How can you identify personal trends?

Contribution image personal trends

In the last trend report I received, there was a detailed report on dresses, skirts and tops – of course, the fashion industry is largely oriented towards women. But as a man in my early 40s, none of these trends suited me. That’s not entirely surprising. But even in the trend reports on the latest sneakers, there was nothing that appealed to me. Is that because of me? Not entirely. It’s because an editor has to decide on a small number of trends, even though he could report on so many. I’m sure there’s a sneaker trend that would suit me, it just wasn’t chosen by him. If a report on all possible trends were set up, an almost infinite number of pages would come out. However, it would be far too general, far too much, and thus inconsclusive, because what is one supposed to derive from it?

Sebastian personal recommendation
Personal Trends Englishwoman
Personal Trends Italian
Personal Trends Bankers

In the best case, a report should list the trends per gender, age, category, culture, background, occasion, etc., so that it also fits the reader. Because for a 20-year-old English woman, something completely different is the trend than for a 30-year-old Italian woman. There are also differences for men. For a 50-year-old banker, for example, something quite different is in than for a 25-year-old student.

Since such distinctions are very relevant for us at Picalike, we tried to get an indicator with the following parameters:

Gender: 3 (men, women, unisex)

Age: 6 (Baby, Child, Teen, Young Adult, Adult, Senior)

Occasion: 6 (Basics, Casual, Chic, Business, Sporty, Rocky)

Categories: 7 (Dresses, Blouses, Trousers, Jackets, Shorts, Shoes, Sweaters)

Attributes: 4 (colour, pattern, length, extras)

Type: 2 (Extrovert, Introvert)

Season: 4 (Spring, Summer, Autumn, Winter)

Weather: 5 (Warm, Cold, Rainy, Sunny, Snowy) Ba

This simple list, which we try to make even more precise every day, would already generate over 30,000 reports. That is a lot of effort! And unfortunately still far too much. Fortunately, we are developers and can have our artificial intelligence monitor this so that it only returns us the relevant results.

When I look back, I see the long way we have come, although we still want to go much further. For example, it was a special experience when we received the first sketches of trends to better visualise the data from the reports. That was our beginning. Now, with OSA and especially with our trend analysis, we are already quite a bit further along. Nevertheless, the sketches of trends are still an event. That drives us on. It remains exciting!

Get relevant informations fast with Newsletter Intelligence

Newsletter Intelligence

The time between the holidays is known to be used for resting. Not so with us. We are busy developing, optimizing and researching. The result is a number of new features in our OnSight Analytics tool. To present all of them now would go beyond the scope of this article… And what fun would it be to shoot all the powder at once instead of finely portioning it and thus benefiting from it for a longer period of time?

Be curious what we have to report in the near future!

But now to an exciting new feature: the Newsletter Intelligence!

What is the Newsletter Intelligence?

With one glance you can see how many newsletters have been sent. When you mouse-over the points, you will receive several pieces of information, for example: How many newsletters were sent on this day? In addition, you can find out which coupons are offered, which collections and freebies are advertised. Of course, you can also click on the individual newsletters from this date below to take a closer look.

Newsletter Intelligence

Instead of subscribing to your competitors’ newsletters yourself and going through them manually, you can let OnSight analyze tens of newsletters for you and stay up to date with your competitors’ marketing strategies.

Newsletter Intelligence

Are you curious? Then test our demo and see for yourself what is possible!

F#ing 10 years already gone

10 Years Picalike
Sebastian Kielmann

A review by Sebastian Kielmann, CEO Picalike GmbH

The idea

Almost exactly 20 years ago, I switched from research on text indexing and search to image analysis and feature extraction. I was attracted by the fact that images are language-independent and rarely have duplicate meanings. When I first started working on image analysis, I mentioned this to my boss at the time, who had a PhD in physics. His reaction to this was decisive for the 20 years that followed: he said that what I was trying to do was not feasible. There was no at that time, AI was not very popular, GPUs were not really used for matrix multiplication yet and corresponding papers from research were hard to get if you had an extremely slow line in deepest Walldorf / Wiesloch like me.

