Nazar Mysholivskyi writes about ecommerce and how brands can become easier for AI assistants to find and recommend. Recent posts focus on practical checks for product visibility, including product-feed details and category-specific searches.
Nazar Mysholivskyi is most useful for making AI product discovery concrete and testable for ecommerce brands.
Best for
A content collaboration for B2B ecommerce software helping brands get products picked by AI assistants.
Not for
A consumer-facing beauty brand wanting shoppers to discover and buy through a creator post.
Key takeaways
Ask an assistant for a category without naming a brand, then read what comes back.
A feed missing sulphate-free may leave an assistant without information found on the product page.
Shopify's 5 August earnings call: Catalog-powered AI searches converted twice the rate of scraped data.
Check first: Confirm current audience results, follower roles, and partnership terms.
Main ideas
Test AI product discovery with an unbranded, specific customer question · Put product attributes where the assistant can read them · Replace broad AI talk with a question about a real category
Nazar Mysholivskyi writes about ecommerce and how product information affects what AI assistants surface and recommend.
Their most useful posts replace broad AI-strategy talk with product-level checks: ask an assistant for a category without naming a brand, see which products it returns, and examine whether the product feed contains the attributes in a shopper’s question.
They also write about customer experience, retail discovery, SaaS and manufacturing.
If a book, show, business or earlier body of work is relevant to your decision, ask Nazar about it; their posts don’t establish whether those are central to their public work.
Nazar’s clearest contribution is making AI product discovery concrete and testable.
They describe two unbranded beauty searches and report what ChatGPT returned, then explain why a feed that omits details such as “sulphate-free” may leave an assistant without information found on the product page.
The practical distinction is useful: appearing in a catalogue is not the same as being selected, and making a product findable does not guarantee a purchase.
They also turn observations into questions brands can use, such as whether they appear in an unbranded category search.
Earlier SaaS posts offer more general commentary, while manufacturing posts describe conversations about warehouse operations without sharing the answers.
Product-level examples are the strongest reason to follow their work, though they also write about beauty, ecommerce, SaaS and manufacturing.
Across the posts, Mysholivskyi’s clearest recurring idea is to make broad technology claims testable against a real product or business process.
Read more
In ecommerce, that means checking what an AI assistant can find and why; in manufacturing, asking how a warehouse is actually tracked. Earlier posts also discuss SaaS customer success and AI, but in more general terms.
Test AI product discovery with an unbranded, specific customer question
Nazar argues that brands should check whether their products appear when someone asks for a category, rather than assume that a catalogue listing means they will be recommended.
Read more
The product criteria in the question matter.
Nazar describes asking ChatGPT for sulphate-free shampoo for fine, oily hair and fragrance-free SPF moisturiser for sensitive, dry skin.
The shampoo answer showed five products before a pick, with buy buttons leading to retailers; the post says the assistant used criteria such as hair type and sulphate-free.
Try the same kind of question without naming a brand.
Replace broad AI talk with a question about a real category
Nazar criticizes discussion of AI for retail that never gets down to an actual product.
Read more
Their proposed test is whether a customer asking about a specific category would find the brand in the answer.
Nazar contrasts broad talk about “AI strategy” with asking whether a beauty brand appears when someone asks ChatGPT about shampoo, serum, or a moisturiser for dry skin. They use the same product-level focus in unbranded beauty searches.
Look for operational gaps with questions a team can answer
Nazar turns ecommerce concerns into checks about whether a business tracks the information it needs.
Read more
They developed the checks through three months of research; ask what evidence supports them before treating them as benchmarks.
Before eCommerce Expo, Nazar suggested checking whether ChatGPT names the brand in its category, what share of stockists saw the latest range launch, and how many customer questions on the last five TikTok or Instagram posts received replies.
Ask about the work itself, not just the technology around it
Nazar’s event posts favor concrete questions about products and operations over abstract technology discussion.
Read more
The approach extends beyond retail to manufacturing.
