Odkrivanje izdelkov in spletno trgovsko prikazovanje

Kaj pokazati kupcem, ki iščejo rešitev težave in ne vrste izdelka

Posted on July 18, 2026 by

Ko kupec brska glede na težavo — “noge me bolijo po celem dnevu stoje”, “moj pes se ves čas praska” — mu najprej pokažite izid in šele nato kategorijo izdelka. Ti kupci ne vedo in jih ne zanima, v katerem oddelku živi odgovor. Poznajo svojo situacijo in želijo olajšanje. Zato je prava stvar, ki jo pokažete, majhen nabor izdelkov, uokvirjen s težavo, ki so jo opisali, s kratko razlago, zakaj vsak od njih rešuje to težavo — ne surov seznam, urejen po vrsti izdelka. Če to zgrešite, kupcu, pripravljenemu na nakup, izročite taksonomijo, ki jo mora prevesti sam. Če to zadenete, ste videti kot trgovina, ki je končno razumela, kaj sprašuje. Ta vodnik govori o tem, kaj natanko pokazati tem kupcem, na strani in pri spremljanju.

Dva kupca, dva popolnoma različna miselna modela

Nekateri ljudje kupujejo po vrsti izdelka. Že vedo, da želijo litoželezno ponev premera 38 cm, in se pripeljejo naravnost do nje. Vaša struktura kategorij jim ustreza povsem.

Kupci, ki na prvo mesto postavljajo težavo, razmišljajo v povsem drugačnem jeziku. Pridejo s simptomom ali ciljem — “nekaj, da mi kozarci ne bodo rjaveli”, “darilo za nekoga, ki ima vse”, “pomoč, da bo malček prespal noč”. Nimajo izdelka v mislih. Imajo težavo in upajo, da jo boste povezali z izdelkom. Trgovina, ki govori le v kategorijah, tega kupca prisili v prevod, ki ga pogosto ne zmore, ker vašega kataloga ne pozna dovolj dobro, da bi vedel, da je odgovor uvrščen pod “nego litega železa”.

Ta razcep je pomemben, ker so kupci, ki dajejo prednost težavi, pogosto zelo nagnjeni k nakupu. Simptom je močnejši nakupni signal kot brskanje po kategoriji. Ta oseba ne razgleduje po oddelku — želi, da težava izgine.

Pravi problem: vaša trgovina odgovarja na vprašanje, ki ga niso zastavili

Vtipkajte težavo v večino iskalnih vrstic v spletnih trgovinah in opazujte, kaj se zgodi. Bodisi ne dobite nič bodisi dobite po ključnih besedah ujeto zmešnjavo, urejeno po vrsti izdelka. Kupec je vprašal “zakaj mi tekaški copati delajo žulje”, trgovina pa je odgovorila z mrežo 60 copatov, urejenih po ceni.

To je neujemanje. Kupec je govoril v izidih; trgovina je odgovorila v SKU-jih. Zdaj je delo, ugotoviti, kateri od teh 60 copatov dejansko prepreči žulje — natanko tisto, po kar je prišel — vrnjeno osebi, ki je za to najmanj opremljena. Večina tega ne bo storila. Domnevali bodo, da nimate tega, kar potrebujejo, in odšli, tudi ko popoln izdelek sedi na prvi strani.

Za trgovca je frustrirajoč del ta, da je bil ta kupec blizu. Svojo težavo je izrekel na glas. Trgovina pa preprosto ni bila nastavljena tako, da bi ga slišala.

Zakaj več kategorij in boljši filtri tega ne rešijo

Mamljiva rešitev je več navigacije — dodati filter “preprečevanje žuljev”, podkategorijo “občutljiva koža”, meni na podlagi simptomov. Včasih kurirana kolekcija pomaga (o tem več spodaj). A pripenjanje jezika težav na sistem kategorij ima meje.

Filtri predpostavljajo, da kupec že ve, kateri atribut ustreza njegovi težavi. Tekač, nagnjen k žuljem, ne ve nujno, da naj filtrira za “oblazinjen ovratnik pete” ali “odvajanje vlage” — to so vaše besede, ne njegove. Zato filter ostane neuporabljen. In podkategorije ne morete zgraditi za vsako ubeseditev vsake težave; kombinacij je neskončno, meni pa postane svoja vrsta preobremenitve.

