Počasi prodajana zaloga

Kako uporabiti podatke o ogledih za odkrivanje skritega povpraševanja po presežni zalogi

Posted on July 18, 2026 by

Vaša presežna zaloga ni ena težava, temveč dve, ki nosita isto etiketo. Nekateri od teh izdelkov dobijo veliko ogledov in skoraj nobenega naročila. Drugih skoraj nihče ne pogleda. Podatki o ogledih vam povedo, v kateri kup sodi obtičan izdelek, in ta odgovor odloči vse o tem, kako ga boste prodali. Da bi odkrili skrito povpraševanje, izvozite podatke o ogledih izdelkov in dodajanju v košarico za vsak počasi prodajan izdelek v zadnjih 60–90 dneh, nato pa jih razvrstite po ogledih. Izdelki z veliko ogledi in nizko prodajo so tisti s povpraševanjem, ki ga še niste izkoristili. Ti so vredni ciljane kampanje. Tisti, ki jih nihče ne pogleda, potrebujejo povsem drugačno rešitev in nobena količina e-pošte jih ne bo rešila, dokler prej ne odpravite težave z vidnostjo.

Ta stran govori o prvem kupu: o branju podatkov o vedenju, da najdete presežno zalogo, ki jo ljudje že želijo, a je ne kupijo.

Napaka: obravnavanje vse počasne zaloge kot nezaželene

Takole se navadno zgodi. Izdelek obtiči. Nadzorna plošča ga označi kot počasnega. Nekdo sklene, da je zguba, in ga bodisi pocenijo po vsej trgovini bodisi tiho pozabijo v kotu skladišča.

Težava je, da “počasen” in “nezaželen” nista isto. Izdelek lahko obtiči iz razlogov, ki nimajo nič opraviti s tem, ali ga kupci želijo: pokopan je na četrti strani zbirke, glavna slika je šibka, cena leži v nerodni vrzeli ali pa ga vidijo napačni ljudje. Povpraševanje je lahko resnično in se vseeno ne pokaže v vašem poročilu o prodaji. To je povpraševanje, ki ga puščate na mizi, in podatki o ogledih so mesto, kjer se skriva.

Zakaj vas poročilo o prodaji samo po sebi zavaja

Poročilo o prodaji šteje samo kupce, ki so prišli vse do blagajne. To je konec zgodbe, ne sredina. Ko izdelek v tem poročilu izgleda počasen, ste že izgubili informacijo, ki jo v resnici potrebujete: koliko ljudi ga je premislilo in odšlo.

Dva izdelka lahko oba prikazujeta deset prodaj ta mesec. Eden je za teh deset naročil dobil 90 ogledov – zdrava 11-odstotna stopnja pretvorbe ogleda v nakup, in počasen je le zato, ker je prometa na strani malo. Drugi je za enakih deset naročil dobil 2.000 ogledov – 0,5-odstotna stopnja, ki pove, da nekaj aktivno odbija ljudi, ki jih očitno dovolj zanima, da kliknejo. Ista prodajna vrstica. Popolnoma različni težavi. Popolnoma različni rešitvi. Poročilo o prodaji obe splošči v eno besedo, “počasen”, in ta beseda vas stane denarja, ker vas usmeri k napačni ročici. Ugotoviti, katero od teh dveh gledate, je diagnostika zase – opisal sem jo v kako ugotoviti, ali ima počasi prodajan izdelek težavo s prometom ali s pretvorbo.

Kje pravzaprav živi skrito povpraševanje

Trije vedenjski signali ločijo resnično povpraševanje od pristnega nezanimanja. Izvozite jih za vsak počasi prodajan izdelek.

Ogledi izdelka. Surovo število ljudi, ki so pristali na strani. Veliko ogledov na obtičanem izdelku je najmočnejši posamezen znak, da zanimanje obstaja in da prodaja nekje pušča po kliku.

Stopnja dodajanja v košarico. Koliko od tistih, ki so si izdelek ogledali, ga je dodalo v košarico? Izdelek s spodobnimi ogledi in spodobno stopnjo dodajanja, a malo naročili, se ustavlja na blagajni – cena, strošek dostave ali dvom pri plačilu. Izdelek z ogledi, a skoraj brez dodajanj, se ustavlja že na sami strani – slika, opis, prikazana cena ali neskladje med tem, kar je obiskovalec pričakoval, in tem, kar je našel.

