SEO, smart messaging in the age of AI
Most startup content sounds the same. Lots of noise, not much signal.
Adomas Šulcas, cofounder and COO at Growth Bite, runs a digital marketing agency that lives in the intersection of philosophy, SEO and practical communication for SaaS and tech companies. In this conversation he explains why clear thinking is a growth lever, why most early stage messaging is off, and how to use AI without turning your brand into generic soup.
Below is the full story.
Journey to Growth Bite
Adomas did not follow a straight path into marketing.
He tried an HR agency, did sales for about 1 and a half years, then switched into freelancing as a copywriter while finishing his studies. One of his freelance clients was an online poker school. He learned online poker, played for 2 to 3 years, and wrote on the side.
Then a friend’s LinkedIn post pulled him into tech. He joined Oxylabs as a content manager, met his cofounder and now fiance Rasa there, and the Growth Bite story slowly started.
For a while they both freelanced. Rasa handled SEO, Adomas handled copy. They kept sending 2 separate invoices to the same clients, until they realised they should just form 1 agency and work as a single team.
Growth Bite was born: a senior first boutique agency focused on SEO, content and strategy for SaaS and other B2B companies.
How philosophy shapes better business communication
Alongside all this, Adomas kept studying. Bachelor in philosophy, then master, now finishing a PhD.
He never treated university as a direct career tool. Instead, philosophy trained 2 core skills that now show up in his work every day.
- He never stops at the first idea
- When he sees a possible solution, his brain automatically generates 3 to 5 alternatives. That makes strategy and messaging conversations more robust. You do not just ask “what should we do” but “which of these options is best and why”.
- He always brings arguments, not just opinions
- In philosophy you are forced to justify every claim. That carried into business. When Adomas proposes something in a company, he comes with reasons and tradeoffs. For him this is normal. For many teams it is a huge upgrade in communication quality.
Philosophy did not make him abstract. It made him concrete. It trained him to think from first principles in any situation:
- What problem are we solving
- For whom
- How do we solve it better
- What does it cost us
- How does this become a real business, not just a cool idea
Everything else is decoration.
The biggest mistakes early stage startups make
Adomas keeps seeing the same patterns when founders come asking for SEO or content help.
1. Choosing a channel just because “everyone does it”
A startup will say “we need SEO” because another agency promised great results. Growth Bite runs the numbers, checks keyword volume and competition, and sometimes finds there are maybe 500 monthly searches in total.
In those cases they say no.
Not “no to a retainer”, but “no, do not spend money on SEO, put that budget into events, sales or partnerships”. He even gives examples where a client had maybe 10 possible customers worldwide. For that kind of business, conferences matter, not TikTok.
The lesson: you do not start with tactics. You start with market size, buyer type and realistic channel potential.
2. Messaging about how cool the product is
Especially in the AI wave, Adomas sees many founders leading with “look what we built in a weekend” or “this is so cool”.
Nobody pays you for “cool”.
Companies pay you because:
- You solve a painful problem nobody else solves
- Or you solve it meaningfully better
Unless you are a lifestyle brand, no user cares about your tech stack or weekend hack. They care about outcomes.
Cool might help with investors for a while. For customers, problem clarity wins.
SEO in the age of AI: what actually changed
A lot of people are loudly declaring SEO is dead.
Adomas disagrees.
From what he sees across clients, SEO still brings tens of times more traffic than all AI search tools combined. What changed is the type of traffic.
Many purely informational queries are moving into AI chat. Those visits often did not convert anyway. The traffic that actually brings pipeline still comes from classic search.
AI models also frequently call external search to answer questions cheaply. That means search engines and SEO remain part of the stack for a long time.
There is early talk about “GEO” generative engine optimisation, and some odd tricks do seem to influence AI answer selection, like over emphasising relevancy with explicit phrases. But many of those tricks conflict with Google’s rules. You can win in 1 place and get penalised in another.
So his current stance is simple:
- SEO is still very much alive
- Conversion oriented traffic still depends on it
- Do not torch your search presence for short term AI games
How a startup should choose content channels
Founders have limited resources. They cannot do everything.
Adomas uses a simple mental model.
The bigger and more complex your target customer, the more your growth relies on relationships, events and direct contact, and the less you can lean on broad digital channels.
Rough guide:
- B2C products with wide appeal
Almost every channel can work. SEO, paid search, social video, email.