But where there’s a will, there’s a way… It was then that I realized how open and willing to share the research community was. There was no Twitter, Facebook, Instagram or anything like that. For me, that time doesn’t seem so long ago, but my kids can hardly imagine such a world. Conversations and exchanges were through conventions or mailing groups. For those who have also been analyzing images for a while: Open-CV was just getting started at that point.

I was researching image analysis ⎼ as a hobby on the side ⎼ and studying management and marketing, working in e-commerce or consulting. Then I met Dr. Tilo Höpker during a flight and we became friends. Over a meal together, I presented my idea to him: a recommendation technology whose basis would come from data from product images, rather than from texts. The idea had come to me because I was working on a product search engine at the time and saw how much erroneous data there was in the product feeds and how much information about the product was missing, even though everyone could see it, based on the product image.

Tilo thought the idea was great, would also jump right in to start a company that offered that. But he wanted to get a second opinion. At the next meeting, Daniel Raschke, who has since become a good friend of mine, was also present. We didn’t know each other at that point, but we were immediately on the same wavelength. The three of us started Picalike GmbH in October 2010. 10 years ago now.

The Start

F#ing 10 Jahre schon vorbei 3
Our first logo

My first lesson was that primarily a market has to be ready for a technology, not the technology has to be ready for the market. Anyone who has ever had to bring a technological product to market knows what I mean. I’m not a born sales guy, like to avoid people and crowds, don’t want to be in the spotlight, and talk a lot about the technology and too little about the benefit. But anyone who founds a start-up also has to take care of sales. The first customers were found quickly. and quickly went from customers to development partners and then friends and part of the family.

F#ing 10 Jahre schon vorbei 4

Our first office in Frankenstraße

In addition to the pure data from the images, customer behavior and correlation among products, as well as filtering by all data about the product, were soon included to provide even better recommendations. Similarity Recommendation (similarity detection, like “similar items”) was then joined by Complete-the-Look Recommendation (outfit recommendation, like “this goes with that”). For us, as for any startup, the time quickly began when you have to do a balancing act between customer support and further development. Not every customer needs the same thing, and we were a fine but still small team. During the day I helped with customer support, in the afternoon I pitched Picalike wherever I could, and in the evenings there was further development. There were days when I thought we were going to take the next giant step (technologically or revenue-wise) and there were days when I wondered if it was really a good idea to leave a secure job. I don’t regret taking the leap into self-employment, but there are many things I would do differently today.

One day my phone rang (it was a Hansenet line back then, who remembers it?).

F#ing 10 Jahre schon vorbei 5

One of our first home

It was a young guy from the OTTO Group who had read an article in the Chamber of Commerce magazine. We talked and he did some intros on, Bonprix, Shopping24 and others. This gave me the opportunity to introduce our technology to the big players. This phone call was crucial for the next 9 years. I presented not only to the aforementioned companies, but also to the OTTO Group itself, which decided to invest in us. Before signing the shareholder agreement, we had resolved that when it was signed, we would all go mega partying. We had wild ideas about parties at the Doll House on the Reeperbahn or about organizing a big private party. And then this: after the signing, everyone went home, went to bed, slept for a few hours and the next morning the work started all over again.

“After the game is before the game”

In no time, we assembled a highly motivated team, moved to a larger office and invested in marketing, product development and research. We quickly grew from a handful of customers to dozens of customers, opened an office in the US (first SF and then moved to NY), started partnerships with prestigious companies, and were pleased with the growing interest. Everything was looking good. We grew, new clients came in, became more and more international and had very exciting projects and cases in mind with us.

F#ing 10 Jahre schon vorbei 6

One of our legendary summer parties


As time went on, we noticed that the momentum of our growth was slowing down. Projects were getting more complex, budgets were getting smaller, competition was getting more diverse. Our approach of “customer request -> research -> project -> product” was showing its weaknesses. We had to admit to ourselves that research is fun, but not so easy to scale financially. And so we analyzed the weaknesses of our business model, the strengths of our team and system, and went for a start-up within the start-up. In other words, a soft opening.