Ponovni ogledi brez nakupa. Tiho zlato. Ko se isti ljudje vračajo na stran izdelka dva- ali trikrat in nikoli ne kupijo, je to oklevanje, ne brezbrižnost. Želijo ga. Nekaj konkretnega jih zadržuje – navadno cena, negotovost ali čakanje na razlog za ukrepanje. Ta skupina je najbolj obnovljivo občinstvo, ki ga imate za presežno zalogo, in večina trgovin je nikoli ne pogleda.

Izdelek brez ogledov, brez dodajanj in brez ponovnih obiskov nima skritega povpraševanja, ki bi ga lahko našli. Bodite pošteni glede tega. Nekaj presežne zaloge je presežna zaloga zato, ker je bila odločitev o nakupu napačna, in podatki o ogledih vam bodo to jasno povedali. Ko izdelka res nihče nikoli ne vidi, je težava višje v toku, v tem, kako ga vaša trgovina prikazuje – to je vprašanje predstavitve izdelkov, obravnavano v zakaj lahko počasi prodajana zaloga razkrije težavo pri predstavitvi izdelkov.

Kako signal spremeniti v kratek seznam

Ne poskušajte analizirati vsega naenkrat. Delajte se navzdol po kratkem seznamu.

Izvozite oglede izdelkov, dodajanja v košarico in naročila za vsak počasi prodajan izdelek v zadnjih 60–90 dneh. Četrtletje zgladi tedenski šum.
Za vsak izdelek izračunajte dve razmerji: ogled proti košarici in košarica proti naročilu.
Označite vse, kar ima nadpovprečno število ogledov in nekje v lijaku pokvarjeno razmerje. To so vaši kandidati za skrito povpraševanje.
Odložite izdelke brez sleherih ogledov. Ti niso težava kampanje, temveč težava odkritja, in e-pošta jih ne bo popravila.

Ostane vam peščica izdelkov, ki ljudi že zanimajo. Tam ima kampanja resnično možnost, saj povpraševanja ne ustvarjate – čistite vse, kar blokira odločitev, ki je bila že napol sprejeta.

Kaj avtomatizirati: dosezite ljudi, ki so že pogledali

Kupci, ki so si ogledali počasi prodajan izdelek in ga niso kupili, so vam natančno povedali, kaj želijo. Občinstva vam ni treba ugibati – vedenje je segment že narisalo namesto vas. To je opuščanje brskanja, usmerjeno na zalogo, ki jo morate konkretno premakniti.

Tukaj je tok, ki deluje za izdelek s presežno zalogo, ki ima veliko ogledov in malo naročil:

Sprožilec: kupec si ogleda izdelek (ali počasi prodajano zbirko) dvakrat ali večkrat v določenem obdobju, ne da bi kupil. Ponovni ogled je signal povpraševanja – en pogled je brskanje, trije so zanimanje.
Segment: tisti, ki so si ta izdelek ogledali v zadnjih 14–30 dneh in ga niso kupili. Izključite vse, ki ga že imajo ali imajo kaj, kar ga nadomešča.
Časovnica: prvo sporočilo nekaj ur do en dan po kvalificiranem ogledu, dokler je izdelek še v mislih. Drugi, nežnejši sunek dva ali tri dni pozneje, če niso ukrepali.
Kanal: e-pošta za podrobnosti in slike; SMS samo, če je obiskovalec znan stik in izdelek upravičuje nujnost.
Vsebina: nagovorite konkretno oklevanje, ki ga podatki nakazujejo. Če so se zataknili na strani izdelka (nizko dodajanje v košarico), začnite z boljšo informacijo – kako se uporablja, s čim se ujema, zakaj je vreden. Če so se zataknili pri košarici, obravnavajte oviro na blagajni – pomirite glede dostave, vračil ali zaloge. Ne začnite s popustom; morda vam ga ni treba porabiti, in obstaja resen razlog, da ga zadržite, opisan v zakaj znižanje cene ni vedno najboljša rešitev za počasi prodajano zalogo.
Cilj: pretvoriti obstoječe zanimanje, ne da bi pocenili celotno trgovino. Ena povrnjena prodaja od nekoga, ki si je izdelek ogledal trikrat, je čist najden prihodek.

