Zadržanje in zvestoba kupcev

Kakšna je zdrava stopnja ponovnih nakupov glede na vrsto izdelka?

Posted on July 19, 2026 by

Ni ene same zdrave stopnje ponovnih nakupov in vsak članek, ki vam v roke stisne eno natančno merilo, ugiba. Kako je videti “zdravo”, je skoraj v celoti določeno s tem, kaj prodajate: kako hitro izdelek poide, kako dolg je naravni nakupni cikel in koliko vašega kataloga vabi k drugemu nakupu. Trgovina s kavo, kjer vrečka zdrži tri tedne, in trgovina s pohištvom, kjer kavč zdrži desetletje, sta lahko obe povsem zdravi s stopnjama ponovnih nakupov, ki se razlikujeta za faktor pet ali več. Zato koristno vprašanje ni “kolikšno je povprečje?” — je “glede na cikel porabe in katalog mojega izdelka, kaj bi moji kupci logično počeli, in ali to počnejo?” Ta članek vam da to razmišljanje, vrsto izdelka za vrsto izdelka, in način, kako postaviti izhodišče iz svojih podatkov namesto iz blog objave nekoga drugega.

Najprej se dogovorite, kaj merite

Stopnja ponovnih nakupov je običajno opredeljena kot delež kupcev, ki so oddali dve ali več naročil znotraj nekega okna. Okno je tam, kjer se primerjave tiho zlomijo. “Stopnja ponovnih nakupov”, merjena čez ves čas, bo vedno videti bolje kot tista, merjena čez 12 mesecev, in trgovina, ki trguje šest let, bo pri številki za ves čas premagala dveletno trgovino, ne da bi bila kaj boljša pri zadržanju.

Dve pravili, preden karkoli primerjate:

Fiksirajte okno. Dvanajst mesecev na kohorto kupcev je razumna privzeta izbira za večino trgovin.
Ne mešajte stopnje ponovnih nakupov (delež kupcev, ki so se vrnili) z deležem prihodka od ponovnih nakupov (delež prodaje od vračajočih se kupcev). Drugi se dvigne, ko pridobivanje pade — trgovina, ki krvavi nove kupce, lahko opazuje, kako njen delež ponovnih nakupov plezá, medtem ko se posel krči. To je resnično zavajajoča številka, če jo bereš sam.

Zakaj googlanje merila ne pomaga

Objavljene številke “povprečne stopnje ponovnih nakupov v spletni trgovini” mešajo naročniške britvice z trgovinami z vzmetnicami, povprečje teh dveh poslov pa ne opisuje nobenega. Onkraj problema mešanja redko veste, kako je bila številka izračunana — okno, ali so bili izključeni veleprodajni računi, ali gre za stopnjo ali delež prihodka. Primerjanje vaše trgovine s tem je kot primerjanje vaše vožnje v službo s povprečno hitrostjo vseh vozil, vključno s kolesi in tovornimi vlaki.

Merilo, ki šteje, je vaša lastna zgodovina: kohorte tega kvartala proti lanskim. Več o gradnji tega spodaj.

Kaj v resnici določa vašo naravno stopnjo ponovnih nakupov

Štiri značilnosti izdelka opravijo večino dela. Predstavljajte si jih kot gumbe:

Hitrost porabe. Se izdelek porabi? Kava, prehranski dodatki, hrana za hišne ljubljenčke in nega kože se praznijo po razporedu. Kavč ne. Hitreje ko poide, višji je strop ponovnih nakupov.

Dolžina nakupnega cikla. Tudi nepotrošno blago ima cikle — moda se osvežuje po sezonah, otroški čevlji po sunkih rasti, smučarska oprema po zimah. Daljši ko je cikel, daljše mora biti vaše okno merjenja, preden “še ni drugega naročila” kaj pomeni.

Širina kataloga. Enoizdelčna trgovina nima kam usmeriti drugega nakupa, razen če se izdelek izprazni. Trgovina s 400 SKU-ji v dopolnjujočih se kategorijah da vsakemu kupcu naslednjo stvar za nakup. Trajen izdelek s širokim katalogom lahko preseže ponovne nakupe potrošnega izdelka s katalogom enega samega.

Cena in premislek. Dokup za 12 € je refleks. Nakup za 900 € je odločitev, pogosto gospodinjska. Trgovine z visokovrednim blagom imajo lahko zveste kupce, ki ponovijo nakup vsakih nekaj let — nevidne v 12-mesečnem oknu, a vseeno resnične.

Razmišljanje skozi razpone, vrsta izdelka za vrsto izdelka

Kar sledi, namerno ni tabela meril. To so oblike, ki bi jih pričakoval iz logike izdelka, uokvirjene tako, da lahko svojo trgovino preizkusite proti njeni lastni fiziki. Vse številke so ponazoritveno razmišljanje, ne izmerjeni panožni podatki.

