Počasi prodajana zaloga

Kaj preizkusiti, preden izdelek odpišete

Posted on July 18, 2026 by

Preden počasi prodajan izdelek razprodate ali umaknete iz kataloga, po vrsti opravite pet poceni preizkusov: cena, stran izdelka, občinstvo, umeščanje in združevanje v komplete. Večina jih ne stane nič drugega kot en teden in nekaj pozornosti, odgovorijo pa na edino vprašanje, ki šteje – je to slab izdelek ali dober izdelek, ki se slabo prodaja? Odpis izdelka je trajna odločitev z resnično ceno: denar, ki ste ga vezali v nakup, znižanje, ki ga boste požrli pri razprodaji, in tiha možnost, da bi se prodajal povsem dobro, če bi eno stvar spremenili. Zato preizkusite, preden odpišete. Ta stran je zaporedje preizkusov v takem vrstnem redu, da najprej najde najcenejšo rešitev.

Prava cena prezgodnjega odpisa izdelka

Odpis se zdi odločen, in odločnost je prijetna, ko vas izdelek nadleguje že mesece. A prezgodnji odpis je ena dražjih napak, ki jih trgovina naredi, prav zato, ker izgleda kot zaključek.

Pomislite, česa se pravzaprav odrekate. Zalogo ste plačali. Če jo počistite z velikim popustom, na vsakem kosu požrete maržo. Če jo umaknete iz trgovine, požrete celoten strošek. In izgubite informacijo – nikoli ne boste vedeli, ali bi izdelek obrnila 4-evrska sprememba v vrstici o dostavi ali boljša druga fotografija, ker ste izdelek odstranili, preden ste to izvedeli. Nagon, da bi se ga “preprosto znebili”, je razumljiv, ko strmite v zalogo, ki se ne premakne. Je pa tudi način, kako dobri izdelki umrejo iz slabih razlogov.

Zakaj “ne prodaja se” ni diagnoza

“Ne prodaja se” opisuje simptom. O vzroku vam ne pove ničesar, vzrokov pa je vsaj pet pogostih, vsak s svojo rešitvijo.

Izdelek se lahko ne prodaja, ker ima napačno ceno, ker mu stran ne dela pravice, ker ga vidijo napačni ljudje, ker je umeščen proti napačni alternativi ali ker za smisel potrebuje spremljajoči izdelek. Le en vzrok – res ni trga – upravičuje odpis. Ostali štirje so odpravljivi, pogosto poceni. Segati po znižanju ali gumbu za brisanje obravnava vsakega od teh kot isto težavo, in zato se toliko počistljive, prodajljive zaloge zavrže. Včasih je pošten odgovor res, da ste kupili napačen izdelek, in to je ločen položaj, ki ga velja jasno poimenovati – to mejo sem začrtal v razlika med počasi prodajanim izdelkom in napačnim izdelkom.

Preizkusi, od najcenejšega naprej

Opravite jih po vrsti. Vsak je zasnovan tako, da vzrok potrdi ali izključi, preden porabite denar. Ustavite se, takoj ko eden premakne številke – našli ste svoj odgovor.

1. Cena

Cena je najhitreje preizkusljiva stvar in najpogostejša posamezna rešitev, zato začnite tukaj. A “preizkusi ceno” ne pomeni le “znižaj jo”.

Poskusite z zmerno spremembo – navzgor ali navzdol – za dva tedna in opazujte stopnjo pretvorbe, ne surovega števila naročil. Navzdol je očitno. Navzgor je tisto, kar ljudje preskočijo: prenizko ovrednoten izdelek lahko deluje poceni ali nizkokakovostno, zlasti v kategorijah, kjer cena signalizira vrednost, in dvig navzgor včasih dvigne prodajo. Preverite tudi vrstico o dostavi, saj je številka, ki ustavi nakup, pogosto vsota na blagajni, ne cena na strani. Izdelek za 39 EUR z 12-evrskim stroškom kosovne dostave, razkritim pozno, je težava s ceno, ki se skriva kot težava z izdelkom. Preden se zavežete trajnemu znižanju, preberite zakaj znižanje cene ni vedno najboljša rešitev za počasi prodajano zalogo – popust je lahko dati in težko vzeti nazaj.

2. Stran izdelka

Če je cena v redu, poglejte, kaj obiskovalec dejansko vidi. Izdelek s spodobnim prometom in slabo stopnjo pretvorbe navadno izgublja ljudi na sami strani.