Bistvo avtomatizacije je, da se ti sprožilci sprožajo neprekinjeno. Logiko nastavite enkrat in vsak prihodnji obiskovalec tega obtičanega izdelka samodejno pade v tok – brez ročne kampanje vsakič.

Primer iz trgovine

Recimo, da vodite trgovino z izdelki za dom in sedite na 300 kosih srednje cenovne keramične svetilke. Presežna zaloga, označena kot počasna. Nagon je 30-odstotno znižanje.

Preden to storite, izvozite podatke. V zadnjem četrtletju je svetilka dobila 1.400 ogledov izdelka – res veliko za vaš katalog – s 6-odstotno stopnjo dodajanja v košarico in skoraj nobenim dokončanim naročilom. Torej jo ljudje najdejo, precej si jih jo želi, nato pa izginejo na blagajni. Ta vzorec ne pravi “te svetilke nihče noče”. Pravi, da se nekaj zlomi pri košarici. Preverite in ugotovite, da dostava ene same kosovne svetilke stane 12 EUR, razkrita šele na blagajni, pri izdelku za 39 EUR. Povpraševanje je bilo ves čas tam. Rešitev je bila brezplačna dostava za ta izdelek ali jasnejša vrstica o dostavi na strani – ne popust po vsej trgovini, ki bi maržo podaril 300 kupcem, ki bi morda plačali polno ceno, če bi bila dostava urejena. (Ponazoritveni primer – vaše lastne številke se bodo razlikovale, a oblika napake je pogosta.)

Kako izmeriti, ali je delovalo

Stopnja ogleda v naročilo za ciljni izdelek pred in po. To je številka, ki jo dejansko poskušate premakniti.
Prihodek na prejemnika iz toka, sproženega z brskanjem – pove vam, ali si kampanja zasluži svojo pošiljko, ne da le odhaja.
Prodaja zaloge izdelka s presežno zalogo v obdobju kampanje v primerjavi s prejšnjim obdobjem.
Ohranjena marža. Če ste zalogo počistili brez splošnega popusta, zabeležite, koliko marže ste obdržali v primerjavi s tem, kolikor bi vas stalo znižanje.

Če je bilo ogledov veliko in tok kljub temu ni pretvarjal, je ovira globlja, kot jo lahko popravi opomnik – verjetno je to cena ali sama stran izdelka. To je vaš znak, da preizkusite temelje, kar je ločena naloga: kaj preizkusiti, preden izdelek odpišete.

Kako se v to vključi Omnisend

Za to potrebujete dve stvari, ki vam ju navadno e-poštno orodje ne da: podatke o brskanju na ravni izdelka, povezane s posameznimi stiki, in avtomatizacije, ki se sprožijo na to vedenje. V svojih trgovinah uporabljam Omnisend, potem ko sem ga preizkusil proti Klaviyu, in opuščanje brskanja je eden od tokov, ki jih obvlada čisto – sledi, katere izdelke si je ogledal znan stik, omogoča vam, da zgradite segment ljudi, ki so si ogledali določen izdelek ali zbirko, a niso kupili, in samodejno sproži sporočilo. Točno tisti izdelek, ki so si ga ogledali, lahko z dinamičnim blokom vstavite v e-pošto, tako da je sporočilo o svetilki, na katero so se vračali, ne pa splošno “pogrešamo vas”. Omnisend je partner Shopimation prek partnerskega programa; priporočam ga iz vsakodnevne uporabe, brezplačni paket pa je dovolj, da to zgradite in preizkusite na enem samem izdelku s presežno zalogo, preden ga razširite.

Ena poštena omejitev: orodje razkrije povpraševanje, ki že obstaja, in ukrepa na njem. Ne more ustvariti povpraševanja po izdelku, ki ga nihče ne pogleda. Za te je odgovor višje v toku, ne v e-poštnem toku.

