Personalizacija in priporočila izdelkov

Kako personalizirati emaile za nove obiskovalce

Posted on July 18, 2026 by

Emaile lahko personalizirate tudi novim obiskovalcem, čeprav o njih veste skoraj nič — kajti “skoraj nič” ni nič. Prvi obiskovalec, ki se pridruži vašemu seznamu, vam da tri uporabne signale: kje se je zgodila prijava (katera stran, kateri pojavni obrazec), kaj je brskal pred prijavo in po njej ter kako je prišel (oglas za določen izdelek pove več kot obisk domače strani). Personalizacija na tej stopnji pomeni, da s temi tremi signali oblikujete pozdravni tok, namesto da vsakemu novemu naročniku pošljete iste tri splošne emaile. Ta članek govori prav o tem oknu prvega stika. Ko nekdo enkrat naroči, se igra popolnoma spremeni — to je obravnavano v kako personalizirati emaile za obiskovalce, ki se vračajo.

Položaj: vaši najdražji naročniki dobijo vaše najbolj splošne emaile

Pomislite, koliko vas je nov naročnik dejansko stal. Prišel je prek oglasa, ki ste ga plačali, brskal po straneh, ki ste jih plačali zgraditi, in vam predal email naslov — pogosto v zameno za pozdravni popust, ki bo pojedel maržo pri njegovem prvem naročilu. To je najdražji stik na vašem seznamu.

In potem mu večina trgovin pošlje pozdravno zaporedje, napisano za nikogar posebej. “Dobrodošli v družini. Tukaj je naša zgodba. Tukaj je 10 % popusta.” Isti trije emaili, ne glede na to, ali je oseba deset minut primerjala pohodne čevlje ali odskočila od bloga s seznamom daril. Medtem trgovec vso energijo za personalizacijo vloži v tokove za obstoječe stranke — v ljudi, ki trgovino že poznajo in jih je treba najmanj prepričevati.

Okno je pomembno, ker je kratko. Radovednost novega naročnika je največja v prvih dneh po prijavi. Če v tem obdobju pošiljate splošno mašilo, drugega prvega vtisa ni.

Zakaj “pošljimo pač večji popust” ni odgovor

Običajen odziv na šibko konverzijo pozdravnega toka je posladkati ponudbo — 10 % postane 15 %, nato se na vrh naloži še brezplačna dostava. Včasih to dvigne konverzijo. Vedno pa reže maržo in nič ne naredi glede prave težave: email novemu naročniku še vedno prikazuje izdelke, ki jih ne zanimajo.

Večji popust na napačnem izdelku izgubi proti manjšemu popustu na pravem. Relevantnost je cenejši vzvod in pri prvih obiskovalcih je tisti, ki ga nihče ne potegne, ker “o njih še nimamo podatkov”. Imate jih več, kot mislite.

Trije signali, ki jih že imate

Kontekst prijave. Pojavni obrazec na strani izdelka vam pove, kateri izdelek je bil dovolj zanimiv, da je bil na zaslonu ob prijavi. Prijava iz vašega vodnika “za začetnike” pove raven izkušenosti. Prijava v nogi domače strani pove malo — kar je samo po sebi uporabno, ker vam pove, da uporabite širši prvi email. Če vaši obrazci ta kontekst trenutno zavržejo, je to prvi popravek: večina orodij za obrazce zna naročnika označiti s stranjo ali zbirko, kjer se je prijava zgodila.

Brskanje pred prijavo in po njej. Ko je nekdo prepoznan (prijavil se je, torej je), se njegovo vedenje v seji priloži k profilu. Dva ogleda izdelkov v eni kategoriji sta šibek signal; šest ogledov in obisk vodnika za velikosti sta močan. Kaj s tem signalom početi, ima svoj celoten vodnik — kako uporabiti vedenje pri brskanju za boljša priporočila izdelkov — a za pozdravni tok je kratka različica: naj brskana kategorija usmerja, katere izdelke prikazujejo emaili.

Vir prometa. Naročnik, ki je kliknil na Meta oglas za vaše tekaške jakne, ni skrivnost. Prenesite kontekst kampanje v tok: pozdravni email lahko začne z linijo jaken, namesto z obhodom po vsej trgovini.

En iskren opomin, preden karkoli od tega uporabite: med relevantnim in vznemirljivim je meja. “Ker ste si ogledovali nepremočljive jakne”, izpisano dobesedno, lahko na nekoga, ki vam je email dal pred štiridesetimi minutami, deluje kot nadzorovanje. Pokažite relevantnost; ne pripovedujte je. Meje je vredno dobro razumeti — glejte kako personalizirati emaile spletne trgovine, ne da bi bili nadležni.