Popuščanje je tu še bolj mimo bistva. Kupec, ki daje prednost težavi, ni obtičal pri ceni — še ni prepričan, da odgovor sploh imate. Cena postane pomembna šele, ko kupec verjame, da izdelek reši njegovo težavo. Vodite s popustom in popust ste dali na rešitev, ki je še ni našel.

Pravi odgovor ni več strukture. Je povezovanje težav z izdelki prek namere, kar je celotna tema gradnje priporočil izdelkov okoli namere strank.

Kje odteka prihodek

Odtok tiči v vrzeli med težavo in izdelkom. Kupec se s pomočjo tipkanja ali klikanja pripelje z jasno namero, zadene ob steno nerazložene ponudbe in odskoči — noseč denarnico, ki je bila odprta.

Tukaj je oblika tega. Recimo, da poizvedbe in vstopne točke, ki dajejo prednost težavi, predstavljajo, denimo, 15 % sej na straneh izdelkov, te seje pa konvertirajo z polovico stopnje sej, ki dajejo prednost kategoriji, ker kupec svoje težave ne zna preslikati na vaš katalog. (Ilustrativno — preverite lastne podatke o iskanju po strani in pristajanju.) Ta vrzel je čist povračljiv prihodek. To niso ljudje, ki jih je treba prepričevati, da imajo težavo. Težavo so vam povedali. Vi le niste odgovorili v jeziku, v katerem bi lahko ukrepali.

Praktična rešitev: odgovorite z izidi, nato pokažite izdelek

Predelajte izkušnjo tako, da vodi jezik težav in sledi vrsta izdelka. Po vrsti:

1. Ustvarite vstopne točke, uokvirjene s težavo

Zgradite peščico kuriranih kolekcij, poimenovanih po težavah, ne po izdelkih: “za utrujene noge”, “pomiri srbeč kožuh”, “premagaj popoldanski upad ob treh”. Vsaka zbira izdelke, ki resnično rešujejo to težavo, ne glede na to, v kateri kategoriji živijo. Kupec prepozna svoje besede in klikne.

2. Na kratko pojasnite “zakaj” ob vsakem izdelku

Kupec, ki daje prednost težavi, potrebuje razlog, ne le fotografije. Ena vrstica na izdelek — “oblazinjena peta in mehka podloga brez drgnjenja, tako da ni žuljev na dolgih izmenah” — naredi več kot katerakoli razpredelnica specifikacij. Zanj zapirate vrzel prevoda.

3. Priporočajte po atributu, ne po kategoriji

V ozadju je stvar, ki težavo poveže s pravimi izdelki, atribut — material, lastnost, namen uporabe. Preslikajte težavo na atribute, ki jo rešujejo, in naj ti poganjajo to, kar prikažete. Metoda je opisana v kako uporabiti atribute izdelkov za bolj relevantna priporočila.

4. Spremljajte, katero težavo je brskal, in si to zapomnite

Če je znani kupec svojo sejo preživel v “za občutljivo kožo”, je to trajno dejstvo o njem, ne enkraten dogodek. To bi moralo oblikovati vse, kar mu pokažete naslednjič — drugačno delo od ponovnega prikazovanja najbolje prodajanih izdelkov, kar je natanko razlika v razlika med najbolje prodajanimi izdelki in najboljšim izdelkom za tega kupca.

Kaj avtomatizirati — spremljanje na podlagi težave

Težava, ki jo je kupec poimenoval, je darilo za vaše avtomatizacije, ker vam natanko pove, kaj pokazati naslednje.