Vaš naslednji korak

Ta teden izvozite eno poročilo: oglede izdelkov, dodajanja v košarico in naročila za vaših deset najpočasnejših izdelkov v zadnjih 90 dneh. Razvrstite po ogledih. Vrstice z veliko ogledi in malo naročili so vaše skrito povpraševanje – začnite pri tisti z najvišjimi ogledi in okoli nje zgradite en tok, sprožen z brskanjem. Ko boste vedeli, kateri obtičani izdelki imajo zanimanje in kateri ne, se odločite, kako močno promovirati vsakega z kdaj naj počasi prodajana zaloga vstopi v avtomatizirano promocijo.

Slow-Moving Inventory

Using Browsing Data to Find Hidden Demand for Overstock

Posted on July 18, 2026 by

Your overstock isn’t one problem — it’s two piles wearing the same label. Some of those products get plenty of views and almost no orders. Others barely get looked at. Browsing data tells you which pile a stuck product is in, and that answer decides everything about how you sell it. To find hidden demand, pull product-view and add-to-cart data for every slow item over the last 60–90 days, then sort by views. The products with high views and low sales are the ones with demand you haven’t captured yet. Those are worth a targeted campaign. The ones nobody looks at need a different fix entirely, and no amount of email will save them until you solve the visibility problem first.

This page is about the first pile: reading behavioral data to find overstock that people already want but aren’t buying.

The mistake: treating all slow inventory as unwanted

Here’s what usually happens. A product sits. The dashboard flags it as slow. Somebody decides it’s a dud, and it either gets marked down across the whole store or quietly forgotten in a warehouse corner.

The trouble is that “slow” and “unwanted” are not the same thing. A product can be sitting for reasons that have nothing to do with whether customers want it: it’s buried on page four of a collection, the main image is weak, the price sits in an awkward gap, or the people seeing it are the wrong people. Demand can be real and still not show up in your sales report. That’s the demand you’re leaving on the table — and browsing data is where it hides.

Why the sales report alone misleads you

A sales report only counts the customers who made it all the way to checkout. It’s the end of the story, not the middle. By the time a product looks slow in that report, you’ve already lost the information you actually need: how many people considered it and walked away.

Two products can both show ten sales this month. One got 90 views to reach those ten orders — a healthy 11% view-to-purchase rate, and it’s slow only because traffic to the page is thin. The other got 2,000 views for the same ten orders — a 0.5% rate that says something is actively repelling people who are clearly interested enough to click. Same sales line. Completely different problems. Completely different fixes. The sales report flattens both into one word, “slow,” and that word costs you money because it points you at the wrong lever. Figuring out which of the two you’re looking at is its own diagnostic — I’ve laid it out in how to identify whether a slow seller has a traffic or conversion problem.

Where the hidden demand actually lives

Three behavioral signals separate real demand from genuine disinterest. Pull them per product for your slow SKUs.

Product views. The raw count of people who landed on the page. High views on a stuck product is the single strongest sign that interest exists and the sale is leaking somewhere after the click.

Add-to-cart rate. Of the people who viewed, how many added it? A product with decent views and a decent add rate but few orders is stalling at checkout — price, shipping cost, or a payment doubt. A product with views but almost no adds is stalling on the page itself — the image, the description, the price shown, or a mismatch between what the visitor expected and what they found.

Repeat views without purchase. The quiet gold. When the same people come back to a product page two or three times and never buy, that’s hesitation, not indifference. They want it. Something specific is holding them back — usually price, uncertainty, or waiting for a reason to act. This group is the most recoverable audience you have for overstock, and most stores never look at it.

A product with no views, no adds, and no repeat visits has no hidden demand to find. Be honest about that. Some overstock is overstock because the buying decision was wrong, and browsing data will tell you so plainly. When a product genuinely never gets seen, the problem is upstream in how your store surfaces it — that’s a merchandising question, covered in why slow inventory can reveal a merchandising problem.

Turning the signal into a shortlist

Don’t try to analyze all of it at once. Work down a short list.

Export product views, add-to-carts, and orders for every slow SKU over the last 60–90 days. A quarter smooths out weekly noise.
Calculate two ratios per product: view-to-cart and cart-to-order.
Flag anything with above-average views and a broken ratio somewhere in the funnel. Those are your hidden-demand candidates.
Set aside the products with no views at all. They’re not a campaign problem; they’re a discovery problem, and email won’t fix them.