After MACH Exhibition, Nazar said they had walked all five halls and asked people how they actually track warehouse inventory. Before the event, they had singled out mid-market manufacturers without a dedicated IT department as a group they wanted to hear from.
The emphasis has moved from broad SaaS and AI claims to specific tests
Earlier posts describe customer success as important for SaaS retention and discuss AI in terms of prediction and automation.
Read more
Later ecommerce posts offer more concrete questions and examples.
An earlier SaaS post links customer success with retention and customer lifetime value; another describes AI’s potential for predicting trends and automating tasks. Later posts ask whether a product category appears in ChatGPT or whether product attributes are present in a feed.
The posts include earlier broad SaaS and AI discussion alongside later, more specific ecommerce tests; they do not establish a precise timeline for the change.
Answers the first question a newcomer has: does this person run their own tests or just repeat what others say?
Read more
This post shows the test.
Nazar ran two beauty searches as a customer would: one for sulphate-free shampoo for fine, oily hair, and another for fragrance-free SPF moisturiser for sensitive, dry skin.
Neither query named a brand.
ChatGPT returned five shampoo products before picking one; every buy button in that answer led to a retailer, not the brand.
The moisturiser answer returned three products and one pick.
ChatGPT chose according to the criteria in the question — hair type, oily scalp and sulphate-free.
The post adds a sourced fact: every Shopify store sits in ChatGPT’s catalogue automatically, so being listed is not the same as being selected.
For brands planning a campaign
A cheap, repeatable test a brand can watch unfold. For a collaboration on AI search visibility, run the same unbranded query on a client’s category and publish the gap between being in the catalogue and being the pick.
Explains why a well-written product page can be invisible to an assistant that reads only the feed.
Read more
Nazar quotes Harley Finkelstein from Shopify’s call on 5 August: AI searches powered by Catalog converted at twice the rate of searches using scraped data.
They explain the consequence: a beauty brand’s feed holds a title, a description and a category; the phrase ‘sulphate-free’ is absent if it appears only in a paragraph the assistant never opens.
The fix has two parts — put every attribute in a field the assistant can read, and write descriptions the way people ask.
Nazar gives an example from a UK agency where ‘comfy shorts’ was expanded to ‘elasticated waist, lightweight fabric’ before the search ran.
They close with a limit: this makes a product findable; it will not make AI buy for customers.
For brands planning a campaign
The most directly usable piece for a B2B ecommerce SaaS brief: it names the fix and sets a credible boundary on what AI visibility delivers. A collaboration could turn it into a feed audit with client examples rather than another general piece about AI search.
A compact example of Nazar turning a conference into research rather than networking: three questions that expose where an ecommerce brand is losing revenue.
Read more
Before eCommerce Expo at Excel London, Nazar posted three questions for brands.
Ask ChatGPT about your own category and check whether you or a competitor appears.
Divide the number of stockists who saw your last range launch by the number of stockists you have.
Scroll your last five TikTok or Instagram posts and count customer questions against replies.
Nazar says the questions came out of three months spent identifying where ecommerce revenue leaks.
They have not surveyed brands, so ask clients for their own scores rather than presenting a ‘most brands’ benchmark.
All three questions share one shape: nobody receives a weekly report on any of them.
For brands planning a campaign
A ready-made question set a brand can answer in a meeting. It could anchor a collaboration: run the three questions with a client, publish the answers and let the gaps make the argument.
Shows Nazar reporting a pattern from many conversations rather than one source, while keeping private meeting content private.
Read more
After two days at eCommerce Expo in London, Nazar reports that ‘AI presence’ came up in nearly every meeting, side conversation and talk.
They infer that customers already look for products through large language models, so products, services and catalogues need to be structured for a site AI can read without blocking it from the context.
Nazar also flags live shopping and TikTok’s growing role, describing TikTok as becoming product engine optimisation for ecommerce because people search there for specific items.
They thank named attendees and keep the substance of private meetings out of the post.
For brands planning a campaign
Evidence of a working read of the market, not a single-source claim. Supports a campaign linking AI visibility with TikTok discovery as two routes to the same customer.
