Personalized Cigar Recommendations – Discovery

Personalized Cigar Recommendations
Published October 13, 2018 11:01 am PT

We continue our series of articles about personalized cigar recommendations and what we consider to be the key success factors for a personalized service. We also share with you some of the results of our latest customer satisfaction survey…

Science vs. Pragmatism

You may already know that we operate as a lean-agile team. In alignment with our continuous improvement remit, we recently conducted a non-scientific survey among our members to understand where we’re doing well and where we can further improve our service. We selected the areas that we consider to be key success factors for our personalized service:

  • Accuracy
  • Discovery
  • Transparency
  • Member Control
  • Sales Motive
  • Personalization
  • Cigar Knowledge
  • Member Knowledge
  • Privacy

We thank our members for the terrific feedback we received on all these fronts.

Today, we’ll focus on the discovery aspect. Said simply, discovery means providing recommendations that help the consumer discover valuable cigars she/he would not find otherwise. According to the study “Recommendations as personalized marketing”, “The customer will be satisfied if recommendations are related to his/her interest, offering variety in recommendations or even challenging him/her, in order to find great things to enjoy. The customer would be dissatisfied if recommendations are repetitive – ‘more of the same things’, ‘over and over again’ – thus failing to help him/her find great things to enjoy.”

Discovery to us is finding something new that can shake our world and make us look at things in a different way. We want to bring a new approach to the table, so we can challenge what we thought we liked and get recommendations that take us (David and I use Cigar Sense too) out of our comfort zone and away from the ordinary go-to choices.

Steve Jobs said “a lot of times, people don’t know what they want until you show it to them.” I firmly believe he was one of the few people entitled to make such a statement, he proved it through products we would have dreamed of. But such serendipitous discovery idea has turned into multi-billion-dollar races for the world’s biggest digital companies and most don’t get it right.

What drives “discovery” in personalized recommendations?

We once talked to someone who tried our free service and wanted to provide feedback. According to him, the recommended cigars should have all been Dominican, because he believes he only likes Dominican cigars. We appreciated the feedback, but had to conclude that this person was not open to the discovery of new cigars: he was trying to challenge the service without knowing the principles that drive it, and he was applying his biases.

Another Cigar Sense user was very surprised that we do not recommend new cigars based on them having the same wrapper as the cigars he likes. There would be a lot to say about this very topic but, to make it short, let’s say that a blend is much more than the sum of its leaves. There are amazing seminars organized by master blenders, and intentionally designed to highlight the influence of the wrapper component on the overall flavor. It’s an interesting theoretical exercise, and it educates on individual leaves, but in the reality of the cigars on the shelves it is rare to find the same organoleptic characteristics in a cigar based solely on the wrapper.  On the contrary, our expert panel analyses show that cigars using similar wrappers can present an amazing diversity of aroma and taste.

As Haruki Murakami said, “If you only read the books that everyone else is reading, you can only think what everyone else is thinking.”

I’d love to choose if I want to discover new products

If I know I like a particular product and a particular brand, I don’t waste my time on services that help me discover other products and brands, unless I decide I am ready to let go of my conviction that only that product and brand will make me happy.

However, it is unavoidable, on many websites, even if I am only going to buy an item I have already identified, I get distracted by many suggestions. On Amazon, for instance, before I even get to finish reading about the authors of a book I am about to buy, I need to scroll down through:

  • Frequently bought together
  • Customers who bought this item also bought
  • Sponsored products (twice)

Like Amazon, many recommendation engines make wild leaps of inference.  In my case, simply because I had been shopping for small backpacks, I was presented a recommended book on growing mushrooms.  Do you see the connection between a backpack and a book on mushrooms? True enough I was conducting my backpack search at the outset of truffle season in Europe, but still this seems a huge leap to me.  And sure enough, when I discovered the very last link on the page allowed me to edit my browsing history, and remove that ugly backpack that I never bought, the mushroom book disappeared.  At least there is an attempt to provide some transparency on the control Amazon allows you to have over what is recommended to you. Both transparency and user control are topics we will cover in separate articles.

The right data right

Recommendations are a service to the customer in the first place, rather than a technology, and I am not qualified to write about technologies. What I know is that you need to get the right data right (the repetition of right is deliberate).

If I don’t do that, in order to find a product similar to X, I can seriously limit the discovery for my customer. Recommendations would be repetitive and would not help the customer discover valuable items. A typical cigar description is mainly made of biographical information, such as country of origin, blend, size, price. Taking only this data into consideration would mean satisfying only those customers who believe that, for example, they only like other cigars produced in the same country or with the same wrapper. It’s easy to make a simple search feature based on mainstream data available, but does this make the customers happy in their discovery? The truth, just like the judgement of quality or crap, belongs to the beholder.

One of our principles at Cigar Sense is helping members be aware of what they like, and why they like it. If you take this information into account, it’s often a mix of cigars coming from different countries and made with different blends that they are likely to appreciate in their discovery journey.

In order to do this, we use meticulous descriptions of the sensory characteristics of each cigar and we match them to every recommendable cigar in our database. The results are ranked based on how likely it is the customer will enjoy each cigar. We call this personal fit.

How do we get the personal fit of each customer? Not based on where members browsed on our website, but based on their actual individual priorities and preferences, and it is confirmed by the feedback the member provides on each cigar that she/he tried based on our recommendations. The more feedback we get, the more we can be spot on.

How do our members rate us on Discovery?

We asked our members to rate the statement “Cigar Sense recommendations often help me find new cigars that I would not think of otherwise”.

Even if we might challenge their beliefs, we are very honored to see their score: 9.1 out of 10.

Share this with your friends!