Recommendations from what you actually smoke

Three things happen, and none of them is a hidden model. The cigars you rated well get matched against the catalogue on their attributes — wrapper, origin, strength, body and brand. On the web, the flavour families you record in your tasting notes count too. And your ratings get compared against other members' ratings, so cigars that members with overlapping taste rated well get surfaced as well.

On the web, once you have rated a few cigars, the cigars you browse show a match score for you where there is something honest to say, with the reason written out beside it. That's the whole mechanism: attribute matching, on the web your flavour profile too, and a comparison against similarly-rated members. Nothing you can't reason about, and no part of it that is not built from something you did.

A top-ten list is the same for every visitor, and it describes the average smoker — who does not actually exist. If you consistently rate Connecticut shades poorly and full-bodied maduros highly, a popularity list has no way to know that. It just keeps recommending the same generic middle of the catalogue.

Your ratings are specific to you. The more of them there are, the further the recommendations move away from generic and toward what you've actually shown you like.

The engine looks at a defined set of inputs from your journal, not your entire account history.

Cigars you have already logged are taken out. A recommendation you have smoked is not a recommendation, whichever way you rated it.

So are one-off customs. A shop roll somebody logged is not something you could buy, so it is never suggested to you.

And on the similar-taste half, only other members' public entries count. A private entry never surfaces its author to anyone, directly or through a cigar it caused to be recommended.

On the web, signed in and once you have rated a few cigars, a cigar's page and your recommendation lists show how well it fits you as a percentage, with a short reason in plain words — the wrapper you rate well, the brand you keep coming back to, how many members with your taste rated it highly. Where a cigar is neither a clear fit nor a clear miss, the percentage is left off and the reason stands alone, because a score near the middle says nothing useful.

Where there is nothing honest to compare, nothing is shown, rather than a made-up number.

On the web, the catalogue's Browse menu has three. For You is built from the attributes and flavours you rate well, with a Stretch pick mixed in now and then. Similar Users is built from members whose ratings track with yours. Explore does the opposite of the other two on purpose — a wrapper you have never tried, a strength outside your usual — because a profile built only from what you already like narrows until nothing new gets in.

Until you have logged enough for For You to say anything specific, it shows what is Popular with members, labelled as exactly that. It switches to picks built from your own ratings as you log and rate.

Separately, every cigar's page lists the cigars closest to it, which works whether or not you have logged anything.

On the web, a Stretch pick is a cigar outside both your usual wrapper and your usual strength that members rated very highly — members with your taste where there are enough of them, members in general where there are not. It is marked as a Stretch pick and says why it was chosen.

On the web, as you log and rate, they start to appear in For You. With too little to go on there are simply none, and the list is made up of matches instead.

On the web, in your journal's Insights, Try next lists the cigars most worth smoking next, each with its reason. Your palate draws a radar of the flavour families you taste most, What changed shows where your taste has moved, and Discoveries are small facts read back from your own entries.

On the web, Members with your taste is the other side of it. As you log and rate, the Community's People tab can list members whose public ratings line up closely with yours, and a cigar's page can tell you how those members rated it. Only other members' public entries count towards that, and a match needs enough ratings on both sides, so it appears as you build a history rather than on day one.

The first few entries move it further than the hundredth will. A handful of ratings is already enough for the attribute half to have something to match on, and the picks visibly stop being generic.

From there it keeps sharpening. We would rather say that plainly than promise pinpoint accuracy on your first session, which would not be true of any system built on data it does not have yet.

Recommendations are a starting point, not the only way to find a cigar. Browsing, filtering and searching the catalogue directly is always available, whether or not you've logged a single entry.