A short, clear statement of the standard Nazar holds others to — and applies to themself — showing the angle a collaboration would take.
Read more
Nazar writes that the same note recurred across a year of AI talks: everyone spoke too generally about AI for retail, AI strategy and the future of shopping, with no product on the slide.
They admit nodding along to some of those talks themself.
Nazar has stopped asking brands about their AI strategy and now asks beauty brands one question: when a customer asks ChatGPT for your category, are you in the answer?
Their reasoning: nobody sells ‘retail’ — they sell shampoo, serum or a moisturiser for dry skin.
AI helps one category at a time, with real products in the answer.
For brands planning a campaign
Frames a collaboration around a narrow, product-level question rather than another AI strategy piece. It also works as a title or hook for a campaign asset.
Shows the method extending beyond ecommerce and what Nazar does with the answers, including what stays unpublished.
Read more
Back from MACH Exhibition in Birmingham, Nazar reports walking all five halls alone and having more candid conversations about warehouse operations in one day than in five months normally.
They thank named people for making time and share their standing question: ‘How do you actually track what's in your warehouse?’ — an operational question, not a technology one.
The answers fascinated them, but they did not publish them, so the post records the method and the room rather than the findings.
For brands planning a campaign
Confirms the approach works in industrial and mid-market manufacturing settings without a dedicated IT department, which matters if a brief reaches beyond beauty and retail. Ask the client which answers can be published; the post does not share them.
Retail technology, grounded in products and operations3
Earlier SaaS and customer-success themes3
AI-assisted product discovery2
Product feeds and findability1
Practical checks for ecommerce revenue gaps1
AI-assisted product discovery
Nazar tests product discovery by asking ChatGPT for a category without naming a brand, then examining which products appear and what details shape the answer.
Read more
Examples include shampoo for fine, oily hair and fragrance-free SPF for sensitive, dry skin.
Use customer-style questions and product criteria; inspect the products and answers returned.
For brands planning a campaign
A beauty or retail campaign could demonstrate a repeatable, category-specific visibility test without implying that catalogue inclusion guarantees recommendations.
Nazar argues that product attributes need to appear in readable feed fields, not only in page copy.
Read more
The example contrasts a feed that omits “sulphate-free” with customer searches that specify product needs.
Compare the product feed’s structured attributes with the criteria customers use in their searches.
For brands planning a campaign
Relevant for ecommerce software or feed-management brands: build a practical collaboration around checking attributes and descriptions against real customer queries.
At an ecommerce event, Nazar proposed three checks: whether an AI assistant names the brand, how many stockists see a new range, and whether customer questions on recent social posts receive replies.
Read more
They say these came from three months of research, but give no survey results.
Ask a concrete operational question, then identify whether the business has a way to track the answer.
Nazar says they have not surveyed brands and do not claim what most brands score.
For brands planning a campaign
A campaign could turn the checks into a diagnostic for ecommerce teams. Ask for industry-wide benchmarks; the post does not provide them.
Retail technology, grounded in products and operations
Nazar criticizes broad AI-for-retail discussion when it lacks specific products.
Read more
They also connect AI visibility with live shopping and TikTok, and ask manufacturers how they track warehouse inventory, particularly where mid-market firms lack dedicated IT staff.
Bring the discussion down to a product, customer search, or operational process.
For brands planning a campaign
Retail or operations campaigns may suit this grounded, question-led approach. Ask for detailed implementation case studies if those are important to the brief.
Other posts address customer success and retention in SaaS, and AI applications such as predictive analytics and task automation.
Read more
These posts are broad and less concrete than the later product-search examples.
The available posts show a shift in emphasis, not when or why it occurred.
For brands planning a campaign
A SaaS brief could connect product adoption or customer success with practical AI use. Ask for evidence of technical depth if the campaign requires it.
Nazar Mysholivskyi's posts do not mention sponsorships, affiliate relationships, ownership, employment, clients, investments or disclosed partnerships with named brands.
