Od prvega do drugega nakupa

Zakaj nekateri kanali pridobivanja ustvarijo več drugih nakupov kot drugi

Posted on July 18, 2026 by

Nekateri kanali pridobivanja ustvarijo več drugih nakupov kot drugi, ker privabijo kupce z resnično namero in pravim ujemanjem s tem, kar prodajate, medtem ko drugi privabijo enkratne kupce, ki lovijo ugodnost ali kliknejo impulzivni oglas. Kupec, ki vas je našel z iskanjem točno tega, kar ponujate, se čez šest mesecev vede zelo drugače kot tisti, ki je impulzivno kupil iz socialnega oglasa, vodenega s popustom. Enaka vrednost prvega naročila, povsem drugačna stopnja drugega nakupa. Če kanale merite le po stroških na pridobitev in ROAS prvega naročila, ste napol slepi — optimizirate za najcenejšo prvo prodajo, medtem ko spregledate, kateri kanali prinašajo kupce, ki ostanejo. Ta članek pojasni, zakaj razmik obstaja in kako ga najti v svojih številkah.

Kratek odgovor: kanali izbirajo tip kupca

Vsak kanal pridobivanja je filter. Ne dostavlja le prometa; izbira vrsto kupca, z vrsto namere, v vrsti trenutka.

Iskanje ujame nekoga, ki aktivno išče rešitev — visoka namera, pogosto prava potreba. Prijava prek priporočila ali od ust do ust prispe že vnaprej zaupljiva, ogreta s strani osebe, ki jo pozna. Dobro zgrajen email seznam, ki ste ga skozi čas gradili, je poln ljudi, ki so se odločili slišati o vas. Širok socialni oglas, voden s popustom, nasprotno, ujame nekoga sredi drsenja, ki sploh ni nakupoval in je zagrizel, ker je bila cena nizka. Vsi lahko ustvarijo prvo naročilo. Ne ustvarijo istega kupca.

To je mehanizem. Stopnja drugega nakupa v resnici ni lastnost kanala — je lastnost kupca, ki ga kanal privabi, in različni kanali privabljajo kupce z zelo različnimi možnostmi za vrnitev.

Pravi problem: vaš najlepše videti kanal je morda vaš najslabši

Tu to stane denar. Večina trgovin sodi kanale po stroških na pridobitev in donosu na porabo za oglase za tisto prvo naročilo. Po tem semaforju lahko poceni kanal, voden s popustom, izgleda kot vaša zvezda — nizek CPA, hitra prva prodaja, zdrav ROAS prvega naročila.

Nato nikoli ne preverite, kaj ti kupci storijo naprej. Če ta “učinkoviti” kanal ustvarja kupce, ki se skoraj nikoli ne vrnejo, je njegova prava vrednost delček tega, kar nakazuje matematika prvega naročila. Medtem lahko kanal z višjim začetnim CPA — recimo iskanje ali program priporočil — prinese kupce, ki se vrnejo dva- ali trikrat, s čimer je vsak skozi svoje življenje vreden veliko več. Na semaforju prvega naročila izgleda drag. Na semaforju življenjske vrednosti je vaš najboljši vir kupcev.

Optimizirajte le na ekonomiko prvega naročila in sistematično boste premikali proračun proti kanalom, ki prinašajo poceni, nezveste kupce, in stran od kanalov, ki prinašajo zveste. Trgovina izgleda učinkovita, stopnja ponovnega nakupa pa tiho erodira. Zakaj ta ponovni posel šteje več kot prva prodaja, je celoten argument zakaj je drugi nakup pomembnejši, kot se večina lastnikov trgovin zaveda.

Zakaj strošek na pridobitev tega ne vidi

CPA in ROAS prvega naročila odgovarjata na eno vprašanje: kako poceni je ta kanal ustvaril prvo prodajo. Sta koristna in sta nepopolna. Nobeden od njiju ne ve ničesar o tem, kaj se zgodi po prvem naročilu — ali se je kupec vrnil, koliko je nato porabil, ali ga je vredno pridobiti po višji ceni.