Gradnja personaliziranega pozdravnega toka

Ohranite strukturo toka, ki jo imate — ponavadi dva do štiri emaile — in vsebino naredite pogojno.

Email 1, takoj po prijavi. Hitro dostavite tisto, kar je obrazec obljubil. Nato en dinamični blok: če obstaja brskana kategorija ali kategorija iz konteksta prijave, prikažite izdelke iz nje; če ne, prikažite najbolje prodajane izdelke po vsej trgovini. Najbolje prodajani so pravilna nadomestna rešitev za prazen profil — kompromisi med obema pristopoma so predstavljeni v najbolje prodajana proti personaliziranim priporočilom: kaj bolje konvertira.

Email 2, dan ali dva pozneje. Do zdaj profil pogosto ni več prazen — kliknili so nekaj v emailu 1 ali spet brskali. Uporabite to. Blok z nedavno ogledanimi izdelki je tukaj naravno središče, temu pa je namenjen tudi poseben vodnik v uporaba nedavno ogledanih izdelkov v emailu. Ovijte ga v nekaj, kar gradi zaupanje: mnenja, vaše jamstvo, po čem se trgovina razlikuje.

Email 3, spet dva ali tri dni pozneje. Če popust ni bil uporabljen, je to opomnik — usmerjen na konkretne izdelke, za katere so pokazali zanimanje, ne na katalog na splošno. Koda, ki poteče ob sliki točno tistih čevljev, ki so si jih trikrat ogledali, je povsem drug email kot koda, ki poteče ob logotipu.

Kaj pa, če nekdo res nima nobenih vedenjskih podatkov — prijavil se je iz noge, ni brskal ničesar? Potem se ne pretvarjajte. Čist, dobro napisan splošni pozdrav premaga gledališče personalizacije. Ena uporabna poteza za te prazne profile: eno preprosto vprašanje z majhnim trenjem v emailu 1 ali 2 (“nakupujete za cesto ali za pot?”), pri katerem vsak odgovor označi profil. Dve možnosti, en klik. Tega vprašanja ne dajajte v sam prijavni obrazec — vsako dodatno polje tam vas stane prijav, klik v emailu pa je zastonj.

Kaj natančno avtomatizirati

Sprožilec: stik se naroči (oddaja obrazca ali privolitev na blagajni — razmislite o ločevanju teh dveh, saj naročnik z blagajne že kupuje in ne bi smel dobiti zaporedja “vrni se in kupi”).
Segment/pogoji: razdelite glede na to, ali obstaja brskana kategorija ali kategorija iz konteksta prijave. Ena veja dobi vsebino, vodeno po kategoriji, druga dobi različico z najbolje prodajanimi. Dve veji sta za začetek dovolj; pet vej že prvi dan je način, kako pozdravni tokovi ostanejo napol zgrajeni.
Časovnica: email 1 v nekaj minutah; email 2 na dan 1–2; email 3 na dan 4–5. Ob nakupu izstop iz toka.
Kanal: email nosi to stopnjo. SMS le, če so se izrecno naročili ob prijavi, in zmerno — zaupanje prvega obiskovalca je krhko.
Vsebina: dinamični bloki izdelkov, napolnjeni z brskanjem/kontekstom; statične nadomestne rešitve za prazne profile.
Cilj: prvi nakup, sledeno po veji, da vidite, ali si personalizirana veja dejansko zasluži svojo zapletenost.

Primer iz trgovine

Trgovina z nego kože ima en pojavni obrazec za vso trgovino. Obiskovalec prebere dve strani o retinolu, dobi pojavni obrazec in se naroči. Različica A pozdravnega toka — splošna — pošlje zgodbo znamke in mrežo osmih najbolje prodajanih izdelkov, večinoma čistil in krem za sončenje. Različica B označi prijavo s kontekstom retinola: email 1 prikaže dva retinolna izdelka in opombo o začetniški jakosti, email 2 odgovori na tri vprašanja, ki jih ima o retinolu vsak, in prikaže začetni komplet, email 3 opomni na popust ob istem kompletu. Isti popust, isto število emailov, isti izdelki v katalogu. En tok govori z osebo; drugi govori mimo nje. (Ponazoritveni primer.)

Kako to izmeriti

Štiri številke vam povedo, ali se personalizacija ob prvem obisku splača:

Stopnja konverzije pozdravnega toka — naročniki, ki kupijo znotraj okna toka, razdeljeni po vejah.
Stopnja klikov na dinamičnih blokih v primerjavi s statičnimi nadomestnimi bloki.
Prihodek na prejemnika pozdravnega toka, najbolj čist posamezni kazalnik.
Stopnja odjav znotraj toka — če se “personalizirana” veja odjavlja pogosteje, ste zašli v nadležnost ali pa napačno označujete interese.

