conclusion does not move.

A customer paying 25 euros a month who stays a year is worth 300 euros. There is no headline that fixes a division like that. Paid works when one of the free channels has already found a sentence that reliably makes strangers reply, because then you are buying reach for something that converts rather than buying a fast answer to a question you have not asked properly. Newsletter sponsorships sit highest of the four purely because a 300 euro slot in a small well-targeted list reports back in 72 hours, which is the fastest paid feedback loop money can buy.

The failure that actually happens is duller than a bad ad. You point the platform at signups because signups are the event you can measure early, the platform delivers exactly that, and you buy a pile of people who will never pay. We spent 258 euros on that lesson: 21 signups, zero customers. Send the paid subscription event back to the ad platform instead, accept that this makes every test slower, and set a total cap and a stop date before the first euro leaves.

The three most people expect to be higher

Product Hunt, at eleven. A launch is an event, not a channel. Twenty to forty hours of preparation buys one day, the spike decays inside a fortnight, and you get to do it once per product. The same forty hours spent on cold email is 800 messages and about five customers who are still paying at Christmas. Launch when the product is ready, take the backlink, then go back to the things that repeat.

Build in public, at eight. It works, and it mostly works at getting you founders. If founders are your buyer, move it to four and take the promotion. If they are not, you are building an audience of people who will cheer for your MRR chart and never open their wallet, which feels like traction for about five months. Follower count and revenue are different variables and the feed rewards the wrong one.

YouTube, at thirteen. Nothing on this list compounds harder and nothing has a worse unit size for a ninety-minute morning. Three to eight hours a video means two or three sessions before anything ships, and about twenty videos before the algorithm has enough to work with. That is a year of your mornings against a channel that reports in month nine. Correct choice at hour 2,000. Wrong choice at hour 20.

What changes the order for you

The list above is a default, and defaults are meant to be overridden by facts about your business rather than by preference. Eight conditions genuinely reorder it. If none of these describes you, run the ranking as written.

If this is true of you	Move	Why
You ship an iOS or Android app	App Store optimisation, 9 to 3	Store search is the one place where strangers are already typing the problem and the shortlist is four results long. A day on the listing is the highest-return day in this table.
Founders and indie hackers are your actual buyer	Indie Hackers and build in public, 10 and 8 to about 4	The usual objection to those rooms is that everyone in them is selling rather than buying. That objection disappears entirely when the sellers are the customers.
Your price is above 100 euros a month and there is a call in the flow	Cold email 3 to 1, paid search 16 to about 8	At 1,200 euros of first-year value, a thousand-dollar customer is a slow yes rather than a no. Warm still beats it, but the warm list runs out and this does not.
Your price is under 20 euros a month with no sales call	Cold email 3 to about 8, paid off the board entirely	Seven hours per cold customer against 75 euros of lifetime revenue is nine euros an hour, before churn. Communities, directories and SEO are the honest set at that price.
You cannot describe your buyer well enough to find 500 of them	Communities 4 to 1, cold email 3 to about 9	Cold outreach to a list you invented burns a sending domain to learn nothing. Sitting in the room where they complain is the cheapest way to find out who they are.
You already have 5,000 people who read what you write	Posting to 1, and warm folds into it	This whole ranking assumes no audience. With one, the slowest channel on the list has already been paid for and the first four rows are a rounding error next to it.
You have a message that converts and 1,000 euros spare	Newsletter sponsorships, 15 to about 6	Sponsorship is the only paid row that reports in 72 hours for a few hundred euros. It stays at 15 until something else has proved the sentence you are buying reach for.
You need revenue inside 60 days or the project stops	Nothing below rank 6 exists this quarter	Six of the seventeen can report back inside eight weeks. Ranking the rest is an interesting exercise for a founder who still has a runway to spend on it.

One condition can fire at a time or several. When two conflict, follow the one about price, because price decides whether the labour arithmetic works and nothing else on the list can overrule it.

Running the top of the list without doing any of it badly

A ranking is only useful if it survives a Tuesday. The way this one dies is a founder reading the table, agreeing with it, then splitting a ninety-minute block across four channels and finishing the week with twelve cold emails, half a post, two community comments and no information about any of it.