Potrošno blago po razporedu (kava, prehranski dodatki, hrana za ljubljenčke)

Izdelek sam vam pove, kdaj naj se kupec vrne: 30-dnevna zaloga pomeni, da zdrav kupec dokupi približno mesečno. Tu najbistrejša preverba zdravja sploh ni letna stopnja ponovnih nakupov — je to, kolikšen delež prvih kupcev se vrne v enem do dveh ciklih porabe. Če vrečka zdrži šest tednov in večina prvih kupcev do dvanajstega tedna ni dokupila, so preskočili cel cikel in verjetno kupujejo drugje (ali pa izdelek ni zalegel). Za to vrsto izdelka nizka stopnja ponovnih nakupov nikoli ni “normalna za panogo”. Je puščanje in običajno popravljivo — kdaj naj spletna trgovina pošlje opomnik za dopolnjevanje je prva stvar za preveriti.

Prožno dopolnjivo blago (kozmetika, sveče, nega)

To poide, a po raztegljivem razporedu, kupci pa med potjo z veseljem zamenjajo znamke. Pričakujte resnični vzorec ponovnih nakupov, počasnejši od strogo potrošnega blaga in bolj odvisen od tega, kako nepozabni ste med nakupi. Veliko tega poganja novost — nov odtenek ali vonj da obstoječemu kupcu razlog za vrnitev, preden je lonček prazen.

Moda in oblačila

Nič se ne izprazni, zato je vedenje ponovnih nakupov vodeno s katalogom in sezonsko. Zdrava modna trgovina vidi kupce, ki se vrnejo, ko izidejo kolekcije, kar pomeni dva do štiri naravne nakupne trenutke na leto za zavzetega kupca — in dolge tihe odseke, ki niso osip. Zdravje sodite po vračanju iz sezone v sezono, ne po 30-dnevnih oknih. Sezonske trgovine imajo svojo lastno logiko avtomatizacije; kako zgraditi avtomatizacijo zadržanja za sezonske kupce gre globlje.

Trajno in visokovredno blago (pohištvo, gospodinjski aparati, oprema)

Ponovni nakup istega izdelka v enem letu je naravno blizu ničle, in to je v redu — trgovina z vzmetnicami s 5-odstotno dvanajstmesečno stopnjo ponovnih nakupov morda ne dela nič narobe. Zdravje se tu pokaže drugje: pri nakupih dodatkov in dopolnil, širjenju kategorij (kupec mize se vrne po stol), ocenah, priporočilih. Če prodajate trajno blago in vaš katalog nima dodatkov ali sosednjih kategorij, je vaša stopnja ponovnih nakupov omejena z asortimentom, ne z marketingom, in noben email tok tega ne bo spremenil.

Širokokataložne hobi in specializirane trgovine

Materiali za ustvarjanje, ribiška oprema, domače varjenje piva: izdelek je trajajoč interes, zato je potencial ponovnih nakupov visok, tudi ko so posamezni izdelki trajni. Tu povprečna stopnja ponovnih nakupov običajno kaže na odkrivanje — kupci ne vedo, kaj imate — bolj kot na zadovoljstvo. Navzkrižna prodaja in brskalno vodeni tokovi so ponavadi pomembnejši od časa dopolnjevanja.

Kako najti svojo zdravo številko

Preskočite panožno povprečje in v enem popoldnevu zgradite svoje izhodišče:

Kupce združite v mesečne ali kvartalne kohorte glede na datum prvega nakupa.
Za vsako kohorto izmerite delež, ki je oddal drugo naročilo v 30, 90 in 365 dneh.
Kohorte primerjajte med seboj skozi čas. Ravne ali rastoče stopnje drugih naročil pomenijo, da je vaše zadržanje zdravo za vaš izdelek; upadajoč trend je resnični alarm, kakršna koli je že absolutna raven.

Ta trendna črta je vredna več kot katero koli zunanje merilo, ker drži vrsto izdelka, cenovno točko in občinstvo konstantne. Polna metoda je v merjenje zadržanja kupcev po kohortah, in ko poznate svojo tipično vrzel med naročili, kako uporabiti pogostost nakupov za izboljšanje zadržanja pokaže, kako po tem ukrepati.

