Avtomatizacija zapuščenega ogleda

Kako prilagoditi emaile za zapuščen ogled glede na kategorijo

Posted on July 19, 2026 by

Prilagajanje emailov za zapuščen ogled glede na kategorijo pomeni krojenje sporočila glede na vrsto izdelka, ki ga je nekdo pogledal — ker je človek, ki brska tekaške copate, in človek, ki brska kavni aparat, v povsem drugačnem miselnem stanju, na drugačnih časovnicah, z drugačnimi vprašanji. Točno ogledani izdelek gre v vsak email samodejno. Kar doda prilagoditev po kategoriji, je vse okoli njega: podporna priporočila, pomiritev, ki šteje za to kategorijo, ton in celo časenje. Copati hočejo samozavest glede velikosti in sorodne sloge; kavni aparat hoče primerjavo specifikacij in opombo o garanciji. Pošiljanje obema kupcema istega splošnega emaila “tukaj je, kar ste si ogledali” pušča najlažjo relevantnost neizkoriščeno. Ta članek pokaže, kako segmentirati po kategoriji, ne da bi gradili petdeset ločenih tokov.

Zakaj en splošen email za ogled slabo deluje

Osnovni email za ogled — ogledani izdelek, povezava nazaj, topla vrstica — je v redu. A vsako kategorijo obravnava enako, kategorije pa se ne obnašajo enako.

Tisti, ki brska modo, pogosto primerja videze in ga skrbi glede prileganja in vračil. Tisti, ki brska tehniko, primerja specifikacije in ga skrbi, ali je to pravi model. Tistemu, ki brska potrošni material (kava, dodatki, nega kože), morda zadostuje majhen opomnik, ker je nakup nizko tvegan in ponovljiv. Ista mehanika toka, zelo drugačna nakupovalna psihologija. Splošen email ne nagovarja nobene od teh konkretno, zato pretvarja kot splošen email — kar pri hladnem občinstvu za ogledovanje ni veliko.

Kaj se dejansko spremeni po kategoriji

Prilagoditev po kategoriji ne pomeni pisanja romana za vsako izdelčno linijo. Gre za zamenjavo nekaj elementov z velikim vzvodom:

Podporna priporočila. Oblačila: drugi izdelki v istem slogu ali dopolnilni kosi. Elektronika: isti izdelek v drugih konfiguracijah ali dodatki. Pohištvo: usklajeni kosi iz zbirke. Točno ogledani izdelek ostane junak; podpora se zaveda kategorije. (Kako uporabiti priporočila izdelkov v emailih za zapuščen ogled.)
Pomiritev, ki šteje. Moda → brezplačna vračila in pomoč pri prileganju. Visoka cena → garancija, dostava, financiranje. Potrošni material → naročnina ali enostavno ponovno naročilo. Uskladite namig zaupanja z resničnim ugovorom za to kategorijo.
Ton in nuja. Nizkocenovna impulzivna kategorija je lahko kanček bolj igriva in hitra; premišljena kategorija naj bo mirnejša in bolj v pomoč.
Časenje. Impulzivne kategorije hočejo krajšo zakasnitev; premišljene hočejo daljšo. Kategorija je naravno mesto za razvejitev koraka čakanja. (Kako hitro naj pošljete email za zapuščen ogled.)

Praktičen način segmentiranja — začnite z nekaj vedri

Ne gradite toka na SKU. Svoj katalog združite v peščico veder, temelječih na vedenju, in prilagodite tem. Koristna začetna razdelitev za mnoge trgovine:

Impulzivno / nizka premislek — poceni izdelki s hitro odločitvijo. Kratka zakasnitev, igriv ton, sorodni priljubljeni izdelki.
Premišljeno / srednja cena — glavnina normalnega kataloga. Standardna zakasnitev, pomiritev glede prileganja/specifikacij, sorodni izdelki.
Visoka cena / raziskano — dragi izdelki, počasne odločitve. Daljša zakasnitev, osredotočeno na garancijo in pomoč, primerjalna vsebina. (Ustvarjanje zapuščenega ogleda za dražje izdelke.)

Tri vedra pokrijejo večino trgovin in so vzdržljiva. Znotraj vsakega vedra lahko dodate besedilo pomiritve, specifično za kategorijo, ne da bi množili tokove.

Kako ljudi usmeriti v pravo vedro

Kategorija ogledanega izdelka določi razvejitev. V praksi izdelke označite (ali uporabite obstoječo strukturo kategorij) in pustite, da tok prebere kategorijo ogledanega izdelka, nato pošlje ustrezno različico. Pri velikem ali neurejenem katalogu se tu izplača nekaj urejanja vnaprej — čiste kategorije naredijo prilagoditev po kategoriji skoraj samodejno. Če je vaš katalog ogromen, glejte Kako ustvariti zapuščen ogled za velike kataloge izdelkov za obvladovanje obsega.