Spreminjajte po eno stvar naenkrat, da veste, kaj je delovalo:

Glavna slika. Zamenjajte jo za jasnejši, bolje osvetljen ali kontekstni posnetek. To samo po sebi premakne pretvorbo pogosteje kot katera koli druga posamezna sprememba na strani.
Prvi dve vrstici opisa. Začnite s tem, kaj izdelek naredi za kupca, ne s tehničnimi podatki.
Očitno neodgovorjeno vprašanje. Velikost, material, združljivost, kaj je v paketu. Če mora kupec vprašati po e-pošti, večina ne bo – odšli bodo.

Ali je stran sploh težava, je odvisno od tega, ali ljudje sploh prihajajo. Stran ne more pretvoriti prometa, ki ga ne dobiva, in ta razcep – težava s prometom proti težavi s pretvorbo – je prvo razpotje, ki ga velja razrešiti: kako ugotoviti, ali ima počasi prodajan izdelek težavo s prometom ali s pretvorbo.

3. Občinstvo

Dober izdelek, pokazan napačnim ljudem, izgleda natanko kot slab izdelek. Zato preizkusite, kdo ga vidi.

Namesto da izdelek potiskate na ves svoj seznam, ga pošljite majhnemu segmentu, izbranemu zaradi pristne ustreznosti – ljudem, ki so kupili nekaj sorodnega ali so brskali po kategoriji. Če ozek, dobro izbran segment kupuje po zdravi stopnji, medtem ko ga je široko občinstvo prezrlo, je izdelek v redu; težava je bilo vaše ciljanje. To celoten položaj preokvirja iz “ubij ga” v “boljše ga usmeri”. Kako zgraditi te segmente, je priročnik zase: kako najti prave kupce za izdelke, ki niso najbolje prodajani.

4. Umeščanje in izobraževanje

Nekateri izdelki se ne prodajajo, ker kupec ne razume, čemu so namenjeni ali zakaj bi jih želel. To ni rešitev za ceno ali sliko – je vrzel v razlagi.

Preizkusite kratek kos izobraževanja. Navodilo, primer uporabe, primerjavo z znano alternativo, dvovrstično “tukaj je, zakaj to obstaja”. Poseben kuhinjski pripomoček, tehnična komponenta, nenavaden material – ti pogosto obtičijo, dokler nekdo ne razloži njihovega smisla, in tedaj se premaknejo. Če malo konteksta dvigne prodajo, ste imeli težavo z razumevanjem, ne s povpraševanjem. Celovitejši pristop je v uporaba izobraževanja o izdelkih za boljšo prodajo zaloge.

5. Združevanje v komplete

Zadnji poceni preizkus: ali izdelek deluje kot del nečesa, ne pa sam zase? Počasi prodajan izdelek je lahko naravni spremljevalec močnega prodajnika – dodatek, polnilo, stvar, ki dopolni komplet.

Poskusite ga povezati s priljubljenim izdelkom, bodisi kot popustni komplet bodisi preprosto kot priporočen dodatek ob nakupu. Če se dobro prodaja kot priloga, ne pa sam zase, v mislih vašega kupca nikoli ni bil samostojen izdelek in pravkar ste našli njegovo pravo polico.

Primer iz trgovine

Recimo, da imate na zalogi obteženo odejo za 55 EUR, ki se že tri mesece ne premakne. Lahka poteza je razprodajno znižanje. Namesto tega opravite zaporedje.

Preizkus cene: znižate jo s 55 EUR na 49 EUR in dodate jasnejšo vrstico o dostavi. Majhen dvig, ne dovolj. Preizkus strani: raven posnetek izdelka zamenjate z življenjskim posnetkom odeje na kavču in prepišete uvodno vrstico tako, da začnete s koristjo namesto z gramsko težo. Stopnja pretvorbe na strani se z nizke osnove približno podvoji. Preizkus občinstva sploh ni bil potreben – ovira je bila ves čas stran. Trije meseci “ta izdelek je zguba” so se izkazali za eno šibko fotografijo. (Ponazoritveno – vaš izdelek in številke se bodo razlikovali, a nauk se ponavlja: vzrok je navadno ceneje popraviti kot odpis, ki ga sproži.)

Kako pošteno izmeriti preizkus

Stopnja pretvorbe na strani izdelka, pred in po vsaki spremembi – najčistejši pokazatelj, ali je rešitev prijela.
Prodaja na teden v poštenem obdobju. Spremembi dajte vsaj teden; en tih dan ne dokaže ničesar.
Stopnja dodajanja v košarico proti dokončanju nakupa – pove vam, ali je ovira na strani ali pri košarici.
Ena spremenljivka naenkrat. Če hkrati spremenite ceno in sliko in prodaja naraste, o tem, katera od njiju je zaslužna, niste izvedeli ničesar. Potrpežljivost tukaj je tisto, kar preizkus naredi vrednega izvedbe.

