Odkrivanje izdelkov in spletno trgovsko prikazovanje

Kako kupcem pomagati odkriti izdelke, za katere niso vedeli, da jih morajo iskati

Posted on July 18, 2026 by

Izdelek, ki ga nekdo nikoli ni pomislil poiskati, najlažje postavite v ospredje tako, da izhajate iz njegovega vedenja in konteksta, ne iz iskalnega polja. Opazujte, kaj si ogleduje, kaj kupuje in kaj združuje med seboj, nato pa mu sorodni ali dopolnilni izdelek ponudite v trenutku, ko to zares nekaj pomeni — na strani izdelka, v košarici in predvsem v nadaljnjih e-poštnih in SMS sporočilih, ko že odide. Iskalnik najde le tisto, kar zna človek poimenovati. Večina vašega kataloga živi zunaj tega besednjaka, zato odkrivanje pomeni, da pravega neznanega izdelka predstavite pravemu človeku, in ne, da čakate, da vtipka poizvedbo, za katero nima besed.

Pravi problem: vaš katalog ima žaromet, ki pa je obtičal na 30 izdelkih

Poglejte svojo analitiko izdelkov za zadnje četrtletje. V večini trgovin, ki sem jih videl, ozek pas SKU-jev pobere skoraj vse oglede. Ljudje iščejo tisto, po kar so prišli, pristanejo na tem in bodisi kupijo bodisi odidejo. Preostanek kataloga — pogosto polovica z višjo maržo, novejša ali zanimivejša — komaj kdaj pride na oči.

To je oblika problema. Kupec poišče »brezžične slušalke«, jih najde in nikoli ne izve, da prodajate tudi etui, nastavke, ki se bolje prilegajo njegovemu ušesu, ali polnilno postajo, ki reši prav tisto sitnost, ki jo bo začutil drugi teden. Teh stvari ni iskal, ker ni vedel, da obstajajo, ali pa še ni vedel, da ima ta problem. Povpraševanje je resnično. Zavedanje ni. In iskalnik po svoji zasnovi ne more ustvariti zavedanja — lahko ga le ujame.

Zakaj »boljši iskalnik in več kategorij« tega ne reši

Prvi nagon je izboljšati orodja, s katerimi ljudje iščejo: izostriti iskalni algoritem, dodati filtre, zgraditi več strani kategorij, napisati boljšo navigacijo. Vse to je v redu. Nič od tega pa se ne dotakne dejanske vrzeli.

Iskalnik in filtri pomagajo tistim, ki že približno vedo, kaj hočejo, da to hitreje najdejo. Za človeka, ki ne ve, da vaš najboljši izdelek zanj sploh obstaja, ne storijo ničesar. Do stvari, ki je ne znate poimenovati, se ne morete prefiltrirati. Dodajanje kategorij pogosto naredi stvari slabše — več krajev, kjer se je mogoče izgubiti, več odločitev, isti 30 izdelkov se prebija na vrh vsake od njih. Če bi bil večji meni odgovor, bi veliki katalogi konvertirali bolje kot majhni, a običajno konvertirajo slabše. Razlog, zakaj se veliki katalogi mučijo, je bliže nadaljnjemu nagovarjanju kot strukturi, o čemer teče beseda v zakaj veliki katalogi potrebujejo boljše nadaljnje nagovarjanje, ne več kategorij.

Torej je več navigacije rešitev, namenjena napačnemu kupcu. Človek, ki ga skušate doseči, ni izgubljen v vašem meniju. Že je odšel, zadovoljen s tisto eno stvarjo, ki jo je našel, ne da bi vedel za pet stvari, ki bi jih obožaval.

Kje odteka denar

Predvsem na dveh mestih.

Prvič, obisk z enim samim izdelkom. Nekdo kupi en izdelek in nikoli ne vidi njegovega naravnega spremljevalca, zato naročilo za 45 €, ki bi lahko bilo 70 €, ostane pri 45 €. Pomnožite skromno vrzel, kot je ta, s nekaj sto naročili na mesec in na mizi puščate pravi denar — ne zaradi pokvarjenega lijaka, ampak zaradi tišine po prodaji.