Dva kanala z enakim CPA 15 EUR imata lahko stopnji drugega nakupa 12 % in 30 %. Poročilo o prvem naročilu ju kaže kot enaka. Življenjska resničnost kaže, da je eden vreden približno dvakrat toliko kot drugi. Sodite po nepopolni metriki in sprejemali boste samozavestne, dobro utemeljene, napačne odločitve o tem, kam dati proračun. Rešitev je dodati drugi stolpec k temu, kako vrednotite kanale — koliko je kupec vreden skozi čas, tik ob tem, koliko vas je stala prva prodaja.

Kako najti, kateri kanali prinašajo zveste kupce

To je kohortna analiza po viru pridobivanja. Prvo različico lahko zgradite iz podatkov svoje platforme.

Vsakega kupca ob prvem naročilu označite s kanalom pridobivanja. Večina platform za spletno trgovino in email zajame vir — iskanje, plačano družbeno omrežje, priporočilo, email, neposredno in tako naprej. Poskrbite, da je shranjeno na kupcu, ne le na naročilu.
Prve kupce razvrstite v kohorte po kanalu.
Vsaki kohorti dajte enako okno za vrnitev. Primerjajte le kupce, ki so imeli enako časa za vrnitev — na primer “drugi nakup v 180 dneh od prvega naročila”. Kanala, ki ste ga začeli uporabljati šele prejšnji mesec, ni mogoče pošteno primerjati z enim, ki teče vse leto, dokler obe kohorti enako ne dozorita.
Vsako kohorto kanala izmerite na stopnji drugega nakupa, stopnji ponovnega nakupa v 6–12 mesecih in življenjski vrednosti — metrikah, ki vam jih sam CPA nikoli ne bo pokazal.
Kanale razvrstite po zvestobi in življenjski vrednosti, nato pa to postavite ob CPA. Kanali, ki izgledajo najbolje na obeh, so tam, kjer naj se proračun zgošča. Poceni kanali, ki ne zadržijo ničesar, dobijo trd drugi pogled.

Eno svarilo: pripis je zapleten. Kupec se lahko pred nakupom dotakne več kanalov, “zadnji klik” pa pripiše preveč zaslug temu, kar je bilo najbližje nakupu. Ne potrebujete popolnega pripisa, da ukrepate — potrebujete dosleden pripis, uporabljen enako za vsako kohorto, da je primerjava med kanali poštena, tudi če je katera koli posamezna številka nepopolna.

Kaj storiti z odgovorom

Prerazporedite proti zvestobi, ne le proti ceni. Premaknite proračun proti kanalom, ki prinašajo kupce, ki se vračajo, tudi tam, kjer je CPA prvega naročila višji, saj kupujete v resnici življenjsko vrednost.
Popravite, ne nujno opustite, šibke kanale. Poceni kanal, ki prinaša nezveste kupce, lahko še vedno deluje, če tem kupcem daste močnejšo pot po nakupu. Težava je morda v nadaljevanju, ne v kanalu.
Zmanjšajte odvisnost od krhkih. Če se vaši zvesti kupci zgoščajo v lastnih kanalih — email, priporočila — je to znak, da te rastite in se manj naslanjate na plačane platforme, katerih stroški nenehno rastejo. Zmanjševanje odvisnosti vaše spletne trgovine od oglasov na Meti in Googlu to v celoti utemelji.
Nadaljevanje segmentirajte po kanalu, kjer se vedenje razlikuje. Kupci iz različnih virov morda potrebujejo drugačno pomiritev, preden bodo znova naročili. Temelj za to je v segmentiranje kupcev, ki bodo najverjetneje kupili znova.

Kaj avtomatizirati

Analiza kanalov je periodična. To, kar napaja, naj teče samo.