Kako se v to vključi Omnisend

Dva od teh preizkusov – občinstvo in združevanje v komplete – sta vedenjska, in tam si e-poštno in SMS orodje prisluži svoje mesto. V svojih trgovinah uporabljam Omnisend, izbran po preizkušanju proti Klaviyu, da preizkus občinstva opravim, ne da bi se dotaknil celega seznama: zgradim segment ljudi, ki so kupili sorodni izdelek, počasi prodajan izdelek pošljem samo njim in preberem rezultat. Za preizkus kompleta njegovi bloki s priporočili izdelkov omogočajo, da obtičani izdelek ponudite kot dodatek kupcem močnega prodajnika in opazujete, ali se pripne. Omnisend je partner Shopimation prek partnerskega programa; uporabljam ga vsak dan in ga priporočam na tej podlagi, brezplačni paket pa pokrije majhen preizkusni segment, preden se čemur koli zavežete.

Orodje preizkuša povpraševanje – ne ustvarja ga. Če izdelek pade na vseh zgornjih preizkusih, vključno s pošteno priložnostjo pri pravem občinstvu, ste opravili delo in odpis je zdaj obveščena odločitev namesto ugibanja.

Vaš naslednji korak

Izberite svoj najbolj nadležen počasi prodajan izdelek in ta teden opravite prvi preizkus: prilagodite ceno in vrstico o dostavi ter mu dajte sedem dni. Če se nič ne premakne, delajte se navzdol po seznamu – stran, občinstvo, umeščanje, komplet – in se ustavite v trenutku, ko se kaj dvigne. Če preživi vseh pet nedotaknjen, je odpis zaslužen, in naslednje vprašanje je, ali ga umakniti tudi iz svojih avtomatiziranih priporočil: kdaj naj se počasi prodajan izdelek odstrani iz avtomatiziranih priporočil.

Slow-Moving Inventory

What to Test Before Writing Off a Product

Posted on July 18, 2026 by

Before you liquidate a slow product or pull it from the catalog, run five cheap tests in order: price, the product page, the audience, positioning, and bundling. Most of them cost nothing but a week and a little attention, and they answer the only question that matters — is this a bad product, or a good product being sold badly? Writing off a product is a permanent decision with a real cost: the cash you tied up buying it, the markdown you’ll eat clearing it, and the quiet possibility that it would have sold fine with one thing changed. So test before you write. This page is the testing sequence, in the order that finds the cheapest fix first.

The real cost of writing a product off too early

A write-off feels decisive, and decisive feels good when a product has been nagging at you for months. But a premature write-off is one of the more expensive mistakes a store makes, precisely because it looks like closure.

Think about what you’re actually giving up. You paid for the inventory. If you clear it at a steep discount, you eat margin on every unit. If you drop it from the store, you eat the whole cost. And you lose the information — you’ll never know whether a €4 change to the shipping line or a better second photo would have turned it around, because you removed the product before finding out. The instinct to “just get rid of it” is understandable when you’re staring at stock that won’t move. It’s also how good products die for bad reasons.

Why “it doesn’t sell” isn’t a diagnosis

“It doesn’t sell” describes a symptom. It tells you nothing about the cause, and there are at least five common causes, each with a different fix.

A product can fail to sell because it’s priced wrong, because the page doesn’t do it justice, because the wrong people are seeing it, because it’s positioned against the wrong alternative, or because it needs a companion product to make sense. Only one cause — genuinely no market — justifies a write-off. The other four are fixable, often cheaply. Reaching for the markdown or the delete button treats every one of those as the same problem, which is why so much clearable, sellable stock gets dumped. Sometimes the honest answer really is that you bought the wrong product, and that’s a distinct situation worth naming clearly — I’ve drawn that line in the difference between a slow seller and the wrong product.

The tests, cheapest first

Run these in order. Each one is designed to rule a cause in or out before you spend money. Stop as soon as one moves the numbers — you’ve found your answer.

1. Price

Price is the fastest thing to test and the most common single fix, so start here. But “test price” doesn’t only mean “lower it.”

