# Di Lyngholm

Di Lyngholm is a retail e-commerce product and growth leader focused on personalization. Lyngholm says they joined Malachyte as VP of Product and Growth after a decade working across brand, web experience, CRM, and operations at FUN.com and HalloweenCostumes.com.

- Profile: https://b2binfluencer.marketing/creators/di-lyngholm/
- LinkedIn: https://www.linkedin.com/in/dilyngholm
- LinkedIn followers: 2,401
- Topics: artificial-intelligence, ecommerce, ai-adoption, customer-research, retail-technology
- Based on 13 dated posts, latest 2026-09-30

## Summary

Di Lyngholm is most useful for operator-side retail personalization, site search, and AI adoption insights.

- Best for: A retail technology vendor or e-commerce team seeking practitioner-led explainers on search and personalization.
- Not for: A direct-to-consumer brand seeking sponsored lifestyle posts, discount codes, and broad consumer reach.
- Check first: Confirm whether they can provide audience demographics, engagement figures, and any documented sponsorship format.

### Takeaways

- Zero-result searches split into findability, assortment, and intent problems, each needing a different fix.
- One focused AI bet on a photo studio pointed to saving 1,800 hours a year.
- Informal customer conversations can reveal frustrations surveys miss, though both methods matter.

## Main ideas

- **Judge personalization by revenue per visitor, not features**: Di Lyngholm says e-commerce personalization must show a measurable revenue effect and be validated in a real pilot. They tie Malachyte's product to first-click adaptation without login or history, and cite a 31% revenue-per-visitor lift with FUN.com. In a hiring post, they define the solutions engineer role around technical demos, pilot setup and data validation that makes results “defensible, not just directional.” Their argument: brand operators have seen the downstream mess of bad software promises, so claims need proof before rollout.
- **Pick one AI use case and go deep**: At a Minneapolis eCom Summit panel, Di Lyngholm urged teams to stop tip-toeing around AI and go all in on one high-leverage use case rather than spread experiments across many options. A focused bet, they argued, produces measurable operational savings faster. They also warn buyers about “LLM wrappers” charging premium prices for custom prompts around existing models, and suggest evaluating what strong in-house prompting could achieve before signing a contract.
- **Treat zero-results searches as a revenue-recovery opportunity**: Di Lyngholm calls the zero-results search page an overlooked revenue leak: on-site searchers reveal exact intent in their own words and convert at much higher rates than browse traffic, but leave when they hit a blank wall. They group null searches into three operational buckets: findability gaps, where products exist but search rules, synonyms or metadata fail; assortment gaps, where shoppers ask for inventory the retailer does not carry; and intent mismatches between shoppers' phrasing and site taxonomy. Each bucket needs a different fix, and they say the breakdown can usually be diagnosed in about 30 minutes from the query log.
- **Operator experience should change how retail tech is sold**: Di Lyngholm says running the brand side changed how they approach vendor booths. They favor less theater and more architecture: operators want to know how a product works with legacy systems and whether it moves shopper intent without breaking workflows. Shared practitioner context beats sales scripts, with conversations about margin compression and tech debt instead of canned pitches. They also say attendees should be treated as peers looking for breathing room, not leads to trap. In their view, people who have lived with bad software promises ask different questions before a demo.
- **Protect budget for unprompted customer conversations**: Di Lyngholm argues that customer research budgets may be better spent on dinner reservations than on some formal studies. Surveys capture what people think they should say; informal conversations reveal what actually annoys them. They say both matter, but on a tight budget they would protect the conversation over the survey people ignore in their inbox. The most useful insight of the week, they say, came from a customer casually complaining about checkout over lunch, unprompted, because nobody told them they were in a study.