Kaj storiti, ko poznate svoj cikel

Razlog, da določite svojo naravno stopnjo, je, da vam pove, katera avtomatizacija se izplača:

Potrošno blago: opomniki za dopolnjevanje, umerjeni na iztek, plus spodbuda od prvega do drugega nakupa — spreminjanje prvih kupcev v dolgoročne stranke pokriva to ključno prvo konverzijo.
Moda in sezonsko: objave novih kolekcij preteklim kupcem te kategorije in predsezonska ogrevanja.
Trajno blago: ponakupna vsebina o negi, navzkrižne prodaje dodatkov, umerjene tedne po dostavi, in prošnje za ocene.
Široki katalogi: navzkrižna prodaja kategorij in personalizirana priporočila, saj je ozko grlo odkrivanje.

Dve ali tri številke za spremljanje

Semafor naj ostane majhen: stopnja drugih naročil v 90 dneh po kohorti (vaša glavna zdravstvena črta), mediana dni med prvim in drugim naročilom in — za potrošno blago — delež kupcev, ki dokupijo v 1,5 cikla porabe. Če spremljate delež prihodka od ponovnih nakupov, ga vedno berite ob številu novih kupcev, da vas krčeči se imenovalec ne more zavesti.

Kje se vklopi Omnisend

Merilno polovico tega lahko naredite z izvozi naročil in preglednico. Polovica ukrepanja — doseganje ljudi na njihovi točki cikla — je tam, kjer orodje šteje. V svojih trgovinah zanjo uporabljam Omnisend: segmenti, zgrajeni na številu naročil in datumu zadnjega naročila, in tokovi, ki se sprožijo na čas od nakupa, kar je natanko mehanika, ki jo potrebuje vsaka zgornja vrsta izdelka, le z drugačnimi okni in sporočili. Ne bo vam povedal, kolikšna je vaša zdrava stopnja. Nič ne bo, razen vaših lastnih kohort. (Omnisend je partner v pridruženem programu Shopimation; priporočilo izhaja iz vsakodnevne uporabe.)

Vaš naslednji korak

Izvlecite svojih zadnjih dvanajst mesecev naročil in izračunajte eno številko: delež prvih kupcev, ki so oddali drugo naročilo v 90 dneh, po mesečni kohorti. Nato jo presodite glede na logiko svojega izdelka — cikel porabe, širina kataloga, sezona — ne glede na povprečje neznanca. Če trend upada, začnite tam, kjer vrsta vašega izdelka pravi, da je puščanje.

Customer Retention and Loyalty

What a Healthy Repeat Purchase Rate Looks Like by Product Type

Posted on July 17, 2026July 17, 2026 by

There is no single healthy repeat purchase rate, and any article that hands you one exact benchmark is guessing. What “healthy” looks like is set almost entirely by what you sell: how fast the product runs out, how long the natural buying cycle is, and how much of your catalog invites a second purchase. A coffee store where a bag lasts three weeks and a furniture store where a sofa lasts a decade can both be perfectly healthy with repeat rates that differ by a factor of five or more. So the useful question isn’t “what’s the average?” — it’s “given my product’s consumption cycle and catalog, what should my customers logically be doing, and are they doing it?” This article gives you that reasoning, product type by product type, and a way to set a baseline from your own data instead of someone else’s blog post.

First, agree on what you’re measuring

Repeat purchase rate is usually defined as the share of customers who have placed two or more orders within some window. The window is where comparisons quietly break. “Repeat rate” measured over all time will always look better than one measured over 12 months, and a store that’s been trading for six years will beat a two-year-old store on the all-time number without being any better at retention.

Two rules before you compare anything:

Fix the window. Twelve months per customer cohort is a sane default for most stores.
Don’t confuse repeat purchase rate (share of customers who came back) with repeat share of revenue (portion of sales from returning customers). The second one rises when acquisition falls — a store bleeding new customers can watch its repeat share climb while the business shrinks. It’s a genuinely misleading number if you read it alone.

Why googling a benchmark doesn’t help

Published “average ecommerce repeat rate” figures blend subscription razor blades with mattress stores, and the average of those two businesses describes neither. Beyond the mixing problem, you rarely know how the number was computed — the window, whether wholesale accounts were excluded, whether it’s rate or revenue share. Comparing your store to it is comparing your commute to the average speed of all vehicles, bicycles and freight trains included.

The benchmark that matters is your own history: this quarter’s cohorts against last year’s. More on building that below.

What actually sets your natural repeat rate

Four product characteristics do most of the work. Think of them as dials:

Consumption speed. Does the product get used up? Coffee, supplements, pet food, and skincare empty on a schedule. A sofa doesn’t. The faster the run-out, the higher the ceiling on repeat purchases.

Purchase cycle length. Even non-consumables have cycles — fashion refreshes by season, kids’ shoes by growth spurt, ski gear by winter. The longer the cycle, the longer your measurement window has to be before “no second order yet” means anything.

Catalog breadth. A one-product store has nowhere for a second purchase to go unless the product depletes. A store with 400 SKUs across complementary categories gives every customer a next thing to buy. Durable product plus broad catalog can out-repeat consumable product with a catalog of one.

