A B2B landing page that checks the company by its tax ID before a lead reaches sales
Salespeople at B2B software companies spend hours on calls just to find out whether a lead is worth a demo. This landing page does that check for them. A multi-step form asks for the company's tax ID, and within seconds an n8n workflow pulls data about the company from Dadata and SPARK, two Russian company databases, and puts the lead in group A, B, C or D. Leads that fit reach sales in amoCRM together with a company profile.
Salespeople call every lead to weed out the random ones
Let's take a B2B software company that builds AI tools for lawyers as an example. Its customers are law firms of 20 or more people and the legal departments of large companies. Every sale starts with a demo, and the sales team is two people.
Anyone who sees "AI for lawyers" and wants to try it for free books a demo: freelancers, students, solo lawyers. For the estimate I assume that 70% of leads are outside the target market. To find out whether a lead is worth their time, a salesperson looks up the company and calls. That takes about 20 minutes per lead, which at 40 leads a week comes to more than 13 hours.
The lead itself says nothing about the company. It arrives in amoCRM as "name, email, phone, message", with no revenue, headcount or industry. Who to call first is a guess.
Five steps from the form to a lead card in amoCRM
A visitor fills in a multi-step form on the landing page and enters the company's tax ID. Then n8n, a tool for building automated workflows, takes over. It looks the company up in two databases, Dadata and SPARK, gives the lead a score of A, B, C or D, and uses that score to decide what happens next: the lead goes to a salesperson, gets an email sequence or receives a polite decline.
A Tailwind v4 page built in advance, with no extra code in the browser
Five steps, each on its own screen: tax ID, phone, headcount, task and consent to data processing
Both are queried at once: name, OKVED code, headcount, revenue, whether the company is active
Rules: fit with the target customer, revenue, headcount, industry, whether the company is being wound up
A and B go to amoCRM with a Slack alert to sales; C gets an email sequence; D gets a polite decline
The page opens quickly
Astro 5 builds the pages in advance, so the browser receives a finished page without extra scripts. The target is for the main block to appear in under 2.5 seconds, which is where Google draws the line for good loading speed. Speed affects both search ranking and whether a visitor keeps reading. A lawyer at a large company decides in the first few seconds whether to stay on the site.
The multi-step form works as a filter
Each question in the form has its own screen. People who get to the fifth and last step are seriously interested in the product, so some random leads drop out before they reach sales. Yandex Metrica shows how many people leave at each step, and a step that isn't needed can be removed.
Company data comes from Dadata and SPARK
Once the tax ID is entered, n8n queries Dadata (legal name, OKVED code, headcount, region) and SPARK (revenue for two years, year of registration, whether the company is active or being wound up) at the same time. Within a few seconds you know more about the company than the person wrote in the form. The score is built on that data.
A/B/C/D scoring follows clear rules
The score combines several conditions. For the company in the example they are: the company fits the target customer (a law firm or a legal department), revenue of ₽100M a year or more, 20 or more employees, an active company, and a heavily regulated industry. The result is a letter, A, B, C or D, and the letter sets how quickly and in what way the lead gets an answer: A within 2 hours, B within 24 hours, C gets an email sequence, D gets a polite decline with links to the documentation. The conditions and response times are set to match your target customer.
amoCRM gets the full picture
The lead card in amoCRM has everything: the company, revenue, headcount, OKVED code, score and a hint on how to run the conversation. A salesperson opens the card and knows at once where to start. For leads in groups A and B, sales also gets an alert in Slack.
Prebuilt pages and Russian company databases
Builds pages in advance. By default there are no scripts in the browser; interactive blocks are added only where needed
Styles are configured right in CSS and rebuild instantly after edits
Runs the whole workflow: form intake, parallel requests to the databases, scoring, routing
Data by tax ID: legal name, OKVED code, headcount, region, contacts
Financial data: revenue, profit, taxes, whether the company is active
A lead card with all the data, score tags and a suggested next step
Instant alerts to sales about A and B leads, with a link to the amoCRM card
If plans are sold right on the site, takes payment without a salesperson
Parts of this setup already run in my projects. I built the website of a switchboard manufacturer with Astro 5 and Tailwind v4. In ten Lighthouse runs in mobile mode, the main block of its home page appears in 2.1 seconds. A request from its form is saved on the server first and only then sent to Telegram.
What to expect after launch
The figures are for the same company. The first two cards show the starting figure, then the target after launch. For your company, I'll recalculate them based on the leads you actually receive.
sales gets groups A and B; the workflow answers the rest
the company data is already in the card, so the call is only to book the demo
more than a working day to spend on customers who fit
Where these figures come from. Today each of the 40 leads takes about 20 minutes, which is more than 13 hours a week. After launch, sales gets about 12 leads in groups A and B: the 30% that fit the target customer. Assume a call on each takes about 15 minutes: 3 hours in total. Another hour or so goes on looking through the group C cards in case the score got something wrong. That makes about 4 hours a week.
Demo conversion and cost per lead can't be worked out in advance: they depend on the ads and on the product itself. Yandex Metrica will show them in the first months after launch.
How long the launch takes
from the start of work to launch. The exact time depends on how your amoCRM is set up and how many conditions go into the score
access to both company databases is billed at the services' own rates
Prices for an Astro landing page and for automatic handling of enquiries are on the Pricing page.
Where else this setup works
The general pattern: a fast landing page, a multi-step form, company data by tax ID, a score and routing to the right person. It works wherever a B2B deal starts with a demo or a call and it pays to screen leads before a salesperson talks to them:
- → Other B2B software with a clear target customer: HR, finance, legal, medical
- → Consulting firms: screening by revenue and industry before a free audit
- → Marketing agencies: screening out small clients and freelancers before a proposal
- → Software development: telling casual interest from real projects with a budget
- → Financial services: banks, leasing and factoring, with an initial check in SPARK
- The landing page foundation on Astro 5 and Tailwind v4
- A multi-step form with a progress bar and field checks; for another target customer only the questions change
- An n8n workflow that queries Dadata and SPARK at once, fills in the company data and scores the lead
- An amoCRM lead card with all the data, and Slack alerts about priority leads
Related cases and ready-made solutions
Finding B2B clients with cold emails written by AI
Five steps. Apollo picks senior decision-makers who fit the customer profile, Clay collects 4–7 facts about…
Conversions go to the ad platforms straight from the server, and the team's report builds itself
The Keitaro tracker collects clicks and conversions on your own server. An nginx layer with Lua removes…
Weekly client reports that put themselves together
Every Monday a workflow in n8n, an automation tool, pulls the numbers for all eight clients from Facebook…
Start with the 5,000 ₽ audit: a written report and a payback estimate
I'll tell you what to automate first, when it will pay for itself and whether your task needs AI at all. Sometimes it doesn't.
Or just ask a question. During business hours I reply within 2 hours