Read more
Shopify, ChatGPT, TikTok and eCommerce Expo appear only as examples or venues, not as stated commercial relationships.
Both analyse platform changes affecting how ecommerce brands reach customers.
Read more
Marko’s posts cover OpenAI putting ChatGPT Ads inside Shopify and HubSpot, and Sephora launching new products on TikTok Shop before its own site.
Nazar’s posts test what AI assistants return for product questions and how product feeds affect findability.
Both translate platform announcements into practical implications for sellers.
Marko focuses on advertising channels, creator commerce and TikTok Shop strategy; Nazar focuses on AI product discovery and the product feed an assistant reads.
Both work on ecommerce growth for brands and draw on event experience.
Read more
Vincenzo’s posts discuss Amazon Ads Q4 strategy, AMC and DSP, and EU expansion.
Nazar’s ask how AI assistants surface products and how manufacturers track operations.
Both use conference settings to push brands toward specific preparation rather than general advice.
Vincenzo’s material centres on Amazon advertising, retail media and cross-border expansion; Nazar’s centres on AI search visibility and product-data readiness.
Both attend manufacturing events and ask about the practical state of operations.
Read more
Kevin’s MACH2026 post asks whether management systems keep pace with changing products and services; Nazar’s MACH post asks manufacturers how they track warehouse inventory. Both favour concrete process questions over general technology talk.
Kevin’s work covers ISO management systems, energy management and process governance; Nazar’s covers ecommerce, AI product discovery and product feeds.
Nazar’s posts focus on practical ecommerce and AI product-discovery questions, including whether products appear in category searches and whether product feeds contain the details shoppers ask for.
Read more
That content may suit a campaign about AI visibility for retail brands.
Their profile lists 1,951 LinkedIn followers as of October 9, 2026.
Before assessing professional-audience fit, ask for current audience results and documentation of followers’ roles or industries.
Nazar Mysholivskyi’s posts do not set out partnership formats, rates or deliverables.
Read more
In a September 2026 post about eCommerce Expo at ExCeL London, they said they were open to discussing a partnership and wanted to meet representatives of brands and marketing companies working on ecommerce customer experience and acquisition.
A follow-up post about the event describes private meetings and recurring themes such as AI visibility and live shopping, but gives no terms.
Conference posts from MACH Exhibition in Birmingham in April 2026 follow a similar pattern: attending, asking manufacturers how they actually track warehouse stock and offering to meet for coffee.
The published work suggests possible collaboration topics: category checks that ask an AI assistant for a brand’s product category and report whether it appears; an exercise they describe running with beauty and ecommerce teams; and three questions brought to ExCeL about AI answers, stockist visibility of new ranges and unanswered customer comments.
These are topics they work on, not a defined offer.
No sponsorship policy, media kit, rate card or partnership contact page appears in the posts, and their LinkedIn profile is a general profile rather than a partnership route.
Ask directly about deliverables, timing, cost and how to get in touch.
A content collaboration for a B2B ecommerce software company that helps brands get their products picked by AI assistants through catalogue and feed tooling.
Read more
The goal: show beauty and retail brands how to move from being listed to being chosen.
This is the closest match to what Nazar Mysholivskyi already publishes.
The campaign's central test is one Nazar already runs in public: ask ChatGPT for a category without naming a brand, then read what comes back.
In the shampoo example, two customer-style queries produced five products and one pick in the shampoo answer, with every buy button pointing to a retailer rather than the brand.
The moisturiser answer returned three products and one pick, with the selection driven by criteria in the question — hair type, oily scalp, sulphate-free.
Nazar also supplies the mechanism the campaign would sell against: a quote from Shopify's 5 August earnings call that AI searches powered by Catalog converted at twice the rate of searches using scraped data.
They then explain that a beauty brand's feed holds a title, a description and a category, so a phrase like 'sulphate-free' never reaches the assistant if it only appears in page copy.
The two-part fix — put every attribute in a readable field and write descriptions the way people ask — gives the campaign a usable spine.
The UK agency example ('comfy shorts' expanded to 'elasticated waist, lightweight fabric') offers a concrete rewrite to imitate.