Sprožilec: prvo naročilo, označeno z virom pridobivanja.
Segment: prvi kupci, razvrščeni po kanalu, tako da je mogoče kupca s popustnega družbenega omrežja in kupca z organskega iskanja obravnavati različno.
Čas: zaporedje po nakupu, naravnano na naravni cikel ponovnega naročanja izdelka, uporabljeno za vsako kohorto kanala, da izboljšujete zadrževanje po vsej črti.
Vsebina: za kanale z nizko zvestobo dodatna vrednost in gradnja zaupanja pred kakršno koli ponudbo; za kanale z visoko namero bolj neposredna pot do naslednjega izdelka, saj ti kupci že želijo to, kar prodajate.
Cilj: dvigniti stopnjo drugega nakupa šibkejših kohort kanalov in jo zaščititi pri močnih.

Primer iz trgovine

Recimo, da prodajate nego kože. Dva kanala poganjata večino vaših prvih naročil. Plačano družbeno omrežje vodi kampanjo “40 % popusta na prvo naročilo” s CPA 12 EUR — poceni, velik obseg, priljubljenec financ. Organsko iskanje prinaša ljudi, ki vtipkavajo konkretne težave, pri CPA 22 EUR — skoraj dvakrat več.

Poročilo o prvem naročilu okrona plačano družbeno omrežje. Nato poženete kohortno analizo in ugotovite, da se kupci s popustnega družbenega omrežja vračajo pri 11 %, kupci z iskanja pa pri 29 %. Upoštevajte to in “dragi” kupec z iskanja, ki kupi dva- ali trikrat, je vreden več kot dvakrat toliko kot “poceni” kupec z družbenega omrežja, ki je pograbil popust in izginil. Ravno ste hoteli obrezati proračun za iskanje, da bi hranili družbeno omrežje. Podatki pravijo, naredite obratno — in kupcem z družbenega omrežja dajte pošten tok po nakupu, da rešite del tistih 89 %, ki se nikoli ne vrnejo. Številke so ponazoritvene; vzorec kanalov, vodenih s popustom, da slabo zadržujejo, je dovolj pogost, da ga pričakujete, še preden poženete svojega.

Kako to izmeriti

Stopnja drugega nakupa po kanalu — glavna, na enakem oknu na kohorto.
Stopnja ponovnega nakupa in življenjska vrednost po kanalu — stolpca, ki ju CPA ne vidi.
CPA in ROAS prvega naročila ob njih — še vedno koristna; le nikoli sama.
Trend skozi čas — kanali se premikajo, ko se spreminjajo stroški in občinstva, zato ga poganjajte znova vsako četrtletje.

Svoje številke kanalov postavite ob izhodišče kategorije, da veste, ali je “nizka” stopnja kanal ali le narava vašega izdelka — kako je videti zdrava stopnja prehoda iz prvega v drugi nakup vam da ta okvir.

Kje se vklopi Omnisend

Povezovanje kanala pridobivanja z vedenjem drugega nakupa potrebuje vir, shranjen na kupcu, in segmentacijo, ki lahko reže po njem. Omnisend uporabljam po vseh svojih trgovinah — izbran po tem, ko sem ga preizkusil proti Klaviyu — ker zajame vir prijave in naročila na profilu, tako da lahko zgradim segment, kot je “prvo naročilo je prišlo s plačanega družbenega omrežja”, in izmerim, kako ta skupina znova naroča v primerjavi z iskanjem ali emailom. Nato lahko vsako kohorto kanala usmerim v svoje lastno nadaljevanje, ne da bi za vsako znova gradil tok.

Orodje meri in avtomatizira nadaljevanje; ne bo prerazporedilo vašega proračuna za oglase in ne bo popravilo kanala, ki prinaša napačnega kupca — te odločitve so vaše, podprte s kohortnimi podatki. Omnisend je partner Shopimationa v pridruženem programu in priporočam ga iz vsakdanje uporabe; brezplačni paket zadošča, da kupce označite po viru in preizkusite segmentirana nadaljevanja.