- Small and medium B2B clients
SEO and paid search remain strong. Some social platforms start to drop off. TikTok or Instagram may still work, but you need sharper positioning.
- Enterprise accounts with thousands of employees
SEO becomes more of a long game. Buying often happens via events, networks, outbound and partnerships. Paid search gets very expensive. LinkedIn becomes the only social platform that really matters.
- Ultra niche markets with maybe 10 to 20 possible customers
You do not need most marketing channels. You need a solid site as a trust asset, then trade shows, direct outreach and relationships.
On top of that he adds 1 rule for almost every founder:
Once you pick a channel and a strategy that is “good enough”
stick with it long enough to get signal
Most teams change direction every 2 months, then complain nothing works. The average strategy applied consistently beats the perfect strategy you restart every quarter.
Platform native content vs repurposing
In a perfect world you would create content shaped for each platform.
Reality: almost nobody has those resources.
Adomas still insists that platform native content works better. Cultures differ wildly across platforms. What looks normal on LinkedIn gets mocked on Reddit. There is literally a subreddit dedicated to roasting corporate posts.
If you blindly repurpose, people will smell it.
At the same time, he knows founders and small teams cannot write 5 different pieces for 5 channels every day. So his practical approach:
- Aim for native where it really matters, usually your primary platform
- Repurpose smartly, but always adapt tone and format so it fits the local culture
How to use AI in content without losing your voice
Growth Bite does not paste full AI outputs into client blogs.
They have seen sites lose all organic traffic in a single day. Recovering from that penalty is expensive and slow. It is fine if you are running a throwaway project. It is not fine if you are building a real company.
Instead they use AI in specific ways.
- Idea support
- Writers paste the brief into a model like Claude, then start writing themselves. They glance at AI suggestions for angles, subtopics or examples. They may borrow a phrase or two, but never paste full paragraphs.
- Fact checking and sanity checks
- For technical topics, Adomas will paste tricky sections into AI and ask “is this correct” and “show sources”. Then he manually verifies. AI becomes a helper, not the author.
- Editing small pieces
- For emails, generic replies and some internal communication, AI can help with phrasing and tone. For long form content that moves money, he prefers human control.
He also notes early research suggesting that over reliance on AI can flatten your style. Your writing starts to sound more like the model than like you. For a brand voice that is a real risk.
Trends in SEO and communication
Beyond AI itself, Adomas sees a cultural trend he calls “work slop”.
Someone is too lazy to write an email, so they ask AI. The recipient is too lazy to read, so they ask AI for a summary. Then they reply with another AI draft. At some point there are 3 bots talking and 0 humans paying attention.
He is not against tools. He is against turning your entire job into prompt forwarding.
You still need humans who actually understand:
- The product
- The market
- The nuances of what was agreed
If nobody really knows anything because every part of communication is outsourced to a model, your business will eventually drift.
A simple principle for clear communication at work
Adomas has 1 very practical habit for everyday communication.
Study how developers write bug reports.
A clean bug report usually says:
- What I was trying to do
- What I expected to happen
- What actually happened
- What I need help with
If you message a colleague like that on Slack, they immediately know what you want and how to respond. Compare that with vague pings like “hey” or “quick question” that go nowhere for 2 hours.
This structure also works for any explanation: marketing brief, product spec, strategy doc. Most communication problems are simply missing context.
How to get dramatically better at writing
Adomas gives 1 uncomfortable but powerful exercise for anyone who wants to write better content.
- Find a person you trust who is a strong writer
- Give them your drafts
- Ask them to be ruthless, leave comments everywhere and explain why they change things
Your ego will hate it. Your craft will grow faster in 6 months than in 5 years of writing alone.
For founders and marketers, this is often the difference between “ok content that fills a page” and “sharp content people actually want to read”.
Content that cuts through the noise
If someone asks him how to write content that stands out, his honest answer is not sexy.
Start with the boring fundamentals.
- Cover the core questions your buyers actually search
- Explain things clearly
- Build out topic foundations before chasing fancy angles
- When you have the basics in place and working, then add original research, opinion pieces and data projects
He points to Ahrefs as an example. They publish in depth data articles with unique research. It works incredibly well for them. But they have a large team, strong budget and years of base content behind that.
Most startups try to jump straight into “thought leadership” without first covering the simple questions that actually drive pipeline.
Do the hard and boring work first. Fancy will come later.