Nazar also sets a boundary that protects the client's credibility: this makes a product findable; it will not make AI buy for the customer.
Their working habits also suit a B2B campaign: three diagnostic questions taken to eCommerce Expo at ExCeL London (whether an AI assistant names the brand, how many stockists saw the last range launch, and how many customer questions went unanswered); openness to discussing a partnership with brand and marketing teams; and a published view that broad 'AI for retail' talks fail because no slide has a product on it.
Before agreeing terms, ask for audience breakdowns by industry, role and location, plus rates and paid-collaboration terms.
The profile shows 1,951 followers as of 9 October 2026, but no audience breakdown, rate card or media kit; no paid collaboration is disclosed in the posts.
When to choose a specialist
A consumer-facing beauty brand that wants shoppers to discover and buy a shampoo or skincare product through a creator post aimed at consumers is a poor match.
Read more
Nazar Mysholivskyi's posts address brand, ecommerce and marketing professionals — ecommerce heads and managers, brand representatives, marketing companies and manufacturers — not shoppers deciding what to put in a basket.
They include no consumer-facing work: no product reviews, routines, tutorials or demonstrations of products in use.
The AI-discovery argument would also work against a consumer campaign.
It critiques how brands manage their feeds and product data — a message for the people who own those systems, not those buying shampoo.
The only audience figure available is 1,951 followers as of 9 October 2026, with no data on who they are, so consumer reach cannot be assessed.
A mid-market manufacturing or warehouse software campaign looks more plausible than it is.
Two MACH Exhibition posts from April 2026 show Nazar attending in Birmingham, walking all five halls and asking manufacturers, 'How do you actually track what's in your warehouse?' But the answers stay unpublished, and there is no manufacturing content series behind them.
A manufacturer-facing campaign would start a new category rather than build on demonstrated output.
A worked campaign example
A worked angle for an ecommerce SaaS campaign, built on Nazar's existing method rather than a new format.
Step one: establish the baseline.
Run the unbranded category test on the client's own products: enter a customer-style query without a brand name across three to five categories the client sells.
Record which products appear, which one gets picked, which details in the question drove the pick, and where the buy buttons point.
The shampoo test is the model — five products, one pick, retailer links instead of brand links.
Step two: audit the feed.
Compare the attributes stored in the client's product feed — title, description, category — with the criteria shoppers actually use in those queries.
Look for the gap Nazar describes: a phrase such as 'sulphate-free' appears in a page paragraph the assistant never opens, so the feed carries it nowhere.
List the missing attributes product by product, and note that being in the AI catalogue is automatic for a Shopify store; it is not the same as being selected.
Step three: produce one worked rewrite.
Take a single product and make both parts of the fix: move every attribute into a field an assistant can read, then rewrite the description the way people ask.
The 'comfy shorts' to 'elasticated waist, lightweight fabric' expansion is the working example — customer wording turned into structured criteria before the search runs.
Step four: rerun the queries and publish the difference.
Show what changed and what did not, stating the limit plainly: this improves findability; it does not make an assistant buy on the customer's behalf.
Close with the three questions from eCommerce Expo as a checklist brands can use themselves: are you in the answer, how many stockists saw your last launch, and how many customer questions went unanswered?
The client supplies a copy of the product feed, the categories to test and one or two product pages for the rewrite.
Nazar supplies the queries, recorded answers, feed gap list and written analysis, delivered as a LinkedIn series or a single article, with an event session as an optional extension along the ExCeL pattern.
Before briefing, confirm rates, deliverable count, timeline and partnership terms.
No rate card, sponsor policy or published commercial terms appear in the posts, and the partnership terms Nazar invited discussion about in September 2026 were left open.
Choose Kevin Jordan when the campaign shifts from AI product discovery to industrial process control: helping manufacturers, energy managers or UK mid-market operations clients tighten management systems, align processes with new products, or treat energy as a board-level managed risk.