What you’re left with is a handful of products people are already interested in. That’s where a campaign has a real chance, because you’re not manufacturing demand — you’re clearing whatever is blocking a decision that was half-made already.

What to automate: reach the people who already looked

The customers who viewed a slow product and didn’t buy have told you exactly what they want. You don’t need to guess an audience — behavior already drew the segment for you. This is browse abandonment, pointed at inventory you specifically need to move.

Here’s a flow that works for a high-view, low-order overstock item:

Trigger: a customer views the product (or a slow-moving collection) two or more times within a set window without purchasing. The repeat view is the demand signal — one glance is browsing, three is interest.
Segment: viewers of that product in the last 14–30 days who haven’t bought it. Exclude anyone who already owns it or something that replaces it.
Timing: first message a few hours to a day after the qualifying view, while the product is still in mind. A second, softer nudge two or three days later if they didn’t act.
Channel: email for the detail and the imagery; SMS only if the visitor is a known contact and the item warrants urgency.
Content: speak to the specific hesitation the data implies. If they stalled on the product page (low add-to-cart), lead with better information — how it’s used, what it pairs with, why it’s worth it. If they stalled at cart, address the checkout friction — reassure on shipping, returns, or stock. Don’t open with a discount; you may not need to spend the margin, and there’s a real case for holding it, laid out in why a price cut is not always the best fix for slow inventory.
Goal: convert existing interest without discounting the whole store. One recovered sale from someone who viewed three times is pure found revenue.

The point of automation here is that these triggers fire continuously. You set the logic once, and every future browser of that stuck product drops into the flow automatically — no manual campaign each time.

A store example

Say you run a homewares store and you’re sitting on 300 units of a mid-priced ceramic lamp. Overstocked, flagged slow. The instinct is a 30% markdown.

Before you do that, you pull the data. Over the last quarter the lamp got 1,400 product views — genuinely high for your catalog — with a 6% add-to-cart rate and almost no completed orders. So people find it, a fair number want it, and then they vanish at checkout. That pattern doesn’t say “nobody wants this lamp.” It says something breaks at the cart. You check and find shipping on a single bulky lamp costs €12, revealed only at checkout, on a €39 product. The demand was there the whole time. The fix was free shipping on that SKU or a clearer shipping line on the page — not a store-wide discount that would have handed margin to 300 buyers who might have paid full price with the shipping sorted. (Illustrative example — your own numbers will differ, but the shape of the mistake is common.)

How to measure whether it worked

View-to-order rate on the target product before and after. This is the number you’re actually trying to move.
Revenue per recipient from the browse-triggered flow — tells you the campaign is earning its send, not merely going out.
Sell-through on the overstock SKU over the campaign window versus the prior period.
Margin retained. If you cleared the stock without a blanket discount, note how much margin you kept versus what a markdown would have cost.

If views were high and the flow still didn’t convert, the block is deeper than a reminder can fix — likely price or the product page itself. That’s your signal to test the fundamentals, which is a separate exercise: what to test before writing off a product.

How Omnisend fits

To run this you need two things a plain email tool won’t give you: product-level browsing data tied to individual contacts, and automations that trigger on that behavior. I use Omnisend in my own stores after testing it against Klaviyo, and browse abandonment is one of the flows it handles cleanly — it tracks which products a known contact viewed, lets you build a segment of people who looked at a specific SKU or collection without buying, and fires the message automatically. You can drop the exact product they viewed into the email with a dynamic block, so the message is about the lamp they kept coming back to, not a generic “we miss you.” Omnisend is an affiliate partner of Shopimation; I recommend it from daily use, and the free tier is enough to build and test this on a single overstock product before you scale it.

One honest limit: the tool surfaces and acts on demand that’s already there. It can’t create demand for a product nobody looks at. For those, the answer is upstream, not in the email flow.

Your next step

Pull one export this week: product views, add-to-carts, and orders for your ten slowest SKUs over the last 90 days. Sort by views. The high-view, low-order rows are your hidden demand — start with the single highest one and build one browse-triggered flow around it. Once you know which stuck products have interest and which don’t, decide how hard to promote each one with when should slow-moving stock enter an automated promotion.