Konkreten primer

Trgovina prodaja oblačila, obutev in nekaj tehničnih dodatkov. (Ilustrativno.)

Nekdo brska volnen plašč → različica za obutev-in-oblačila: plašč, vrstica o brezplačnih 30-dnevnih vračilih in enostavnih menjavah, dva dopolnilna kosa (šal, pulover), topel ton, 3-urna zakasnitev.
Nekdo brska par tekaških copat za teren → isto vedro za oblačila, a osredotočeno na prileganje: copat, povezava do vodnika po velikostih, dva sorodna sloga, pomiritev “ustreza velikosti in vračila so brezplačna”.
Nekdo brska slušalke za 220 € → različica za visoko ceno: slušalke, opomba o 2-letni garanciji, kratka primerjava z naslednjim boljšim modelom, daljša 8-urna zakasnitev, mirnejši ton.

Trije različni emaili, ena struktura toka, vodena s kategorijo ogledanega izdelka. Vsak deluje, kot da je napisan za tega kupca, ker podporna vsebina dejansko je.

Kje pušča brez tega

Prihodek, ki ga izgubite splošnim emailom za ogled, ni dramatičen na sporočilo — je stalno slabo delovanje relevantnosti, ki ste jo že imeli na voljo iz podatkov. Veste, katero kategorijo so pogledali; pošiljanje vsebine, ki to prezre, pomeni neizkoriščen enostaven dvig. Ilustrativno: če email, ki se zaveda kategorije, dvigne razmerje klikov in pretvorbo celo za skromno mero nad splošnim, se to čez celoten obseg vašega ogledovanja sešteje v pravi denar za spremembo, ki je večinoma konfiguracija, ne nova kreativa. (Številke ilustrativne.)

Kako meriti, ali deluje

Razmerje klikov in pretvorba po vedru kategorije — vsako vedro primerjajte s splošno izhodiščno vrednostjo, ki jo je zamenjalo. Nekatere kategorije se bodo odzvale veliko bolj kot druge; to vam pove, kam vlagati.
Prihodek na prejemnika na vedro, proti kontrolni skupini.
Katera podporna priporočila se klikajo — povratna informacija o tem, ali je vaša logika kategorij dejansko relevantna.

Bodite pozorni na vedro, ki deluje slabše od splošne različice — to običajno pomeni, da je združevanje kategorij ali pomiritev napačno za tisto vrsto izdelka, ne da je prilagoditev spodletela.

Kako pomaga Omnisend

Prilagoditev po kategoriji potrebuje, da tok prebere lastnosti ogledanega izdelka in se po njih razveji, plus bloke priporočil, ki spoštujejo kategorijo. V svojih trgovinah uporabljam Omnisend po tem, ko sem ga preizkusil proti Klaviyu, deloma zato, ker je razvejanje toka po kategoriji izdelka in vstavljanje blokov priporočil, ki se zavedajo kategorije, preprosta samopostrežba in ne gradbeni projekt. Pomembni deli: sprožilci za ogled, ki nosijo lastnosti izdelka, pogojne razdelitve po kategoriji, dinamični bloki priporočil in poročanje po segmentih. (Omnisend je orodje, ki ga uporabljam in priporočam; morebitna partnerska povezava je razkrita — pristop deluje na katerikoli zmogljivi platformi.)

Vaš naslednji korak

Svoj katalog združite v tri vedra, temelječa na vedenju — impulzivno, premišljeno, visoka cena — in napišite eno različico emaila za ogled na vedro, ki se razlikuje po pomiritvi, podpornih izdelkih in zakasnitvi. Usmerjajte po kategoriji ogledanega izdelka, nato primerjajte pretvorbo vsakega vedra s svojim starim splošnim emailom. Poglobite stran namere z Kako segmentirati zapuščen ogled po nameri stranke in izostrite priporočila z Kako uporabiti priporočila izdelkov v emailih za zapuščen ogled.

Browse Abandonment Automation

Personalizing Browse Abandonment Emails by Category

Posted on July 10, 2026 by

Personalizing browse abandonment emails by category means splitting your flow so that someone who browsed running shoes gets a different email than someone who browsed winter jackets — different subject line, different supporting content, different recommended products. You do it by adding a category condition after the trigger, then writing one email variant per major category. For most stores, three to six category branches cover 80% of browse traffic, and each branch outperforms a generic “you looked at something” email because the message finally matches what the person was actually shopping for.

The generic reminder problem

Most stores that run browse abandonment run exactly one version of it. The email says “Still interested?”, shows the viewed product, and links back. That’s fine as a first build — it’s what I’d tell anyone to start with. But once the flow has been live for a couple of months, you’ll notice something in the reports: certain categories convert well from it and others barely move.