Vaš naslednji korak

Izvlecite svoje prve kupce izpred treh do šestih mesecev, jih razvrstite po kanalu pridobivanja in izmerite stopnjo drugega nakupa vsakega — na istem časovnem oknu. Prvič, ko “poceni” kanal pokaže slabo stopnjo ponovnega nakupa ob “dragem”, ki dobro zadržuje, boste premislili, kam gre proračun. Nato obdelajte nadaljevanje: kupce iz šibkejših kanalov dajte v tok po nakupu, zgrajen tako, da jih pripelje nazaj, začnite pa s pretvarjanje prvega kupca v drugič vračajočega kupca.

First-to-Second Purchase

Why Some Acquisition Channels Produce More Second Purchases Than Others

Posted on July 18, 2026 by

Some acquisition channels produce more second purchases than others because they bring in customers with genuine intent and a real fit for what you sell, while others bring in one-time buyers chasing a deal or clicking an impulse ad. A customer who found you by searching for exactly what you offer behaves very differently, six months on, than one who impulse-bought from a discount-led social ad. Same first order value, completely different second-purchase rate. If you measure channels only on cost-per-acquisition and first-order ROAS, you’re half-blind — you’re optimizing for the cheapest first sale while ignoring which channels bring customers who stay. This article explains why the gap exists and how to find it in your own numbers.

The short answer: channels select for customer type

Every acquisition channel is a filter. It doesn’t just deliver traffic; it selects a kind of buyer, with a kind of intent, at a kind of moment.

Search catches someone actively looking for a solution — high intent, often a real need. A referral or word-of-mouth signup arrives pre-trusted, warmed up by a person they know. A well-built email list you’ve grown over time is full of people who chose to hear from you. A broad, discount-led social ad, by contrast, catches someone mid-scroll who wasn’t shopping at all and bit because the price was low. All of them can produce a first order. They do not produce the same customer.

That’s the mechanism. The second-purchase rate isn’t really a property of the channel — it’s a property of the customer the channel attracts, and different channels attract customers with very different odds of coming back.

The real problem: your best-looking channel might be your worst

Here’s where it costs money. Most stores judge channels on cost per acquisition and return on ad spend for that first order. By that scorecard, a cheap discount-led channel can look like your star — low CPA, quick first sale, healthy first-order ROAS.

Then you never check what those customers do next. If that “efficient” channel produces buyers who almost never return, its true value is a fraction of what the first-order math suggests. Meanwhile a channel with a higher upfront CPA — search, say, or a referral program — might bring customers who come back two or three times, making each one worth far more over their life. On a first-order scorecard it looks expensive. On a lifetime scorecard it’s your best source of customers.

Optimize on first-order economics alone and you’ll systematically shift budget toward channels that bring cheap, disloyal buyers and away from channels that bring loyal ones. The store looks efficient and the repeat rate quietly erodes. Why that repeat business matters more than the first sale is the whole argument of why the second purchase matters more than most store owners realize.

Why cost-per-acquisition can’t see this

CPA and first-order ROAS answer one question: how cheaply did this channel produce a first sale. They’re useful, and they’re incomplete. Neither one knows anything about what happens after the first order — whether the customer came back, how much they went on to spend, whether they’re worth acquiring at a higher price.

Two channels with an identical €15 CPA can have second-purchase rates of 12% and 30%. The first-order report shows them as equals. The lifetime reality shows one is worth roughly double the other. Judge on the incomplete metric and you’ll make confident, well-reasoned, wrong decisions about where to put your budget. The fix is to add a second column to how you evaluate channels — what the customer is worth over time, sitting right beside what the first sale cost you.

How to find which channels bring loyal customers

This is a cohort analysis by acquisition source. You can build the first version from your platform’s data.

Tag each customer with their acquisition channel at first order. Most ecommerce and email platforms capture the source — search, paid social, referral, email, direct, and so on. Make sure it’s stored on the customer, not just the order.
Group first-time buyers into cohorts by channel.
Give every cohort an equal return window. Only compare customers who’ve had the same time to come back — for example, “second purchase within 180 days of first order.” A channel you only started using last month can’t be fairly compared to one running all year until both cohorts have aged equally.
Measure each channel cohort on second-purchase rate, repeat rate over 6–12 months, and lifetime value — the metrics that CPA alone will never show you.
Rank channels by loyalty and lifetime value, then set that beside CPA. The channels that look best on both are where budget should concentrate. The cheap channels that retain nothing get a hard second look.