Drugič, dober kupec, ki misli, da ste trgovina z eno samo kategorijo. Pri vas je kupil tekaške copate in vas v glavi arhiviral pod »copati«. Prodajate tudi nogavice, gele, opremo za regeneracijo — a ker se je odkrivanje končalo pri blagajni, vse to kupuje pri nekom drugem. Strošek pridobitve ste plačali enkrat, nato pa vrednost celotne življenjske dobe kupca predali konkurentu, ki je kupcu preprosto povedal, kaj bi še utegnil potrebovati.

Obe puščanji imata isti vzrok: odkrivanje, ki se ustavi v trenutku, ko se ustavi aktivno iskanje.

Praktična rešitev, po vrsti

Ni vam treba vsega naenkrat. Gradite v tem zaporedju.

Najprej uredite odkrivanje na strani, ker je najcenejše. Na vsaki strani izdelka in v košarici pokažite resnično sorodne izdelke — dopolnila in ujemanja, ne naključne mreže najbolje prodajanih. Beseda »sorodni« tu opravlja veliko dela; leni blok, ki vsakomur pokaže vaše najbolje prodajane izdelke, je razlog, zakaj toliko teh podpovprečno deluje — past, ki jo obravnava zakaj generični bloki priporočil pogosto dajejo generične rezultate.
Ujemanja gradite na resničnih razmerjih. Kaj se dejansko kupuje skupaj? Kaj deli kakšno lastnost — isto kolekcijo, isti namen uporabe, isti tip kože, isti prostor? Lastnosti izdelkov so surovina, zaradi katere priporočilo deluje premišljeno in ne naključno; celotna metoda je opisana v kako z lastnostmi izdelkov ustvariti bolj relevantna priporočila.
Odkrivanje raztegnite čez izhod. Tu večina trgovin odneha, čeprav ne bi smela. E-poštno sporočilo, ki ga pošljete dva dni po nakupu ali po brskanju brez nakupa, je vrhunska priložnost za predstavitev izdelka, za katerega kupec ni vedel, da si ga želi. Prodajno urejanje se ne bi smelo končati pri zavihku brskalnika — zakaj naj se prodajno urejanje nadaljuje, ko kupec zapusti spletno mesto to razloži v celoti.
Predstavite resnično novo. Nove prispele izdelke in kategorije, ki se jih kupec še ni dotaknil, so čisto gorivo za odkrivanje, če jih le uskladite s pravim človekom. Priročnik za to je kako obstoječim kupcem predstaviti nove izdelke.

Opazite vrstni red: najprej najcenejše in najbolj neposredno, nato pa sloj nadaljnjega nagovarjanja, ki ga večina vaše konkurence preskoči.

Kaj avtomatizirati — potek od brskanja do odkrivanja

Tu je konkretna avtomatizacija, ki opravi glavnino dela. Predstavljajte si jo kot potek za opuščeno brskanje z nalogo odkrivanja namesto naloge popusta.

Sprožilec: kupec si ogleda enega ali več izdelkov in odide brez nakupa ali pa nakup opravi in vstopi v ponakupno obdobje.
Segment: ločite glede na to, kaj so si ogledali. Nekdo, ki si je ogledal tri izdelke v eni kategoriji, dobi drugačno sporočilo kot nekdo, ki je enkrat kupil in se ni vrnil. Prvi obiskovalec in stalni kupec ne bi smela prejeti istega e-poštnega sporočila.
Časovnica: pri brskanju prvo sporočilo nekaj ur do dan pozneje, dokler je zanimanje sveže. Pri ponakupnem odkrivanju počakajte, da je prvi izdelek prejet in v uporabi — pogosto en do dva tedna, dlje pri premišljenih nakupih.
Kanal: e-pošta za bogatejše, s slikami podprte predstavitve izdelkov; SMS za kratke, pravočasne spodbude, ko je izdelek resnično relevanten in redek.
Vsebina: začnite s stvarjo, ki je ne bi iskali. »Ogledali ste si [šotor] — tukaj je podloga, na katero večina pozabi do prvega mokrega jutra.« Ena jasna ideja, dva ali trije izdelki, en gumb. Ne izpust celotnega kataloga.
Cilj: klik na izdelek, ki ga še niso videli, in sčasoma drugi nakup iz kategorije, iz katere še niso kupovali.