Sprožilec: znani kontakt brska po kolekciji, uokvirjeni s težavo, ali po skupini izdelkov, vezanih na eno težavo, in odide, ne da bi kupil.
Segment: prepoznani kontakti, razvrščeni po težavi, za katero so pokazali zanimanje (brskalci “utrujenih nog”, brskalci “občutljive kože”), izključujoč nedavne kupce.
Časovnica: nekaj ur do en dan po seji.
Kanal: e-pošta, ki nese razlago; SMS ali potisno sporočilo za kratko spodbudo pri ponovnih obiskovalcih z veliko namero.
Vsebina: izdelki, ki rešujejo njihovo konkretno težavo, vsak z enovrstičnim razlogom, uokvirjeni s težavo v njihovih lastnih besedah — “Se še vedno spopadate z utrujenimi nogami? Ti trije so narejeni za to.” En jasen gumb.
Cilj: pripeljati jih nazaj k ožjemu izboru, ki govori njihov jezik, in konvertirati.

Sporočilo še naprej govori o njihovi težavi, ne o vašem katalogu. Ta kontinuiteta je tisto, zaradi česar zaleže.

Primer iz trgovine

Trgovina z izdelki za hišne ljubljenčke dobi obiskovalko iz iskanja, kot je “ustavi praskanje psa”. V večini trgovin bi pristala v “šamponih” — 50 izdelkov, urejenih po ceni, brez pojma, kateri obravnava srbenje. Odide.

Predelano za nakupovanje, ki daje prednost težavi, pristane na kurirani kolekciji “pomiri srbečo kožo”: šest izdelkov med šamponi, dodatki in razpršili, vsak z vrstico, ki pojasnjuje, kako pomaga. Je znani kontakt, zato ji, ko odide brez nakupa, e-pošta za spremljanje nekaj ur pozneje znova odpre s “Še vedno poskušate ustaviti praskanje?” in prikaže istih šest, na čelu z dvema, ki ju v njeni situaciji izbere največ ljudi. Nikoli ji ni bilo treba vedeti, da je odgovor razdeljen med tri vaše kategorije. To delo ste opravili vi zanjo.

Kako izmeriti, ali deluje

Stopnja konverzije na kolekcijah, uokvirjenih s težavo v primerjavi z enakovrednimi kategorijami po vrsti izdelka — najjasnejši znak, da uokvirjanje pomaga.
Stopnja iz iskanja po strani do nakupa za poizvedbe v slogu težav.
Stopnja klikov na e-pošti za spremljanje na podlagi težave — pove vam, ali se je vaše uokvirjanje ujemalo z njihovo namero.
Prihodek na prejemnika iz potekov na segmentih težav.

En način odpovedi, na katerega je treba paziti: če kolekcije težav dobijo klike, a ne naročil, ste se ujeli z jezikom, ne pa z izdelki — izdelki v tej kolekciji težave v resnici ne rešujejo dovolj dobro. Popravite kuriranje, preden se dotaknete besedila.

Kako pomaga Omnisend

Gradnja kolekcij, uokvirjenih s težavo, je delo na strani. Spremljanje — razvrščanje kupcev po težavi, ki so jo brskali, in ponovno seganje po njih v njihovem jeziku — je tam, kjer uporabljam Omnisend, za katerega sem se odločil za svoje trgovine, potem ko sem ga preizkusil proti Klaviyu. Sledi, katere kolekcije in izdelke si je znani kontakt ogledal, omogoča razvrščanje na podlagi tega vedenja in gradi dinamične bloke izdelkov okoli relevantnih atributov, tako da brskalec “občutljive kože” v spremljanju dobi izdelke za občutljivo kožo, ne da bi jih ročno izbirali. E-pošta nese razlago; SMS in potisna sporočila poskrbijo za kratke spodbude, vse v enem poteku.

Iskren pridržek: orodje lahko razvršča in pošilja, a preslikava iz težave v izdelek je vaša, da jo določite — vi veste, kateri od vaših izdelkov dejansko reši katero težavo, in nobena avtomatizacija tega ne more sklepati namesto vas. Omnisend je partner Shopimationa v pridruženem programu; priporočam ga iz vsakodnevne uporabe, brezplačna raven pa zadošča, da preizkusite en segment težave, preden razširite.