Vsaki spremembi dajte štiri do šest tednov podatkov, preden jo ocenite. Pozdravni tokovi imajo v večini trgovin majhen dnevni obseg; sklepi tretjega dne so šum.

Kako to nastaviti v Omnisendu

Ta tok je mogoče zgraditi v vsaki resni email platformi za spletne trgovine. V lastnih trgovinah uporabljam Omnisend — preizkusil sem Klaviyo in Omnisend drug ob drugem in ostal pri Omnisendu zaradi enostavnejšega vsakodnevnega upravljanja in boljše cene — in koščki se neposredno ujamejo: prijavni obrazci lahko naročnike označijo po obrazcu in strani, pozdravni tok podpira pogojne razcepe glede na te oznake in glede na dejavnost brskanja, bloki s priporočili izdelkov pa brez dela po meri obvladajo logiko brskane kategorije z nadomestno rešitvijo najbolje prodajanih. Za obstoj signalov o brskanju mora biti nameščeno spletno sledenje, zato to preverite, preden gradite veje, ki so od njega odvisne.

Česa orodje ne bo naredilo: odločilo, kaj naj pove prvi vtis vaše trgovine. Pogojna napeljava vzame uro; besede v emailu 1 so še vedno težji del. Omnisend je partner Shopimationa v affiliate programu — priporočam ga, ker ga vsak dan uporabljam, ne zaradi povezave.

Vaš naslednji korak

Danes preverite eno stvar: ko se nekdo prijavi s strani izdelka, ali vaš sistem zabeleži, kje se je prijava zgodila? Če ne, to popravite najprej — vsaka druga taktika tukaj je od tega odvisna. Nato pozdravni tok razdelite na dve veji, vodeno po kategoriji in nadomestno z najbolje prodajanimi, in pustite, da vam mesec podatkov pove, kaj naj druga veja postane.

Personalization and Product Recommendations

How to Personalize Emails for First-Time Visitors

Posted on July 17, 2026 by

You can personalize emails to first-time visitors even though you know almost nothing about them — because “almost nothing” isn’t nothing. A first-time visitor who joins your list gives you three usable signals: where the signup happened (which page, which popup), what they browsed before and after subscribing, and how they arrived (an ad for a specific product says more than a homepage visit). Personalization at this stage means using those three signals to shape the welcome flow, instead of sending the same three generic emails to every new subscriber. This article is about that first-contact window specifically. Once someone has ordered, the game changes completely — that’s covered in how to personalize emails for returning customers.

The situation: your most expensive subscribers get your most generic emails

Think about what a new subscriber actually cost you. They arrived from an ad you paid for, browsed pages you paid to build, and handed over an email address — often in exchange for a welcome discount that will eat margin on their first order. This is the most expensive contact on your list.

And then most stores send them a welcome sequence written for nobody in particular. “Welcome to the family. Here’s our story. Here’s 10% off.” The same three emails whether the person spent ten minutes comparing hiking boots or bounced off a gift-guide blog post. Meanwhile the merchant’s personalization energy goes into flows for existing customers — the people who already know the store and need the least convincing.

The window matters because it’s short. A new subscriber’s curiosity peaks in the first days after signup. Send generic filler during that stretch and there is no second first impression.

Why “just send a stronger discount” isn’t the answer

The usual response to weak welcome-flow conversion is to sweeten the offer — 10% becomes 15%, then free shipping gets stacked on top. Sometimes that lifts conversion. It always cuts margin, and it does nothing about the real problem: the email still shows the new subscriber products they don’t care about.

A bigger discount on the wrong product loses to a smaller discount on the right one. Relevance is the cheaper lever, and with first-time visitors it’s the one nobody pulls because “we don’t have any data on them yet.” You have more than you think.

The three signals you already have

Signup context. A popup on a product page tells you which product was interesting enough to be on screen when they subscribed. A signup from your “for beginners” guide tells you experience level. A footer signup on the homepage tells you little — which is itself useful, because it tells you to use a broader first email. If your forms currently throw this context away, that’s the first fix: most form tools can tag a subscriber with the page or collection where the signup happened.

Pre- and post-signup browsing. Once someone is identified (they subscribed, so they are), their session behavior attaches to their profile. Two product views in one category is a weak signal; six views and a size-guide visit is a strong one. What to do with this signal has its own full guide — how to use browsing behavior for better product recommendations — but for the welcome flow the short version is: let the browsed category steer which products the emails show.