Motor pod tem so podatki o brskanju, dobro branje teh signalov pa je veščina zase — kako vedenje pri brskanju uporabiti za boljše odkrivanje izdelkov se poglobi v mehaniko pretvarjanja surovih ogledov v uporabne segmente.

Primer trgovine (ponazoritveni)

Recimo, da vodite trgovino s kuhinjsko opremo. Kupec poišče »litoželezno ponev«, kupi ponev premera 26 cm in odide. Njegovo naročilo je zaključeno. Njegova kuhinja ni.

Iz lastnih podatkov o skupnih nakupih veste, da se kupci ponev pogosto vračajo po žično krtačo, prevleko za ročaj in olje za nego — običajno v enem mesecu, običajno potem, ko naletijo na sitnost, ki jo ti izdelki rešujejo. Zato deset dni po dostavi prispe samodejno e-poštno sporočilo: »Zdaj, ko ste na njej že nekajkrat kuhali — tri stvari, ki litoželezno ponev ohranjajo prijetno za vsakdanjo uporabo.« Trije izdelki, vsak vezan na resnično nevšečnost, ki jo bo kupec kmalu občutil, ena povezava. Nič od tega prvi dan za tega kupca ni bilo mogoče poiskati, ker še ni vedel, da krtača obstaja ali da je milo problem.

Te številke so ponazoritvene — pred gradnjo česarkoli vstavite lastne podatke o skupnih nakupih in svoje marže. Bistvo je oblika: predstavili ste izdelke, ki jih kupec nikoli ne bi iskal, v trenutku, ko so postali relevantni.

Kako izmeriti, ali odkrivanje deluje

Ne utapljajte se v metrikah. Spremljajte tole:

Ogledani izdelki na sejo in, kar je bolj zgovorno, koliko različnih izdelkov kupec vidi v svoji celotni življenjski dobi. Odkrivanje naj to število razširi.
Širina kategorij na kupca — iz koliko kategorij v povprečju kupuje stalni kupec. Če odkrivanje deluje, se to število veča.
Prihodek na prejemnika pri e-poštnih sporočilih za odkrivanje in klik do izdelkov, ki si jih kupec prej ni ogledal.
Stopnjo pripenjanja pri blokih sorodnih izdelkov na strani.

Eno pošteno opozorilo: prepričajte se, da ustvarjate novo povpraševanje, ne pa preimenujete prodaje, ki bi se zgodila tako ali tako. Priporočilo, ki »zmaga« tako, da si pripiše zasluge za naročilo, ki bi ga kupec vseeno oddal, ni odkrivanje — je knjigovodstvo. Kako to dvoje ločiti, obravnava preizkušanje, ali priporočila povečujejo prihodek ali ga zgolj premeščajo.

Kako se v to umešča Omnisend

Ko se enkrat odločite, kaj in kdaj pokazati, potrebujete nekaj, kar se odzove na vedenje, ne da bi ročno sestavljali vsako pošiljanje. Tu se v lastnih trgovinah opiram na Omnisend, potem ko sem ga preizkusil proti Klaviyu in ostal zaradi intuitivnejše nastavitve, združene e-pošte in SMS-a ter cene. Njegovi bloki priporočil izdelkov lahko izberejo izdelke, povezane s tem, kar si je stik ogledal ali kupil, njegove avtomatizacije sprožijo sprožilce za brskanje in ponakupno obdobje, njegova segmentacija pa vam omogoča, da ločite kupca ene kategorije od zvestega rednega kupca, tako da isto e-poštno sporočilo ne gre obema.

Pošteno o omejitvah: orodje izpostavi kandidate, a ne more odločiti, za kateri neznani izdelek bi se vam kupec dejansko zahvalil — ta presoja izhaja iz poznavanja vašega kataloga in vaših kupcev. In odkrivanje se obrestuje le, če je izdelek v ozadju resnično dober; avtomatizacija ga predstavi, ne odreši ga. Omnisend je partner Shopimationa preko affiliate programa in ga priporočam iz vsakdanje uporabe; brezplačni paket zadošča za preizkus poteka od brskanja do odkrivanja na enem segmentu, preden se zavežete.