One caution: attribution is messy. A customer might touch several channels before buying, and “last click” over-credits whatever was closest to the purchase. You don’t need perfect attribution to act — you need consistent attribution, applied the same way to every cohort, so the comparison between channels is fair even if any single number is imperfect.

What to do with the answer

Reallocate toward loyalty, not just cheapness. Shift budget toward channels that bring customers who come back, even where the first-order CPA is higher, because lifetime value is what you’re actually buying.
Fix, don’t necessarily abandon, weak channels. A cheap channel that brings disloyal buyers can still work if you give those customers a stronger post-purchase path. The problem may be the follow-up, not the channel.
Reduce your dependence on the fragile ones. If your loyal customers cluster in owned channels — email, referrals — that’s a signal to grow those and lean less on paid platforms whose costs keep climbing. Reducing your ecommerce store’s dependence on Meta and Google ads makes that case in full.
Segment follow-up by channel where behavior differs. Customers from different sources may need different reassurance before they’ll reorder. The groundwork for that is in segmenting customers who are most likely to buy again.

What to automate

The channel analysis is periodic. What it feeds should run on its own.

Trigger: first order, tagged with acquisition source.
Segment: first-time buyers grouped by channel, so a discount-social buyer and an organic-search buyer can be treated differently.
Timing: a post-purchase sequence timed to the product’s natural reorder cycle, applied to every channel cohort so you’re improving retention across the board.
Content: for low-loyalty channels, extra value and trust-building before any offer; for high-intent channels, a more direct next-product path since those customers already want what you sell.
Goal: raise the second-purchase rate of the weaker channel cohorts and protect it on the strong ones.

A store example

Say you sell skincare. Two channels drive most of your first orders. Paid social runs a “40% off first order” campaign with a €12 CPA — cheap, high volume, the finance-friendly favorite. Organic search brings people typing in specific concerns, at a €22 CPA — nearly double the cost.

The first-order report crowns paid social. Then you run the cohort analysis and find the discount-social buyers come back at 11%, while the search buyers come back at 29%. Factor that in and the “expensive” search customer, buying two or three times, is worth more than double the “cheap” social customer who took the discount and vanished. You were about to cut the search budget to feed social. The data says do the opposite — and give the social buyers a proper post-purchase flow to rescue some of that 89% who never return. Numbers are illustrative; the pattern of discount-led channels under-retaining is common enough to expect it before you even run yours.

How to measure it

Second-purchase rate by channel — the headline, on an equal window per cohort.
Repeat rate and lifetime value by channel — the columns CPA can’t see.
CPA and first-order ROAS beside them — still useful; just never alone.
Trend over time — channels shift as costs and audiences change, so rerun it quarterly.

Set your channel numbers against a category baseline so you know whether a “low” rate is the channel or just your product’s nature — what a healthy first-to-second purchase rate looks like gives you that frame.

Where Omnisend fits

Tying acquisition channel to second-purchase behavior needs the source stored on the customer and segmentation that can slice by it. I use Omnisend across my stores — picked after testing it against Klaviyo — because it captures signup and order source on the profile, so I can build a segment like “first order came from paid social” and measure how that group reorders compared to search or email. Then I can route each channel cohort into its own follow-up without rebuilding the flow for each one.

The tool measures and automates the follow-up; it won’t reallocate your ad budget or fix a channel that brings the wrong customer — those calls are yours, informed by the cohort data. Omnisend is an affiliate partner of Shopimation, and I recommend it from daily use; the free tier is enough to tag customers by source and test the segmented follow-ups.

Your next step

Pull your first-time buyers from three to six months ago, group them by acquisition channel, and measure the second-purchase rate of each — on the same time window. The first time a “cheap” channel shows a poor repeat rate next to an “expensive” one that retains well, you’ll rethink where the budget goes. Then work the follow-up: put your weaker-channel customers into a post-purchase flow built to bring them back, starting with turning a first-time buyer into a second-time customer.