That gap exists because a single template forces one tone and one argument onto very different purchases. A €25 phone case and a €280 coffee machine don’t stall for the same reason. The phone case browser got distracted; the coffee machine browser needs reviews, a comparison, maybe a delivery-time answer. When the email treats both identically, it’s mediocre for both. Meanwhile you’re paying rising ad prices to bring both of those people to the site in the first place, and the follow-up — the part that costs you nothing per send — is the part left generic.

Why the answer isn’t more sends or a coupon

The reflex fix for a flat flow is volume: add a third email, shorten the delay, bolt on a discount. Volume doesn’t fix relevance. If the message didn’t land once, sending it twice mostly grows your unsubscribe count, and a coupon papers over the mismatch at the cost of your margin. Personalization attacks the actual problem: the email one person receives should read like it was written for the thing they were shopping for, because it was.

Here’s the leak in concrete terms, with invented, illustrative numbers. Suppose your flow sends 1,500 emails a month at €0.55 revenue per recipient — €825. Category-matched emails don’t need to work miracles; if better relevance lifts revenue per recipient to €0.75 across the same sends, that’s €1,125 a month, a €300 difference from copy changes alone. No extra traffic, no discount budget. The lift comes from opens (a subject line naming the category beats a vague one) and from clicks (recommendations from the right category get tapped; random ones don’t).

How to split the flow, step by step

Pull your browse data by category first. Look at the last 60 to 90 days of product views from identified visitors. Rank categories by view volume. You want the short list of categories that generate most flow entries — typically three to six.
Create one branch per top category, plus a fallback. After the browse trigger, add a condition: viewed product’s category is X. Everything that matches no branch drops into your existing generic email, which stays as the safety net.
Rewrite the subject line and opening line per branch. “Your running shoe shortlist is still here” beats “Still browsing?” every time the reader actually was looking at running shoes.
Swap the supporting content per branch. The high-consideration category gets reviews and a buying guide; the impulse category gets speed (“in stock, ships today”); the seasonal category gets urgency that’s actually true.
Later, not now: per-category timing tweaks, category-specific second emails, SMS branches. Prove the first split before multiplying it.

If your catalog has hundreds of categories rather than six, branching by hand stops scaling — the large-catalog approach leans on dynamic content instead of branches.

The full spec for one branch

Take “coffee equipment” as the worked example:

Trigger: identified contact viewed a product, no add-to-cart, no purchase within 2 hours.
Branch condition: viewed product category equals “Coffee equipment”.
Segment filters: exclude buyers from the last 14 days and anyone already in the cart abandonment flow. Recent buyers browsing the same category are often just admiring what they bought — excluding them saves you awkward sends.
Delay: 4 hours. Channel: email.
Message content: subject “Your espresso setup, continued”; the viewed product with image and price; one line on what distinguishes it (“19-bar pump, 2-year warranty”); a block of two or three other coffee-category products; one customer review from that category.
Goal: click back to the viewed product or a sibling product in the same category.

Note the recommendation block is category-locked. Showing a coffee browser your bestselling yoga mat is how generic flows leak clicks; recommendations deserve their own set of rules.

A quick check that this is worth your time

Before building six branches, run this five-minute test. Take last month’s browse abandonment sends and split the results by the category of the viewed product (most platforms let you filter flow analytics by product or you can export and pivot). If your top two categories already convert at, say, double the rate of the rest — those invented numbers again, purely illustrative — the generic email is being carried by categories where it accidentally fits. Branch the weak high-volume categories first; that’s where the recoverable money sits.

Metrics per branch, not per flow

Revenue per recipient, by branch — the number that justifies or kills each variant.
Open rate by branch — tests whether the category-specific subject line earns attention.
Click-to-open rate by branch — tests whether the content matches the promise.
Share of flow entries hitting the fallback — if 60% of entries still land in the generic email, your branches cover the wrong categories.

Doing this in Omnisend

For years I dismissed email marketing as the channel where bad newsletters go to die. What changed my mind was watching acquisition costs eat my margin and realizing behavioral email is a different animal — it responds to what a specific person just did. I compared Klaviyo and Omnisend for my own stores and stayed with Omnisend, mainly because building conditional branches like these is something I can do myself in an afternoon.

Concretely: in the flow editor you add a split after the browse abandonment trigger using the viewed product’s category as the condition, then attach a separate email to each path. The product recommender block can be filtered to the matching category so each branch pulls its own products automatically. Two honest limits: this only works if your product categories are clean in your store’s catalog feed — miscategorized products end up in the wrong branch — and branching multiplies the emails you maintain, so every seasonal update now touches four templates instead of one.

One thing to do this week

Open your browse abandonment report and split its results by product category. Find the highest-traffic category with the weakest revenue per recipient, and build one branch for it — one subject line, one email, one category-locked recommendation block. Compare it against the generic email after two weeks.