Second brand for a service firm: when and how to run it
A second brand makes sense for a service firm when a new service sells to a different buyer. What stays separate, what is shared and how to measure each brand.
Should a service firm sell a new service under a second brand? Usually not: by default you stay with one brand. You start a second one when the new service sells to a different buyer than your current one, meaning a person in a different role who speaks a different language and to whom your current brand says nothing, and when you can afford a separate website, content and proof. If you go ahead, each brand has its own services, ICP (ideal customer profile), proof, channels and results, while the firm, the team and the clients stay shared. You offer a client of one brand the other brand’s service in a conversation, not through a shared website.
In this post I’ll show how to tell whether a second brand makes sense, what to separate and what to keep shared, and when a second brand does harm.
What a second brand is
A second brand is a separate name, website and service line of the same firm, with the same team and the same client base. It doesn’t have to be a new company or a new team: what changes is how the firm presents itself to a specific buyer.
In 2000, David Aaker and Erich Joachimsthaler described the brand relationship spectrum, the ways the brands of one company relate to each other (overview in Branding Strategy Insider, summary in WARC). The examples in the Branding Strategy Insider overview are large corporations, but in my view the four models also name the choice a small service firm faces well:
| Model | How it works | In a service firm |
|---|---|---|
| Branded house | one master brand, services have only descriptive names | “Firm X: audits”, “Firm X: implementation” |
| Sub-brand | a brand connected to the master brand that takes on its associations | “X Analytics” next to “X” |
| Endorsed brand | an independent brand backed by the firm’s brand | “New name, part of Firm X” |
| House of brands | independent brands, each aimed at its own market | two names and two websites that don’t mention each other |
According to the WARC summary, a house of brands lets a firm position each brand clearly and win in niches. This post is mainly about the last two models, because that’s where the questions about separate content, proof and results come up.
A second brand makes sense when someone else is buying
In the 2024 Branding Strategy Insider overview, I read that Aaker and Joachimsthaler suggest the branded house as the default architecture and consider an alternative after four questions:
- Does the master brand contribute to the offering?
- Will the master brand be strengthened by association with the new offering?
- Is there a compelling need for a separate brand?
- Will the business support a new brand name?
In a service firm, the buyer usually answers the third question. In my view, a separate brand makes sense when the new service is bought by someone other than the person who buys your current one: someone in a different role, with a different vocabulary, comparing you with different alternatives. If your brand is known for campaigns for online stores, an IT director at a factory looking for systems maintenance won’t see a supplier for them in it. If the same person buys, the new service usually stays under your current name.
Promethean Research writes in its 2025 report on digital agencies that, with specialization now the norm, the edge comes from how clearly and credibly you define it, and generic claims won’t cut it (report, p. 45). In my view, when one brand serves two different buyers, its description usually becomes generic, because it has to fit both. I wrote about choosing a niche in the post on agency specialization.
What is separate for each brand and what is shared
I recommend writing this down before the second website goes up. Without such a list, the team starts mixing the services, examples and results of both brands.
| Separate for each brand | Shared by the firm |
|---|---|
| services | the firm (company, invoices, contracts) |
| ICP and personas (descriptions of the people who buy) | the team |
| proof: projects, results, references | clients (one client record in the CRM and one person who manages the client, even when they buy from both brands) |
| channels: website, LinkedIn, newsletter | competitor notes (one per competitor, even when it competes with both brands) |
| positioning | |
| a list of what the brand must never claim | |
| marketing and sales results |
The “must never claim” list protects you from promises that fit one brand but not the other. For example, a monthly reporting brand shouldn’t suggest that it prepares audited statements.
One piece of content serves one brand
Every article, campaign and proposal belongs to one brand and talks about that brand’s service to that brand’s buyer. Content “for both brands” talks to two buyers at once, in their vocabularies and about their problems in turn, so each of them spends half the time reading about someone else. The second brand has its own content plan, its own topics and its own channels, even if the same team writes them.
In practice: before the team writes a LinkedIn post, it decides which brand’s profile it will go out from, and only then chooses that brand’s service and audience.
You win a client of one brand for the other in a conversation
In my view, the biggest advantage of two brands in one firm is a client who buys from the first brand and has a problem the second one solves. That client is a reason to talk about expanding the relationship. The person who manages the client and knows their company sees it, not the website analytics.
The brands’ websites don’t have to mention each other. You make the connection in conversations, client by client: the person who manages the client knows there’s someone there in the role of the second brand’s buyer and arranges a conversation with them. You show proof from a first-brand project in a second-brand proposal only if the client has agreed to its use.
You measure results separately for each brand
A number “for the whole firm” doesn’t tell you which brand works. If inquiries for the first brand grow and the second brand has none, the total still grows, even though the second brand is going nowhere. That’s why you count website traffic, inquiries, sales conversations and won proposals separately for each brand and assess them separately. Only the firm’s revenue and revenue from a client who buys from both brands are shared.
Example: a software house with a second brand
This is an illustration, not a real firm: take a software house that has been building online stores for mid-sized retail chains for years, maintains them after launch, and now wants to sell systems maintenance and security to companies with their own manufacturing and logistics systems.
| Field | Brand 1: online stores | Brand 2: systems maintenance |
|---|---|---|
| Services | store design and build, store maintenance after launch | ongoing maintenance, updates, security monitoring |
| ICP and persona | retail chain, e-commerce manager | company with its own manufacturing or logistics, IT director |
| Proof | launched stores | maintenance of brand 1 stores’ integrations with clients’ warehouse systems (with clients’ permission) |
| Channels | brand 1 website, brand 1 LinkedIn, store case studies | brand 2 website, brand 2 LinkedIn, newsletter for IT departments |
| Positioning | stores for retail chains | systems maintenance for manufacturing and logistics |
| Never claims | that it maintains manufacturing and warehouse systems | that it designs online stores |
| Results | brand 1 inquiries and won proposals | brand 2 inquiries and won proposals |
Shared: the firm, the development team, client records and competitor notes. If a retail chain from brand 1 has its own distribution center with warehouse systems, meaning it fits brand 2’s ICP, the person who manages that client arranges a conversation with its IT director about brand 2’s service.
When a second brand does harm
A second brand does harm when:
- the new service is bought by the same person as the current one, so the brand only splits their attention,
- you don’t have time to run a second channel, and the second website stops changing after a few months,
- the second brand has no proof, and the only thing that sets it apart is its name.
When a brand stops making sense, you can sunset it: you stop creating new content and proposals in it, but you keep its history, projects and proof.
Frequently asked questions
Does a second brand have to be a separate company?
No. A second brand can be a separate name and website of the same firm, with the same team. Check legal matters, such as protecting the name, with a lawyer.
Should the two brands’ websites link to each other?
They don’t have to. If the second brand’s buyer would see the connection with the first brand as a minus, it’s better for the websites not to mention each other and for the person who manages the client to make the connection. If the firm’s brand strengthens the new brand, choose the endorsed model and say so openly.
How many brands can a small firm handle?
As many as the team runs regularly with the full set: channels, content, proof and a person responsible. In my view, a 10–50 person firm rarely handles more than two.
Summary
If you have a new service, start by asking who will buy it: if it’s the same person as before, stay with one brand. If it’s someone else and you can afford to run a second brand separately, write down what will be separate for each brand and what will be shared, and set separate results before the second website goes up. In our open repo (files with knowledge about the firm that an AI agent such as Claude Code uses), a firm with two brands keeps each one in a separate folder with its services, ICP, proof and channels, while client records stay shared (open source).
Every method on this blog also runs as a skill in the open-source B2BForce repo. About the repo