Read more
Their posts show ISO 50001 energy management work and 25+ years across ISO management systems, plus an invitation to MACH2026 to discuss management system health, quality management and continuous improvement. For a client selling energy-management, ISO or manufacturing-operations software to UK industrial buyers, that is a closer match than an AI search visibility collaboration.
The tradeoff is that Nazar Mysholivskyi's posts do not address AI assistants, product feeds, ecommerce catalogues, beauty or retail.
A campaign showing brands how to move from being listed to being chosen by ChatGPT would lose the specific category-query test, the Shopify feed mechanism and the beauty examples.
Kevin Jordan brings operational and standards credibility, but the buyer would need a different content foundation for AI-discovery content.
Marko Stupar is the stronger alternative when paid AI placement and social commerce operations take priority over feed hygiene.
Read more
Their posts cover the 16 September launch of ChatGPT Ads inside Shopify and HubSpot, the Shopify app rollout and 23 September international date, Sponsored Agents, attribution through HubSpot URL tagging and Shopify catalogue and measurement, and the warning that product data becomes the creative.
They also cover Sephora launching new products on TikTok Shop before its own site, with creator teasers, LIVE drops and exclusivity windows, and draw out what that means for creator retainers, average order value and premium SKUs.
For a client seeking a content collaboration on ChatGPT Ads setup, Shopify or HubSpot measurement, or TikTok Shop launch strategy for beauty and retail brands, Marko is a close alternative.
The tradeoff is that Marko Stupar's posts focus on platform news and ad operations.
They do not show the same repeated, public, unbranded category test across AI assistants or a step-by-step product-feed attribute audit of the kind Nazar publishes.
If the buyer's core promise is to show brands how to stop losing attributes in page copy and get picked in AI answers, Marko's ad-buying and TikTok Shop angle is adjacent but less specific.
Choose Vincenzo Toscano when the campaign is Amazon-first.
Read more
Their posts show keynote speaking at the Amazon Ads EU Business Accelerator 2026 and an Amazon EU Accelerate meetup at the London Chamber of Commerce, with a Q4 message built around AMC and DSP to understand who is buying, build audiences before peak CPC spikes, and retarget into January.
They also cover European and US expansion, FBA Select and AWD logistics changes, and AI moving from buzzword to application in advertising and operations.
For a B2B ecommerce SaaS client whose beauty or retail customers sell on Amazon and want marketplace advertising, measurement or expansion content, Vincenzo is a better fit than a general AI-search visibility creator.
The tradeoff is that the Amazon marketplace is the frame, not AI assistants broadly.
The posts do not show unbranded ChatGPT category tests, product-feed attribute audits for AI answers, or Shopify and TikTok Shop buying journeys.
A collaboration centred on getting products picked by ChatGPT and other large language model assistants would need to extend beyond Vincenzo's Amazon Ads and Q4 peak expertise.
Nazar Mysholivskyi's LinkedIn posts cover early commentary on SaaS and customer success and later work on ecommerce AI discovery, product feeds, live shopping, and trade events.
The later posts include two unbranded beauty searches run against ChatGPT, a post quoting Shopify's 5 August earnings call on catalogue-powered AI search, the three diagnostic questions taken into eCommerce Expo at ExCeL London, and the two MACH Exhibition posts from Birmingham on 20-21 April 2026.
The LinkedIn account showed 1,951 followers on 9 October 2026.
Scope and gaps.
Not every post displays an exact publication date, so ordering follows the events each post refers to.
Nazar Mysholivskyi's posts do not say whether there are signature books, shows or businesses, establish employment or audience composition, or give commercial terms such as rates or sponsorship terms, and no results from the campaigns and events described are published.
Posts that mention Shopify, ChatGPT, TikTok or trade events treat them as examples or venues, not as stated commercial relationships.
Two posts each are included for comparator creators.
Topic labels, similar-creator comparisons, and campaign-fit judgments reflect the posts, with comparisons to the strongest competing creator pages.
Recommendations are limited to what Nazar Mysholivskyi's posts show; open identity and commercial questions remain unanswered rather than being filled by inference.