Do the rotation at the day level. Four mornings on whichever of the top five is currently yours, one morning on something that compounds, and nothing else on the board. When the warm list empties, cold moves into the four mornings and the shape of the week does not change. The daily marketing routine covers what the ninety minutes actually contains, and marketing a SaaS with no audience and no budget covers the version of this where the budget line is zero rather than a hundred euros.

One thing to do before the first morning, and it takes four minutes. Work out how many actions your target actually needs at your actual rates, because the answer is either "fewer than you feared" or "this target is arithmetic fiction", and both are worth knowing on day one instead of day fifty. The 100 in 100 goal calculator does the sums, and the first 100 customers playbook has the specific asks for each of the top five rows. If your target is revenue rather than logos, how many customers you need to reach 10k MRR does the same division against a monthly number.

That is the ranking, the arithmetic under it, and the conditions that break it. The order matters less than the fact that you now have one, because the founders who lose a year to this rarely lose it to the wrong channel. They lose it to relitigating the choice every Monday for eleven weeks.

Questions founders actually ask

What are the best marketing channels for an indie SaaS?: For a solo founder with no audience and about ninety minutes a morning: warm outreach to your own network first, then referrals from the people already using it, then cold email, then niche communities such as the right subreddit or Slack group, then cold DMs. Those five are the only ones that can report back inside a month at zero cost. SEO, directories and posting in public run underneath them at a low cadence, because their clocks only start when you publish the first thing. Paid channels come last, since they spend money to amplify a message none of the free channels have proved yet.
Which marketing channel should I start with if I have no audience?: Warm outreach, if you can name 30 people who would open a message from you and remember why. It converts at roughly ten times the rate of cold email and it pays back the same afternoon. If that list is honestly under 30 names, start with cold email instead, because it is the only channel that works on day one with no audience, no list and no money. Do not start with posting, ads or a launch, because all three assume something you do not have yet.
Does SEO work for a new SaaS website?: Yes, and not inside your first quarter. Ahrefs, sampling newly published pages, found that 1.74% reach the top ten within a year for even one keyword, revised down from 5.7% in an earlier run of the same study. Fifteen pages published today should therefore be expected to produce nothing measurable for six to twelve months. That is an argument for starting SEO early at one page a week and an argument against making it your main channel, which are different things and often get confused.
Are paid ads worth it for a bootstrapped SaaS?: Not as your first channel, and the reason is arithmetic rather than taste. Dreamdata's benchmark of non-branded B2B Google search ads, covering August 2024 to July 2025, puts the average cost per click at 5.34 dollars, up 29% from 4.13 the year before. At a 2% click-to-trial rate and a 25% trial-to-paid rate, that is 200 clicks and roughly a thousand dollars per paying customer. If your product costs 20 euros a month, no amount of copy fixes that division. Ads amplify a message that already converts, so earn the message somewhere free first.
Is Product Hunt still worth launching on?: It is worth one day of your year and not one month of it. The spike lasts about 48 hours and decays inside a fortnight, so the honest way to value a launch is as a single event with a fixed cost, not as a channel you can run again next Tuesday. Twenty to forty hours of preparation for a one-off spike is a poor trade against the same hours spent on 800 cold emails. Launch when the product is ready, take the backlink and the badge, then go back to the channels that repeat.
Is Reddit good for marketing a SaaS?: It is the best free channel on this list that does not require an audience, and it is fourth rather than first because it takes two to four weeks and one badly judged link to lose the room. The rule that works is boring: answer questions in your area for three weeks without mentioning your product, then mention it only when it is the literal answer to what somebody asked. Three good subreddits or Slack groups is the entire supply, so treat them as a finite resource rather than a scalable one.
How many marketing channels should a solo founder run at once?: One per session, two or three across a week. Splitting a ninety-minute morning across four channels produces four efforts too small to read, so the week ends with a feeling of busyness and no information. Four mornings on the top-ranked channel you can actually stomach, and one morning on something that compounds, is the shape that survives a hundred repetitions.