Vaš naslednji korak

Poglejte iskalne izraze na svoji strani za zadnji mesec in izluščite tiste, ki so ubesedeni kot težave in ne kot izdelki. Vsak ponavljajoč se je kolekcija, ki čaka, da jo zgradite, in segment za spremljanje, ki čaka, da ga ustvarite. Začnite z eno samo najpogostejšo poizvedbo o težavi, zgradite eno kolekcijo, uokvirjeno s težavo, in speljite spremljanje za kupce, ki jo brskajo. Nato uskladite sporočanje s tem, kje v svoji odločitvi je vsak kupec, z kako priporočila izdelkov prilagoditi fazi nakupne poti, in razširite njihovo odkrivanje z kako kupcem pomagati odkriti izdelke, ki jih niso znali poiskati.

Product Discovery and Merchandising

What to Show Customers Who Browse by Problem Instead of Product Type

Posted on July 18, 2026 by

When a customer browses by problem — “my feet hurt after standing all day,” “my dog keeps scratching” — show them the outcome first and the product category second. These shoppers don’t know or care what department the answer lives in. They know their situation and they want relief. So the right thing to show is a small set of products framed by the problem they described, with a short explanation of why each one solves it, not a raw list sorted by product type. Get this wrong and you hand a ready-to-buy shopper a taxonomy they have to translate themselves. Get it right and you look like the store that finally understood what they were asking. This guide is about exactly what to show these shoppers, on-site and in follow-up.

Two shoppers, two completely different mental models

Some people shop by product type. They already know they want a 15-inch cast-iron skillet, and they navigate straight to it. Your category structure serves them fine.

Problem-first shoppers think in a different language entirely. They arrive with a symptom or a goal — “something to stop my jars from rusting,” “a gift for someone who has everything,” “help my toddler sleep through the night.” They don’t have a product in mind. They have a problem, and they’re hoping you’ll connect it to a product. The store that only speaks in categories forces this shopper to do a translation they often can’t do, because they don’t know your catalog well enough to know that the answer is filed under “cast iron care.”

This split matters because problem-first shoppers are frequently high-intent. A symptom is a stronger buying signal than a category browse. The person isn’t window-shopping a department — they want the problem gone.

The real problem: your store answers a question they didn’t ask

Type a problem into most ecommerce search bars and watch what happens. Either you get nothing, or you get a keyword-matched jumble sorted by product type. The shopper asked “why do my running shoes give me blisters” and the store replied with a grid of 60 shoes sorted by price.

That’s the mismatch. The shopper spoke in outcomes; the store answered in SKUs. Now the work of figuring out which of those 60 shoes actually prevents blisters — the exact thing they came for — has been handed back to the person least equipped to do it. Most won’t. They’ll assume you don’t have what they need and leave, even when the perfect product is sitting on page one.

The frustrating part for the merchant: this shopper was close. They named their problem out loud. The store just wasn’t set up to hear it.

Why more categories and better filters don’t solve it

The tempting fix is more navigation — add a “blister prevention” filter, a “sensitive skin” subcategory, a symptom-based menu. Sometimes a curated collection helps (more on that below). But bolting problem-language onto a category system has limits.

Filters assume the shopper already knows which attribute maps to their problem. A blister-prone runner doesn’t necessarily know to filter for “padded heel collar” or “moisture-wicking” — those are your words, not theirs. So the filter sits unused. And you can’t build a subcategory for every phrasing of every problem; the combinations are endless and the menu becomes its own kind of overwhelm.

Discounting is even more beside the point here. A problem-first shopper isn’t stuck on price — they’re not yet convinced you have the answer at all. Price only matters after the shopper believes a product solves their problem. Lead with a discount and you’ve discounted a solution they haven’t found yet.

The real answer isn’t more structure. It’s connecting problems to products through intent, which is the whole subject of building product recommendations around customer intent.

Where the revenue leaks

The leak sits in the gap between the problem and the product. A shopper types or clicks their way in with clear intent, hits a wall of unexplained options, and bounces — carrying a wallet that was open.

Here’s the shape of it. Suppose problem-first queries and entry points account for, say, 15% of your product-page sessions, and those sessions convert at half the rate of category-first sessions because the shopper can’t map their problem to your catalog. (Illustrative — check your own site search and landing data.) That gap is pure recoverable revenue. These aren’t people who need convincing that they have a problem. They told you the problem. You just didn’t answer in a language they could act on.