Vaš naslednji korak

Izvlecite svojo analitiko izdelkov in poiščite svoj najtišji izdelek z visoko maržo — tisti dobri, ki ga nihče ne išče. Zastavite eno vprašanje: kateri obstoječi kupec že ima nekaj, zaradi česar bi bil ta izdelek očiten naslednji nakup? To ujemanje je vaše prvo sporočilo za odkrivanje. Sestavite ga, pošljite majhnemu segmentu in opazujte, ali odpre vrata, ki jih iskalnik nikoli ni mogel. Ko boste pripravljeni sistematizirati celotno pot, je kako z zgodovino nakupov odkriti naslednjo kategorijo, ki jo kupec morda potrebuje logično naslednje branje.

Product Discovery and Merchandising

How to Help Shoppers Discover Products They Did Not Know to Search For

Posted on July 18, 2026 by

The way to surface a product someone never thought to search for is to lead with their behavior and context instead of the search box. Watch what they view, buy, and pair together, then put the adjacent or complementary item in front of them at a moment when it actually makes sense — on the product page, in the cart, and above all in follow-up email and SMS once they’ve left. Search only finds what a person can already name. Most of your catalog lives outside that vocabulary, so discovery means introducing the right unfamiliar product to the right person, not waiting for them to type a query they don’t have the words for.

The real problem: your catalog has a spotlight, and it’s stuck on 30 products

Look at your product analytics for the last quarter. In most stores I’ve seen, a narrow band of SKUs takes almost all the views. People search for what they came in for, land on it, and either buy or leave. The rest of the catalog — often the higher-margin, newer, or more interesting half — barely gets seen.

That’s the shape of the problem. A shopper searches “wireless earbuds,” finds them, and never learns you also sell the case, the tips that fit their ear better, or the charging dock that solves the exact annoyance they’ll feel in week two. They didn’t search for those things because they didn’t know those things existed, or didn’t know they had the problem yet. The demand is real. The awareness isn’t. And search, by design, can’t create awareness — it can only match it.

Why “better search and more categories” doesn’t fix this

The instinct is to improve the tools people use to look: sharpen the search algorithm, add filters, build more category pages, write better navigation. All fine. None of it touches the actual gap.

Search and filters help people who already know roughly what they want find it faster. They do nothing for the person who doesn’t know your best product for them exists. You can’t filter your way to a thing you can’t name. Adding categories often makes it worse — more places to get lost, more decisions, the same 30 products surfacing to the top of each one. If a bigger menu were the answer, large catalogs would convert better than small ones, and they usually convert worse. The reason large catalogs struggle is closer to follow-up than to structure, which is its own discussion in why large product catalogs need better follow-up, not more categories.

So more navigation is a fix aimed at the wrong shopper. The person you’re trying to reach isn’t lost in your menu. They’ve already left, satisfied with the one thing they found, unaware of the five things they’d have loved.

Where the money leaks

Two places, mostly.

First, the single-item visit. Someone buys one product and never sees the natural companion, so a €45 order that could have been €70 stays €45. Multiply a modest gap like that across a few hundred orders a month and you’re leaving real money on the table — not through a broken funnel, but through silence after the sale.

Second, the good customer who thinks you’re a one-category shop. They bought running shoes from you and mentally filed you under “shoes.” You also sell the socks, the gels, the recovery gear — but because discovery ended at checkout, they buy all of that from someone else. You paid the acquisition cost once and then handed the lifetime value to a competitor who simply told the customer what else they might need.

Both leaks share a cause: discovery that stops the moment active searching stops.

The practical solution, in order

You don’t need everything at once. Build in this sequence.

Fix on-site discovery first, because it’s cheapest. On every product page and in the cart, show genuinely related items — complements and pairings, not a random bestseller grid. The word “related” is doing a lot of work here; a lazy block that shows your top sellers to everyone is why so many of these underperform, a trap covered in why generic recommendation blocks often produce generic results.
Base the pairings on real relationships. What actually gets bought together? What shares an attribute — same collection, same use case, same skin type, same room? Product attributes are the raw material that makes a recommendation feel deliberate rather than random; there’s a full method in how to use product attributes to create more relevant recommendations.
Extend discovery past the exit. This is where most stores quit and shouldn’t. The email you send two days after a purchase, or after a browse with no purchase, is prime real estate for introducing the product they didn’t know to want. Merchandising shouldn’t end at the browser tab — why merchandising should continue after a shopper leaves the website makes the case in full.
Introduce the genuinely new. New arrivals and categories a customer hasn’t touched are pure discovery fuel, provided you match them to the right person. The playbook for that is how to introduce new products to existing customers.