## Work to read

- [Why “no results found” is one of the most expensive pages on an e-commerce site](https://www.linkedin.com/posts/dilyngholm_why-no-results-found-is-one-of-the-most-activity-7498786475367546881-3dUf)
- [Talking a room out of overspending on AI: one focused bet beats a test matrix](https://www.linkedin.com/posts/dilyngholm_ecommerce-ai-retailtech-activity-7462857253801332736-Wblu)
- [Ten years at FUN.com and HalloweenCostumes.com, then VP of Product and Growth at Malachyte](https://www.linkedin.com/posts/dilyngholm_newchapter-productandgrowth-retailtech-activity-7451948923663073280-0Svr)
- [What changes when a retail tech booth is staffed by an operator](https://www.linkedin.com/posts/dilyngholm_retailtech-ecommerce-productstrategy-activity-7498733608749371392-oskw)
- [Customer research versus dinner reservations](https://www.linkedin.com/posts/dilyngholm_ecommerce-retailtech-retailstrategy-activity-7483143373755973633-pmd7)
- [Building the sales engineering side of a personalization company](https://www.linkedin.com/posts/dilyngholm_hiring-solutionsengineering-ecommerce-activity-7483534458529263619-I09B)

## Brand relationships

- **Malachyte** (employment)
- **FUN.com** (employment)
- **HalloweenCostumes.com** (employment)
- **Attentive** (mention)
- **Claude** (mention)

## Partnerships

Di Lyngholm's posts show a full-time operator role rather than a creator offering: they say they joined Malachyte as VP of Product and Growth in April 2026, and their public activity since then is largely company-facing — announcing a Malachyte solutions-engineering hire, hosting roundtables and a social event at eTail Boston, and presenting company results such as a 31% revenue-per-visitor lift at FUN.com. On a May 2026 conference panel, they recommended tools including Claude, Attentive, and their own employer, Malachyte — vendor advocacy from a leadership position, not a disclosed sponsorship. For a brand buyer, the practical route is a commercial or partnership conversation through Malachyte, not a creator sponsorship booking. The posts provide no inquiry address, pricing, deliverables list, disclosure practice, or sponsorship terms; ask Malachyte about the available format, exclusivity, and usage rights.

## Alternatives

- Naga Vara Prasad S: When the campaign shifts from on-site search diagnostics and personalization teardowns to conversational or voice AI in retail operations. Naga's posts offer a vendor-side view of AI personalization and loyalty, with specific deployment examples: a KFC Dubai drive-thru assistant built with Sodaclick and Intel, where 75% of customers accepted upsell suggestions, and Arabic/English switching. They also outline a broad retail cost-reduction framework covering inventory, automated service, supply chains and personalized shopping journeys. For a vendor seeking a speaker on AI adoption across the whole retail operation rather than just site search, these posts are a closer fit.
- Richard Newton: When the campaign shifts to external organic search rather than internal site search: product and collection pages that answer buyer questions, structured data, Google Discover visibility and SEO-led revenue. Newton's posts tackle a specific missed-demand problem: structured data cannot make a vague product page useful, and collection pages must match how shoppers search after store discovery, with examples such as linen dresses and sustainable occasionwear. If the vendor wants to demonstrate expertise across external and on-site search, or build a content asset around category- and product-page improvements, Newton is a closer fit. The posts also offer a direct service route through a free audit, which may suit a vendor seeking a clear conversion path.
- Abhijeet Mishra: When the campaign shifts to B2B commerce and distribution—buyer portals, fragmented data across systems, and AI agents for procurement, finance, sales reps and dealers—rather than retail consumer e-commerce. Mishra's posts use FoodLink, a B2B food and beverage distribution portal, as a case study, citing operational metrics such as 18% fewer order errors after moving from phone and fax, 2.5x more selling time for reps, and $7–10M in annual savings from consolidating fragmented data. They also outline a roadmap for an AI agent serving six portal roles. For a vendor selling a B2B commerce platform or seeking an enterprise portal digital-transformation angle, this material addresses the same buyer task—using digital tools to remove friction from a commercial process—but in a wholesale context.

---
AI-assisted summary of public posts. Audience demographics are not available. Engagement appears only where we checked recent posts. Method: https://b2binfluencer.marketing/methodology/