How do you get your first 100 website visitors?

Your first 100 website visitors matter out of all proportion to their number. They are how you find out whether the page communicates, whether anybody clicks the button, and whether the idea survives contact with a stranger who owes you nothing. Below are the five sources that reliably produce them, in the order I would run them, and — because a hundred visitors is really a hundred data points — what each batch actually proves. (Disclosure: published by BidSurvivor, which is source four and is ours.)

First, decide what the hundred is for

Most people go looking for a hundred visitors when what they want is a hundred customers, and then read the result as a verdict on the product. It is not. At this volume you are testing three things and none of them is demand at scale:

1. Comprehension. Does a stranger understand what this is within one screen?
2. Intent. Does anybody who understands it want it enough to click?
3. Findability. Can a person who wants it find it at all?

A hundred visitors from a single source answers one of those. A hundred from five sources answers all three and tells you which of them is broken. That is the entire argument for spreading the first hundred across sources, and it is the opposite of the advice for the first $200 of paid spend, where concentrating on one channel is right. Free traffic has no minimum sample cost; paid traffic does.

Source 1: the people who already know you (~20 visitors)

Your own network, posted once, honestly: I built this, tell me what is confusing. Not "please share".

These visitors are warm and nearly useless for market data. They click out of loyalty and they will tell you it looks great. What they are good for is the one thing strangers will never do, which is tell you where they got stuck. Watch for the sentence "wait, is this for—?" That sentence is the bug report.

Fix what they trip on before inviting anybody who does not like you.

Source 2: one community post (~30–50 visitors)

The subreddit, forum or Discord where your users already complain about the problem you solve. One post that leads with the problem and mentions the product second, once, at the bottom.

These are the highest-intent visitors on this list, which makes them the most informative and the most brutal. If they do not click, you do not have a traffic problem, you have a message problem, and no amount of extra traffic will fix it. Read the subreddit's rules before posting — Reddit's content policy is the floor, and individual subreddits sit well above it on self-promotion.

One post. Not five, in five subreddits, on the same afternoon. That reads as what it is.

Source 3: a small-stage launch (~50–200 visitors)

Uneed, Fazier or Peerlist rather than Product Hunt for a first outing — the full comparison is here, and the short version is that a small directory gives you a real, survivable first launch instead of one shot at a leaderboard you are not ready for. Uneed publishes its queue, so you can see how many products you are competing with on the day.

The visitors are curious, product-minded strangers with no relationship to you. That is exactly the audience for testing whether your one-liner stops a scroll, which is a skill you can practise: one-line pitch examples has the patterns that work and the ones that read as filler.

Keep Product Hunt for later, when the page is good. It works once.

Source 4: a free ad slot (a trickle, fully measured — ours)

BidSurvivor runs two twelve-hour ad slots a day, sold by live auction. The first brand into an empty slot pays $0 — no account and no card — and every visitor to the board starts with $100 of house credit before signing up for anything, so a first attempt costs nothing out of pocket.

Be clear about the size of it: this is a young board, the numbers are small, and they are published on the front page and on each brand's own page rather than described in adjectives. What makes the trickle worth having at this stage is not volume, it is that the clicks are cold. Nobody's friend, nobody's community, no launch-day goodwill. Cold clicks are the strictest test a pitch line gets, and here the test reports its own result: you can see how many people opened the card and how many left for your site.

The honest limitation is that a board this size cannot tell you whether a channel scales. It can tell you whether a sentence works.

Source 5: directory findability (a slow drip, forever)

An evening of directory submissions — BetaList, SaaSHub, AlternativeTo, Startup Stash — will not produce a spike, and anybody who tells you otherwise is selling submissions. What it produces is a permanent drip and a searchable name, which every other source on this list quietly depends on: people who see you in a community post go and search for you afterwards.

It also matters more than it used to. Ahrefs' study of search traffic found the vast majority of pages get essentially no search traffic at all, and directories are how a page with no authority becomes quotable by anything — including the answer engines that increasingly sit between a search and a click.

Read the hundred like data, not like a scoreboard

Line the sources up against what each one proves:

Source	What it validates	What it cannot tell you
Your network	Clarity, first-run bugs	Whether anyone wants it
A community post	Demand, in one niche	Whether it generalises
A small launch	The pitch line	Retention
A cold ad slot	The one-liner, coldly	Scale
Directories	Findability	Anything, quickly

If every source converts badly, the page is the problem. If only the cold sources convert badly, the one-liner is the problem. If the warm sources convert and nothing else sends anybody, you have a findability problem and should spend the next week on source five rather than on ads.

A two-week order of operations

Week one: sources 1 and 5. Network on Monday, fix what they trip on by Wednesday, spend Thursday evening on directories. Nothing here costs money and both take effect slowly.

Week two: source 2, then 4, then 3, in that order, one per day so you can attribute what arrived. Do not run a launch and a community post on the same day; you will get one number and two possible explanations.

Then stop and read it. A hundred visitors is enough to fix a page and not nearly enough to judge a market. The next milestone — turning a working page into an actual launch day — is in how to get traffic on launch day, and the cheapest ways to keep the drip going after that are in the cheapest way to get website traffic.