Notice the order: cheapest and most immediate first, then the follow-up layer that most of your competitors skip.

What to automate — the browse-to-discovery flow

Here’s a concrete automation that does the heavy lifting. Think of it as a browse abandonment flow with a discovery job rather than a discount job.

Trigger: a shopper views one or more products and leaves without buying, or completes a purchase and enters the post-purchase window.
Segment: split by what they looked at. Someone who viewed three items in one category gets a different message than someone who bought once and hasn’t returned. A first-time browser and a repeat customer should not receive the same email.
Timing: for browse, a first message a few hours to a day later, while the interest is warm. For post-purchase discovery, wait until the first product has been received and used — often one to two weeks, longer for considered goods.
Channel: email for the richer, image-led product introductions; SMS for short, timely nudges when the item is genuinely relevant and scarce.
Content: lead with the thing they wouldn’t have searched for. “You looked at the [tent] — here’s the footprint most people forget until their first wet morning.” One clear idea, two or three products, one button. Not a catalog dump.
Goal: a click into a product they hadn’t seen, and over time, a second purchase from a category they hadn’t bought from.

The engine underneath this is browsing data, and reading those signals well is a skill in itself — how to use browsing behavior to improve product discovery goes into the mechanics of turning raw views into useful segments.

A store example (illustrative)

Say you run a kitchenware shop. A customer searches “cast iron skillet,” buys the 26cm pan, and leaves. Their order is done. Their kitchen isn’t.

You know from your own bought-together data that skillet buyers frequently come back for the chainmail scrubber, the pan handle cover, and the seasoning oil — usually within a month, usually after they’ve hit the annoyance those products solve. So ten days after delivery, an automated email arrives: “Now that you’ve cooked on it a few times — three things that keep a cast iron pan easy to live with.” Three products, each tied to a real friction the customer is about to feel, one link. None of these were searchable to that customer on day one, because they didn’t yet know the scrubber existed or that soap was a problem.

Those numbers are illustrative — plug in your own bought-together data and margins before you build anything. The point is the shape: you introduced products the customer would never have searched for, at the moment they’d become relevant.

How to measure whether discovery is working

Don’t drown in metrics. Track these:

Products-viewed per session and, more tellingly, how many distinct products a customer sees over their lifetime. Discovery should widen that number.
Category breadth per customer — how many categories the average repeat buyer purchases from. If discovery is working, it climbs.
Revenue per recipient on the discovery emails, and click-through into products the customer hadn’t previously viewed.
Attach rate on the on-site related-product blocks.

One honest warning: make sure you’re creating new demand rather than relabeling sales that would have happened anyway. A recommendation that “wins” by taking credit for an order the customer was already going to place isn’t discovery — it’s accounting. The way to separate the two is covered in testing whether recommendations increase revenue or just shift it.

How Omnisend fits

Once you’ve decided what to show and when, you need something that reacts to behavior without you hand-building each send. This is where I lean on Omnisend in my own stores, after testing it against Klaviyo and staying for the more intuitive setup, the combined email and SMS, and the pricing. Its product recommendation blocks can pull items related to what a contact viewed or bought, its automations fire off browse and post-purchase triggers, and its segmentation lets you separate the one-category buyer from the loyal regular so the same email doesn’t go to both.

Being honest about the limits: the tool surfaces candidates, but it can’t decide which unfamiliar product a customer would actually thank you for — that judgment comes from knowing your catalog and your buyers. And discovery only pays off if the underlying product is genuinely good; automation introduces it, it doesn’t redeem it. Omnisend is an affiliate partner of Shopimation, and I recommend it from daily use; the free tier is enough to test a browse-to-discovery flow on one segment before you commit.

Your next step

Pull your product analytics and find your quietest high-margin product — the good one nobody searches for. Ask one question: which existing customer already owns something that would make this the obvious next buy? That pairing is your first discovery message. Build it, send it to a small segment, and watch whether it opens a door search never could. When you’re ready to systemize the whole path, how to use purchase history to discover the next category a customer may need is the logical next read.