Traffic source. A subscriber who clicked a Meta ad for your running jackets is not a mystery. Carry the campaign context into the flow: the welcome email can lead with the jacket line rather than a tour of the whole store.

One honest warning before using any of this: there’s a line between relevant and unsettling. “Since you were looking at rain jackets” written out explicitly can read as surveillance to someone who gave you their email forty minutes ago. Show relevance; don’t narrate it. The boundaries are worth understanding properly — see how to personalize ecommerce emails without being creepy.

Building the personalized welcome flow

Keep the flow structure you have — usually two to four emails — and make the content conditional.

Email 1, immediately after signup. Deliver whatever the form promised, fast. Then one dynamic block: if a browsed or signup-context category exists, show products from it; if not, show store-wide bestsellers. Bestsellers are the correct fallback for a blank profile — the trade-offs between the two approaches are laid out in best-selling vs personalized recommendations: which converts better.

Email 2, a day or two later. By now the profile is often less blank — they clicked something in email 1, or browsed again. Use it. A recently-viewed-products block is the natural centerpiece here, and there’s a dedicated walkthrough in using recently viewed products in email. Wrap it in something that builds trust: reviews, your guarantee, what makes the store different.

Email 3, another two or three days on. If the discount hasn’t been used, this is the reminder — pointed at the specific products they’ve shown interest in, not the catalog at large. A code expiring alongside a picture of the exact boots they viewed three times is a different email from a code expiring next to a logo.

What if someone genuinely has no behavioral data — signed up from the footer, browsed nothing? Then don’t fake it. A clean, well-written generic welcome beats personalization theater. One useful move for these blanks: a single low-friction question in email 1 or 2 (“shopping for road or trail?”) with each answer tagging the profile. Two options, one click. Don’t put that question in the signup form itself — every extra field there costs signups, and the email click is free.

What to automate, precisely

Trigger: contact subscribes (form submission or checkout opt-in — consider separating these, since a checkout subscriber is already buying and shouldn’t get the come-back-and-buy sequence).
Segment/conditions: split on whether a browsed or signup-context category exists. One branch gets category-led content, the other gets the bestseller version. Two branches is plenty to start; five branches on day one is how welcome flows end up half-built.
Timing: email 1 within minutes; email 2 at day 1–2; email 3 at day 4–5. Exit the flow on purchase.
Channel: email carries this stage. SMS only if they explicitly opted in at signup, and sparingly — a first-time visitor’s trust is thin.
Content: dynamic product blocks fed by browsing/context; static fallbacks for empty profiles.
Goal: first purchase, tracked per branch so you can see whether the personalized branch actually earns its complexity.

A store example

A skincare store runs one popup shop-wide. A visitor reads two pages about retinol, gets the popup, subscribes. Version A of the welcome flow — the generic one — sends the brand story and a grid of eight bestsellers, mostly cleansers and sunscreen. Version B tags the signup with the retinol context: email 1 shows the two retinol products and a beginner’s-strength note, email 2 answers the three questions everyone has about retinol and shows the starter kit, email 3 reminds about the discount next to that same kit. Same discount, same number of emails, same products in the catalog. One flow talks to a person; the other talks past her. (Illustrative example.)

How to measure it

Four numbers tell you whether first-visit personalization is paying:

Welcome flow conversion rate — subscribers who purchase within the flow window, split by branch.
Click rate on the dynamic blocks versus the static fallback blocks.
Revenue per welcome-flow recipient, the cleanest single scoreboard.
Unsubscribe rate inside the flow — if the “personalized” branch unsubscribes more, you’ve crossed into creepy or you’re mis-tagging interests.

Give any change four to six weeks of data before judging it. Welcome flows have small daily volume in most stores; day-three conclusions are noise.

Setting this up in Omnisend

This flow is buildable in any serious ecommerce email platform. I use Omnisend in my own stores — I tested Klaviyo and Omnisend head-to-head and stayed with Omnisend for the simpler day-to-day management and better price — and the pieces map directly: signup forms can tag subscribers by form and page, the welcome flow supports conditional splits on those tags and on browsing activity, and product recommendation blocks handle the browsed-category-with-bestseller-fallback logic without custom work. Web tracking has to be installed for the browsing signals to exist, so check that before building branches that depend on it.

What the tool won’t do: decide what your store’s first impression should say. The conditional plumbing takes an hour; the words in email 1 are still the hard part. Omnisend is an affiliate partner of Shopimation — I recommend it because I run it daily, not because of the link.

Your next step

Check one thing today: when someone subscribes from a product page, does your system record where the signup happened? If not, fix that first — every other tactic here depends on it. Then split your welcome flow into two branches, category-led and bestseller-fallback, and let a month of data tell you what the second branch should become.