Why the Biggest Online Stores Are the Most Underserved

At Balkan eCommerce Summit 2026, Krasimir Krastev from HighSky.ai explored one of the biggest paradoxes in modern eCommerce:

the larger and more successful an online retailer becomes, the harder it often is to adapt to innovation and fully utilize new technologies. The lecture focused on the hidden operational and customer experience challenges large online stores face, particularly in the era of AI, mobile commerce, and shrinking consumer attention spans.

Large Retailers Often Struggle Most with Innovation

The presentation began with an observation based on years of digital transformation experience:

the bigger the company, the slower it tends to adapt to change.

According to the speaker, large retailers usually:

• invest heavily in massive catalogs
• complex backend systems
• detailed product information
• and operational infrastructure,

yet these same strengths often become:

bottlenecks that slow innovation.

Rather than building another chatbot or generic AI tool,
HighSky.ai chose to focus specifically on solving:

the operational and customer experience challenges of large eCommerce businesses.

The “Catalog Trap” Is One of the Biggest Problems in eCommerce

A major concept introduced during the lecture was:

the catalog trap.

Large retailers may manage:

• 20,000
• 30,000
• or even more SKUs,

while investing enormous resources into:

• product descriptions
• images
• comparisons
• specifications
• and categorization.

However,
the speaker emphasized a key problem:

customers barely interact with most of that information.

According to the lecture,
an average customer only explores:
around 10 products,
despite stores having tens of thousands available.

This creates what the speaker described as:

the “catalog gap” –

where the retailer’s greatest investment becomes one of its biggest barriers to conversion.

Gen Z Attention Spans Are Changing eCommerce

Another central topic was:

the collapse of consumer attention spans.

The speaker compared historical media behavior:

• 10-minute attention spans in the 1980s and 1990s
• 30-second spans around 2021
• and now approximately:

8 seconds.

This creates enormous pressure for online stores,
especially large retailers with complex catalogs.

The key challenge becomes:
how to:

• capture attention
• understand intent
• recommend products
• and convert customers
  within:

a few seconds.

Most Shopping Happens on Small Mobile Screens

The lecture also highlighted the growing mismatch between:
how retailers design stores
and how customers actually shop.

While many online stores still optimize experiences around:

• large desktop layouts
• long descriptions
• and detailed comparisons,

the reality is that:

most users shop on mobile devices.

The speaker emphasized that:
large retailers must now communicate:

• product value
• recommendations
• and purchasing confidence
  on:

a 6-inch screen within seconds.

Customers No Longer Want to Search – They Want Guidance

One of the strongest themes throughout the presentation was:

modern shoppers do not want to search manually anymore.

Instead of typing simple product queries like:
“TV”,
customers increasingly express:

intentions.

Examples included:

• “I need a gift for a 6-year-old”
• “I want something for my niece”
• “What should I buy?”

The speaker argued that:
traditional search systems fail in these situations,
because they are built around:
keywords,
not customer intent.

This is where AI-powered conversational search becomes critical.

AI Must Understand Intent, Not Just Keywords

According to the lecture,
the future of eCommerce search is:

intention-based commerce.

Retailers need systems capable of:

• understanding natural language
• interpreting customer goals
• and recommending relevant products instantly.

The search bar itself becomes:

one of the most important conversion tools in modern eCommerce.

AI Shopping Assistants Can Replace Complex Manual Guidance

The presentation also explored:

AI-driven shopping assistance.

Modern consumers increasingly expect:
guided shopping experiences,
similar to receiving help from an in-store expert.

The challenge for large online stores is that:
human agents cannot realistically:

• master every category
• answer every product question
• or spend long periods assisting each customer.

AI assistants solve this by:

• asking follow-up questions
• understanding context
• recommending complementary products
• and guiding shoppers throughout the journey.

Assisted Selling Dramatically Increases Order Value

One of the most striking examples shared during the lecture involved:
an AI-assisted shopping session where:
a customer initially intended to purchase:

a €15 product,

but after a:

45-minute AI conversation,

the final order reached:

€240 across 17 SKUs.

The speaker emphasized that:
no human consultant would realistically dedicate that amount of time to a low-value customer inquiry,
especially at scale.

This demonstrated the power of:

AI-driven assisted selling

for increasing:

• basket size
• product discovery
• and customer engagement.

Backend Complexity Is Also a Major Barrier

The lecture also addressed operational challenges behind the scenes.

Large retailers operate:

• complex databases
• sophisticated warehousing systems
• intricate logistics
• and layered backend platforms.

Even if front-end experiences improve,
these backend systems often become:

organizational bottlenecks

that slow:

• fulfillment
• customer support
• and innovation speed.

AI Is Most Valuable for Established Retailers

The presentation concluded with the argument that:
large retailers actually have:

the most to gain from AI adoption,

because:

• they already have scale
• traffic
• large catalogs
• and operational maturity.

However,
they are also the businesses most likely to:

• move slowly
• overcomplicate decisions
• and struggle with implementation.

According to the speaker,
AI should not simply automate tasks,
but instead:

bridge the gap between massive catalogs and shrinking customer attention spans.

Key Takeaways

The lecture demonstrated that:
large online retailers often adapt slower to innovation despite having the most resources
the complexity that makes big retailers successful also creates operational bottlenecks
massive product catalogs frequently overwhelm customers instead of helping them
most shoppers interact with only a tiny fraction of available products
Gen Z attention spans are shrinking dramatically, increasing pressure on conversion speed
modern shoppers increasingly shop on mobile devices with limited screen space
customers no longer want manual search — they expect guided experiences
traditional keyword-based search is becoming insufficient
AI-powered conversational search helps retailers understand customer intent
modern eCommerce requires intention-based product discovery
AI shopping assistants can provide scalable personalized guidance
AI-assisted selling significantly increases basket size and product discovery
large retailers struggle to scale expert-level customer assistance with human teams alone
backend complexity often slows innovation and operational agility
the future of eCommerce depends on combining AI assistance with simplified customer experiences
the retailers that successfully connect AI, intent understanding, and guided shopping will gain a major competitive advantage.