The practical solution: answer in outcomes, then show the product

Rebuild the experience so problem-language leads and product-type follows. In order:

1. Create problem-framed entry points

Build a handful of curated collections named for problems, not products: “for tired feet,” “calm an itchy coat,” “beat the 3pm slump.” Each collects the products that genuinely solve that problem, regardless of which category they live in. The shopper recognizes their own words and clicks.

2. Explain the why, briefly, next to each product

A problem-first shopper needs a reason, not only a photo. One line per product — “cushioned heel and a soft, rub-free lining, so no blisters on long shifts” — does more than any spec table. You’re closing the translation gap for them.

3. Recommend by attribute, not by category

Behind the scenes, the thing that connects a problem to the right products is attributes — material, feature, use case. Map the problem to the attributes that solve it, and let those drive what you show. The method is laid out in how to use product attributes to create more relevant recommendations.

4. Watch which problem they browsed, and remember it

If a known shopper spent their session in “for sensitive skin,” that’s a durable fact about them, not a one-off. It should shape everything you show them next — a different job from reshowing best sellers, which is exactly the distinction in the difference between best sellers and the best product for this customer.

What to automate — problem-based follow-up

The problem the shopper named is a gift for your automations, because it tells you precisely what to show next.

Trigger: a known contact browses a problem-framed collection or a cluster of products tied to one problem, and leaves without buying.
Segment: identified contacts, grouped by the problem they showed interest in (“tired feet” browsers, “sensitive skin” browsers), excluding recent buyers.
Timing: a few hours to a day after the session.
Channel: email to carry the explanation; SMS or push for a short nudge on high-intent repeat viewers.
Content: the products that solve their specific problem, each with a one-line reason, framed by the problem in their own words — “Still dealing with tired feet? These three are built for it.” One clear button.
Goal: bring them back to a shortlist that speaks their language, and convert.

The message keeps talking about their problem, not your catalog. That continuity is what makes it land.

A store example

A pet-supplies store gets a visitor from a search like “stop dog scratching.” On most stores she’d land in “shampoos” — 50 products, sorted by price, no idea which one addresses itching. She leaves.

Rebuilt for problem-first shopping, she lands on a curated “soothe itchy skin” collection: six products across shampoos, supplements, and sprays, each with a line explaining how it helps. She’s a known contact, so when she leaves without buying, a follow-up email a few hours later reopens with “Still trying to stop the scratching?” and shows those same six, led by the two most people in her situation choose. She never had to know that the answer was split across three of your categories. You did that work for her.

How to measure whether it’s working

Conversion rate on problem-framed collections vs. equivalent product-type categories — the clearest sign the framing helps.
Site-search-to-purchase rate for problem-style queries.
Click rate on problem-based follow-up emails — tells you whether your framing matched their intent.
Revenue per recipient from the problem-segment flows.

One failure mode to watch: if problem collections get clicks but not orders, you’ve matched the language but not the products — the items in that collection don’t actually solve the problem well enough. Fix the curation before you touch the copy.

How Omnisend helps

Building the problem-framed collections is on-site work. The follow-up — segmenting shoppers by the problem they browsed and reaching back out in their language — is where I use Omnisend, which I settled on for my own stores after testing it against Klaviyo. It tracks which collections and products a known contact viewed, lets you segment on that behavior, and builds dynamic product blocks around the relevant attributes, so a “sensitive skin” browser gets sensitive-skin products in the follow-up without you hand-picking them. Email carries the explanation; SMS and push handle the short nudges, all in one flow.

The honest caveat: the tool can segment and send, but the mapping from problem to product is yours to define — you know which of your products actually solve which problem, and no automation can infer that for you. Omnisend is an affiliate partner of Shopimation; I recommend it from daily use, and the free tier is enough to test one problem segment before expanding.

Your next step

Look at your site-search terms for the last month and pull the ones phrased as problems rather than products. Each recurring one is a collection waiting to be built and a follow-up segment waiting to be created. Start with the single most common problem query, build one problem-framed collection, and wire up a follow-up for the shoppers who browse it. Then match your messaging to where each shopper is in their decision with how to match product recommendations to the stage of the buying journey, and widen their discovery with how to help shoppers discover products they did not know to search for.