CASE STUDY: What to expect from a cold email campaign? Specific examples and benchmarks

catlinpuhkan
Dec 2, 2025
6 min read

Interview with Sander Aavik, co-founder of the agency vDisain.

vDisain’s 2025 revenue is 1.6 million euros. Last year, it was 1.2 million. The team consists of 34 people, with offices in Estonia, Latvia, and Lithuania. Before founding his own agency, Sander sold books in the United States with the Southwestern team.

Let us take as an example your own sales campaign for the website management service. Who was the target group of this campaign, and what kind of companies?

We targeted companies that have a website built on WordPress and that use the WooCommerce e-commerce extension. In other words, it is the combination of WordPress together with WooCommerce. The reason is very simple: it is a popular but also vulnerable platform, and these are precisely the types of websites that benefit most from management and security.

How did you identify who uses WordPress and WooCommerce?

It is a multistep process. First, we rely on company databases. In Baltic countries, this is the business register, along with several other data sources.

Then we use tools that index websites and detect the software they run. There are solutions that can identify whether a site is on WordPress, which plugins are installed, whether WooCommerce is active, what analytics tools are used, and what user tracking systems are present. This allows us to filter out exactly those domains that run WordPress and WooCommerce.

In the end, we have a list such as www.someshop.ee, www.company.ee, and so on. All are WordPress plus WooCommerce users.

Once you have the domains, how do you find the right contacts and their email addresses?

Here, we use a combination of scraping, registers, and LinkedIn. We have our own scraper that goes through all the identified domains and collects all publicly available email addresses listed on the site.

At the same time, we use the business register, where we obtain company contacts, job titles, and sometimes direct email addresses. In addition, we check LinkedIn to see whether the company has an official page and which employees are listed there. This usually leads us to the CEO, marketing manager, or IT manager, depending on who is the most relevant recipient for the specific offer.

If we have the first name, last name, and the company’s domain, deriving the email address through permutations is relatively easy. Most companies use formats such as firstname@domain, firstname.lastname@domain, lastname@domain, and similar variations. Our software tests these combinations and usually identifies the correct one.

In some campaigns, we have also used API based scraping tools or the ChatGPT API to automatically extract contact emails from the domain. We always try to keep our costs under control, because we need very large datasets and cannot afford to pay too much for each email.

How large a list did you gather for this management campaign?

In this particular campaign, we identified approximately 843 emails belonging to personal contacts. These are individuals with identifiable email addresses. In addition, we had around one thousand general addresses, such as info@, which we retrieved from the business register. Altogether, this produced roughly one thousand real contacts and a similar number of general addresses.

What did the email content look like, and how personal did you make it?

We use a fairly structured format, where each part of the message has a clear purpose.

Firstly, the subject line and the opening sentence. We call these the opener. Their single task is to get the recipient to open the email. We try to make this as relevant as possible. For example: “I noticed that [domain] uses the WordPress platform...” This is immediately connected to the company and feels authentic.

Next comes the value proposition, which explains what we are offering and why it is beneficial for them. We briefly describe that the management service keeps their WordPress site secure, ensures a fast response time to changes, and helps develop additional functionalities without the need to hire an in-house development team.

At the end, there is a clear call to action. This can be an invitation to a short call, a request for a price quote, or a question asking whether they would like to see examples of similar projects. We try to make the call to action something that is as easy as possible to accept.

Personalization for us means contextual relevance. We know what platform they use, sometimes which sector they operate in, or whether they export to a certain market. This is far more important than adding comments such as “I saw on LinkedIn that you have a dog” or “I saw that you recently traveled to Greece”. These details may sound impressive, but in practice, they are difficult to automate well and do not necessarily improve conversion rates.

We send all sales emails in plain text format. This means the email contains no HTML code, no designed blocks, no images, and no tracking pixels. Sometimes, we even include a link in a way that requires the recipient to copy and paste it manually rather than clicking directly.

The reason is very simple. If an email consists of plain text, there is practically no technical risk for email servers that the message might contain a virus or malicious code. The primary task of spam filters is to protect users from harmful content. A plain text message is therefore much more likely to pass through.

In addition, open rates are partly misleading. Many email clients, such as Apple Mail, technically open messages automatically, which means open rate statistics are never fully accurate. We measure success based on replies and inquiries rather than attractive report numbers.

Sending mass emails from the company’s main domain is risky, because the domain can end up on spam blacklists.

We use several domains and mailboxes purchased specifically for the campaign. From one account, we send a maximum of 30 emails per day, and from one domain up to 100. This is far safer for both the email servers and our reputation.

We prefer to use domains that resemble the main domain rather than completely unrelated ones. For example, we may add prefixes or suffixes such as go companyname, companyname team, companyname digital, companyname marketing, and so on.

It is important that if someone enters such a domain into a browser, there is a proper redirect to the main website. And if someone searches for that domain in Google, the results should still lead to our company rather than to an empty or suspicious domain. We do not use entirely invented domains ourselves, although this is sometimes done in the market.

What was the result of this specific campaign?

We sent up to three emails in sequence to 843 contacts. If the first message received no response, we sent a follow-up. If the second received no response, we sent a third. The intervals ranged from a couple of days to nine days. It was a human and reasonable cadence rather than aggressive spamming.

Sixty-one people replied to the first email, giving a response rate of about 7.2 percent. Ten of these were genuinely interested and asked for more information, a price quote, or a meeting.

The second email went to 775 recipients and generated 70 replies, or approximately nine percent. Six of these were interested.

The third email went to 690 recipients and received 30 replies, a response rate of about 4.3 percent. One of these was interested.

In total, we received around 160 replies. This includes those who politely declined or asked not to be contacted again. Genuine leads, meaning contacts who were actually interested, amounted to an estimated 17. For a cold email campaign, especially in a sector as competitive as web development and website management, this is a very strong result.

What kind of results do you generally promise clients, and who is this model best suited for?

We usually promise to deliver at least 20 inquiries per month. Most clients fall in the range of 20 to 50 inquiries per month. Some international clients with a large target market can receive up to 300 leads per month.

This model works best for companies whose service or product is B2B and high value, and whose potential target market is sufficiently large, either geographically or in terms of the number of companies. It is very difficult to deliver strong results in fields where competition is extreme and differentiation is low, such as classical software development, web development, or SEO services. This does not mean it is impossible, but it is certainly more difficult, and the conversion rates are lower.