Try a modest change — up or down — for two weeks and watch the conversion rate rather than the raw order count. Down is obvious. Up is the one people skip: an underpriced product can read as cheap or low-quality, especially in categories where price signals value, and nudging it up sometimes lifts sales. Also check the shipping line, because the number that stops a purchase is often the total at checkout, not the price on the page. A €39 item with a €12 bulky-shipping charge revealed late is a price problem hiding as a product problem. Before you commit to a permanent cut, read why a price cut is not always the best fix for slow inventory — a discount is easy to give and hard to take back.

2. The product page

If price checks out, look at what the visitor actually sees. A product with reasonable traffic and a poor conversion rate is usually losing people on the page itself.

Change one thing at a time so you know what worked:

The main image. Swap it for a clearer, better-lit, or in-context shot. This alone moves conversion more often than any other single page change.
The first two lines of the description. Lead with what it does for the buyer, not the spec sheet.
The obvious question left unanswered. Size, material, compatibility, what’s in the box. If a customer has to email to ask, most won’t — they’ll leave.

Whether the page is even the issue depends on whether people are arriving at all. A page can’t convert traffic it isn’t getting, and that split — traffic problem versus conversion problem — is the first fork to settle: how to identify whether a slow seller has a traffic or conversion problem.

3. The audience

A good product shown to the wrong people looks exactly like a bad product. So test who’s seeing it.

Instead of pushing the item to your whole list, send it to a small segment picked for genuine relevance — people who bought something adjacent, or who browsed the category. If a narrow, well-chosen segment buys at a healthy rate while the broad audience ignored it, the product is fine; your targeting was the problem. That reframes the whole situation from “kill it” to “aim it better.” How to build those segments is its own playbook: how to find the right customers for products that are not best-sellers.

4. Positioning and education

Some products don’t sell because the customer doesn’t understand what they’re for or why they’d want it. That’s not a price or image fix — it’s an explanation gap.

Test a short piece of education. A how-to, a use-case, a comparison against the familiar alternative, a two-line “here’s why this exists.” A specialty kitchen tool, a technical component, an unusual material — these often sit until someone explains the point of them, at which point they move. If a bit of context lifts sales, you had a comprehension problem, not a demand problem. There’s a fuller approach in using product education to improve sell-through.

5. Bundling

The last cheap test: does the product work as part of something rather than alone? A slow item can be a natural companion to a strong seller — the accessory, the refill, the thing that completes the set.

Try pairing it with a popular product, either as a discounted bundle or simply as a recommended add-on at the point of purchase. If it sells readily as an attachment but not on its own, it was never a standalone product in your customer’s mind, and you’ve just found its real shelf.

A store example

Say you stock a €55 weighted blanket that hasn’t moved in three months. The easy call is a clearance markdown. Instead you run the sequence.

Price test: you nudge it from €55 to €49 and add a clearer shipping line. Small lift, not enough. Page test: you replace the flat product shot with a lifestyle image of the blanket on a sofa and rewrite the opening line to lead with the benefit rather than the gram weight. Conversion rate on the page roughly doubles from a low base. Audience test wasn’t even needed — the page was the block all along. Three months of “this product is a dud” turned out to be one weak photo. (Illustrative — your product and numbers will differ, but the lesson repeats: the cause is usually cheaper to fix than the write-off it triggers.)

How to measure a test honestly

Conversion rate on the product page, before and after each change — the cleanest read on whether the fix landed.
Sales per week over a fair window. Give a change at least a week; a single quiet day proves nothing.
Add-to-cart rate versus checkout completion — tells you whether the block is on the page or at the cart.
One variable at a time. If you change price and image together and sales rise, you’ve learned nothing about which one did it. Patience here is what makes the test worth running.

Where Omnisend fits

Two of these tests — audience and bundling — are behavioral, and that’s where an email and SMS tool earns its place. In my own stores I use Omnisend, chosen after testing it against Klaviyo, to run the audience test without touching my whole list: build a segment of people who bought an adjacent product, send the slow item only to them, and read the result. For the bundle test, its product-recommendation blocks let you offer the stuck item as an add-on to buyers of a strong seller and watch whether it attaches. Omnisend is an affiliate partner of Shopimation; I use it daily and recommend it on that basis, and the free tier covers a small test segment before you commit to anything.

The tool tests demand — it doesn’t manufacture it. If the product fails every test above, including a clean shot at the right audience, you’ve done the work and the write-off is now an informed decision instead of a guess.

Your next step

Pick your most nagging slow product and run test one this week: adjust the price and the shipping line, and give it seven days. If nothing moves, work down the list — page, audience, positioning, bundle — and stop the moment something lifts. If it survives all five untouched, the write-off is earned, and the next question is whether to also pull it from your automated recommendations: when should a slow-moving product be removed from automated recommendations.