How to run advertising without a website: Instagram, lead forms, and messengers

Yes. Meta doesn't require a website: the algorithm optimizes just as well for a lead-form submission, a Direct or WhatsApp message, or a bot subscription. For services with a short decision cycle — beauty, fitness, food, local services — not having a website often doesn't affect the cost per lead at all.

A website isn't a launch requirement — it's a nurturing tool. It's needed when a person needs time to think, compare, and show the offer to others. If your product is bought on emotion within 1–2 days, an Instagram profile does the job of a landing page better than a raw website built on a site builder.

The real question isn't "do you have a website" but where exactly to send traffic for your deal cycle. We covered all the options in where to send ad traffic: Direct, a landing page, or a quiz — here we'll focus only on no-website setups.

The setup is simple: the ad leads either to a profile (the "Traffic" or "Profile visits" objective) or straight into a conversation (the "Messages" objective). In the second case, the person taps a button and lands in Direct with a pre-filled message like "I want to know the price." You get a live conversation instead of a cold number.

The upside is the shortest path to a sale and the highest-quality contact: the person is already messaging you themselves. The downside is that everything hinges on response speed. If a manager replies three hours later, the leads go cold, and no Meta ads specialist can fix that. The second requirement is that the profile has to look like a storefront: pinned Stories, prices, reviews, up-to-date posts.

The risk with profile ads is that you're paying for a click, not a lead, so CPL depends heavily on content. The risk with Direct campaigns is that the inbox fills up with unqualified "so how much is it?" messages. This is fixed with a screening question in the first message and a proper script for the admin.

• The profile needs to answer the basic questions without a website: price, address, how to book
• An auto-reply within the first seconds is the minimum needed to keep the contact
• The "Messages" objective gives a pricier click but a cheaper conversation than plain traffic

A Meta lead form opens right in the feed: name, phone, and email are pulled from the profile automatically, and the person just makes two taps. It's the cheapest lead format there is — it's exactly what got us 96 leads in a 10-day test at a $1.95 CPL in the B2B medical devices niche and 475 leads at a $2.80 CPL in thermal panel insulation.

The flip side of auto-fill is that it's too easy. A person leaves a number without reading the offer and doesn't remember filling it out a day later. So the raw cost per lead is deceiving: count the cost of a qualified lead, not the CPL. In an auto case, out of 4,263 leads, 45% were qualified — the business was fine with that, but you need to plan around that share specifically.

This is fixed with two settings: the "Higher intent" form type with a confirmation screen, and 1–2 qualifying questions in the form (budget, city, timeline). The lead count drops, the cost rises, and the number of real deals rises too. Be sure to connect instant lead handoff to a manager, or the form turns into an archive of dead numbers.

WhatsApp advertising is the same "Messages" objective, but it routes into the messenger tied to the business's number. The format is strong for the diaspora and overseas markets: in Europe, the US, and Kazakhstan, WhatsApp replaces both a phone call and email for many people. The person writes from their phone, the number is visible right away, and the contact doesn't get lost.

Pros: a high reply rate, and it's convenient for agreeing on times and sending photos of work or a price list. Risks: you need a WhatsApp Business account with a properly confirmed number, and some conversations drop off after the first message. Another nuance is analytics: without proper conversation logging, you don't know how many of them made it to payment.

A practical rule: if your audience is Ukrainians abroad, run WhatsApp alongside Direct and see which one produces better-quality conversations. Meta lets you run campaigns on both messengers at once and compare the cost per conversation within a single ad account.

A Telegram bot is the most technical no-website option, but also the most controllable. The person clicks the link from the ad, taps "Start," the bot asks 2–3 questions, and hands the manager a ready-made profile. In effect, you get a quiz without a landing page, plus a contact base you can message again for free.

The upside is automatic qualification: the bot filters out unqualified people before a human spends any time on them. The downside is that part of the audience never makes it to "Start," and you're paying for the click with no result. So a bot pays off where leads are expensive and every minute of a manager's time costs money.

A separate bonus is nurturing. A bot lets you keep working on people who didn't buy right away: reminders, case studies, a limited-time offer. For info products, courses, and services with a "need a week to think" cycle, this is often cheaper than running retargeting at a cold audience.

The choice between a profile, a lead form, a messenger, and a bot comes down to one thing: how expensive an unqualified contact is for you. The cheaper the lead, the more junk is in it; the more barriers at the entry point, the pricier the lead, but the higher the conversion to payment.

Without a website, leads are usually cheaper because there are fewer steps between the click and the lead. At a fitness club in Kazakhstan, we held a $2.06 CPL at 784 leads a month with 90% qualified. In B2B construction — $2.80 per lead on a $1.3K budget over two months. In medical devices — $1.95 in a ten-day test.

This isn't a promise of what your CPL will be: lead cost depends on niche, geo, competition, and season. We've put together benchmarks for different fields separately — cost per lead by niche. But the pattern holds: a lead form gives the lowest number, messengers give a pricier lead with better conversion to payment.

Count all the way through the chain. A $2 lead with a 5% conversion rate is pricier than a $6 lead with a 30% conversion rate. That's why, during an audit, we don't look at the CPL line in the ad account — we look at what a customer who actually paid costs.

You need a website when the purchase decision can't be made from a phone in five minutes. An expensive service starting at several thousand dollars, B2B with sign-off from several people, a deal cycle spanning months — in these scenarios, the client needs something to open on a computer, forward to a colleague, and come back to a week later.

The second marker is the need for proof. If the sale rests on a portfolio, calculations, certificates, technical specs, or a contract, none of that fits into Direct. Real estate, construction, medical equipment, legal services, dentistry — formats where a page with details closes objections better than a chat thread.

The third is analytics and scaling. Without a website, there's no pixel on key events, so the algorithm has nothing to learn from beyond "message." On small budgets this isn't critical, on large ones it's a ceiling. In B2B we often combine both: a lead form for volume plus a page like marketing for medical and B2B business for those who want to dig into the details.

• Ticket under $300 and a decision in 1–2 days — no site needed
• Ticket from $1000, multiple decision-makers, a sales cycle over a month — a website is a must
• A middle ground — a quiz or a one-page landing page instead of a full website

The clearest case is a beauty salon in Ukraine: in 9 months, 3,100+ inquiries and 1,240 bookings, with 90% coming through DMs. There was no website in the funnel at all: ads led to the profile and DMs, and the administrator closed the booking. How we build systems like this is described on the marketing for beauty salons page.

The second example is a food project in Switzerland: in 3 months, 700+ followers, 3700 CHF in sales, and a ROAS of 3.6×, with sales coming through the account and messages. In a restaurant case in Ukraine, a $145 budget over 10 days delivered $7.5K in revenue and a ROAS of 51.7× — also without a separate landing page.

In 4 years and 150+ projects across 40+ niches, we've seen both scenarios: businesses where a website delivered nothing but a developer's invoice, and businesses that hit a ceiling without one. More numbers are in the cases section, and founder Anton Tysiachnik personally reviews your situation during a free audit.