Price and consideration. A €12 reorder is a reflex. A €900 purchase is a decision, often a household one. High-ticket stores can have loyal customers who repeat every few years — invisible in a 12-month window, real all the same.

Reasoning through the ranges, product type by product type

What follows is deliberately not a benchmark table. These are the shapes I’d expect from the product logic, framed so you can test your store against its own physics. All figures are illustrative reasoning, not measured industry data.

Consumables on a schedule (coffee, supplements, pet food)

The product itself tells you when the customer should return: a 30-day supply means a healthy customer reorders roughly monthly. Here the sharpest health check isn’t the annual repeat rate at all — it’s what share of first-time buyers come back within one to two consumption cycles. If a bag lasts six weeks and most first-timers haven’t reordered by week twelve, they’ve skipped a full cycle and are probably buying elsewhere (or the product didn’t land). For this product type, a low repeat rate is never “normal for the industry.” It’s a leak, and usually a fixable one — when should an ecommerce store send a replenishment reminder is the first thing to check.

Flexible replenishables (cosmetics, candles, grooming)

These run out, but on a stretchy schedule, and customers happily switch brands mid-stream. Expect a real repeat pattern, slower than strict consumables, and more dependent on how memorable you are between purchases. Newness drives a lot of it — a new shade or scent gives an existing customer a reason to return before the jar is empty.

Fashion and apparel

Nothing depletes, so repeat behavior is catalog-driven and seasonal. A healthy fashion store sees customers return when collections drop, which means two to four natural buying moments a year for an engaged customer — and long quiet stretches that are not churn. Judge health by season-over-season return, not by 30-day windows. Seasonal stores have their own automation logic; how to build retention automation for seasonal customers goes deeper.

Durables and high-ticket (furniture, appliances, equipment)

Same-product repeat within a year is naturally near zero, and that’s fine — a mattress store with a 5% twelve-month repeat rate may be doing nothing wrong. Health here shows up elsewhere: accessory and add-on purchases, category expansion (the desk buyer returning for the chair), reviews, referrals. If you sell durables and your catalog has no accessories or adjacent categories, your repeat rate is capped by assortment, not by marketing, and no email flow will change that.

Broad-catalog hobby and specialty stores

Craft supplies, fishing gear, home brewing: the product is an ongoing interest, so repeat potential is high even when individual items are durable. Here a mediocre repeat rate usually points at discovery — customers don’t know what you carry — rather than satisfaction. Cross-sell and browse-based flows tend to matter more than replenishment timing.

How to find your healthy number

Skip the industry average and build your own baseline in an afternoon:

Group customers into monthly or quarterly cohorts by first-purchase date.
For each cohort, measure the share that placed a second order within 30, 90, and 365 days.
Compare cohorts against each other over time. Flat or rising second-order rates mean your retention is healthy for your product; a sagging trend is the real alarm, whatever the absolute level.

That trend line is worth more than any external benchmark because it holds product type, price point, and audience constant. The full method is in measuring customer retention by cohort, and once you know your typical gap between orders, how to use purchase frequency to improve retention shows how to act on it.

What to do once you know your cycle

The reason to pin down your natural rate is that it tells you which automation earns its keep:

Consumables: replenishment reminders timed to run-out, plus a first-to-second-purchase push — turning first-time buyers into long-term customers covers that critical first conversion.
Fashion and seasonal: new-collection announcements to past buyers of that category, and pre-season warm-ups.
Durables: post-purchase care content, accessory cross-sells timed weeks after delivery, and review requests.
Broad catalogs: category cross-sell and personalized recommendations, since discovery is the bottleneck.

The two or three numbers to watch

Keep the scoreboard small: second-order rate within 90 days by cohort (your main health line), median days between first and second order, and — for consumables — the share of customers reordering within 1.5 consumption cycles. If you track repeat revenue share, always read it next to new-customer counts so a shrinking denominator can’t flatter you.

Where Omnisend fits

The measurement half of this you can do with order exports and a spreadsheet. The acting half — reaching people at their cycle point — is where a tool matters. In my stores I use Omnisend for it: segments built on order count and last-order date, and flows that fire on time-since-purchase, which is exactly the mechanic every product type above needs, just with different windows and messages. It won’t tell you what your healthy rate is. Nothing will, except your own cohorts. (Omnisend is an affiliate partner of Shopimation; the recommendation comes from daily use.)

Your next step

Pull your last twelve months of orders and compute one number: the share of first-time buyers who placed a second order within 90 days, by monthly cohort. Then judge it against your product’s logic — consumption cycle, catalog breadth, season — not against a stranger’s average. If the trend is sagging, start where your product type says the leak is.