“A few steps back for the run-up”

F#ing 10 Jahre schon vorbei 7

Our current home

Welcome to the world of Portfolio Analysis 🙂 Through several analyses, we realized that we were too dependent on the availability of technical resources at our clients, that we needed to create a quick and easy entry point into our Picalike world, and that we needed to deliver demonstrable, repeatable value to our clients. Thus, we started with multiple A/B tests, case studies, surveys, interviews, and new positions within the team. It was also important that research remained a high priority in our day-to-day. With the help of our customers, partners and the team, we started the development of OnSight analytics. The planned development time was half a year. However, reality is not a good friend of planning: two years later, we launched the open beta. Some challenges were bigger than expected, the day-to-day business could never be neglected, and we had to learn a lot from our pilot partners. But from the beginning we had committed ourselves not to compromise on quality, even if it would take longer, and we still stand by that. Even at a time when Corona has changed all our lives.

The last few months have been like the first few months: a lot of ups and downs. Moods of optimism followed by setbacks. But what remains is that in the last 10 years no day has been like the other, many stories have been written, we have learned a lot, and now for the next 10 years we want to continue to learn and put into practice what we have already learned.

Thank you to the team for the incredible support and loyalty, to our clients for the great and always exciting collaboration, and to everyone else we have met along the way over the last 10 years for enriching our lives and shaping our path.

Stay healthy and take part in the next stages of our journey.
Many greetings
Sebastian Kielmann

Information extraction from websites (focus on product details)

Information extraction from websites

Last Tuesday, a crowd of software developers and data specialists gathered in our company to listen to the words and tips of Timo Schulz. Timo is a former employee of Picalike and now a consultant at ITGAIN Consulting. As a specialist for artificial intelligence and in particular machine learning, deep learning and data processing, Timo advises companies on advanced analytics and AI.

The topic of the workshop, which was attended by 20 participants from different industry sectors, was “Information extraction from websites with a focus on product details”. In other words, how do you get structured data from unstructured texts?

The first part of the workshop dealt with the theory: From RegEx to Neural Networks Timo tried to explain the topic text analysis and text mining to the interested tech professionals and to clarify which problems can be encountered with product texts in e-commerce. After a short break, it was time to get down to business: The laptop keys were actively typed with many “hands-on” examples and a lively exchange took place on learned techniques and new application examples with many tips and tricks.

Afterwards we had a cool beer and a delicious pizza and I had the chance to ask the workshop participants and Timo a few questions.

Interview with Timo Schulz

I’ve always wanted to get more out of data.


Where to find which product data and how to process them in a structured way

Where to find which product data and how to process them in a structured way

Why did you decide to work in artificial intelligence?
Already in 2005, during my computer science studies, I started to work with data. I’ve always wanted to get more out of data and did a lot of research in this area. But then I wanted to get out of research and put my knowledge and technology into practice. That’s how I came to Picalike.

Then why did you go to consulting later?
I wanted to get out of the e-commerce business at some point. It was very exhausting and nerve-wracking to bring the technology close to the companies. Often the companies were convinced by the product, the technology, that it worked, but then it partly failed because of political decisions within the company or there was no far-reaching understanding for it. Of course, it is difficult to remain highly motivated. In consulting, I can now advance AI in all areas and show companies without pressure what is possible and how they can implement AI in their companies.

You often have to do a lot of convincing.

What challenges do you see for e-commerce in terms of AI?
The biggest challenge is actually to correctly recognize and assess the potential of AI. And the acceptance: The company has to recognize for itself what AI can do for itself, i.e. for the company. You often have to do a lot of convincing.

Has there ever been a case where you advised a company not to use AI?
No, not really, because AI is so versatile. But sometimes you have to be careful that AI is not just seen as a trend. According to the motto: “We absolutely have to do something with AI now”. Here it is often sufficient to simply structure the existing data in the company better and to see what we can already get out of this data.

As a consultant you should always stay up to date. How and where do you find out about the industry, about new developments in the field?
As far as possible, I dedicate a whole day to research. I read a lot about the topic, follow blogs, listen to lectures by people I follow and then try to implement my own use case as a prototype. So I can then decide whether this approach makes sense in my eyes, whether the topic should be pursued further or not.

The tech professionals are eagerly listening to KI guru Timo Schulz

The tech professionals are eagerly listening to KI guru Timo Schulz

And which trends are exciting at the moment? Where is the journey going?
I think everything about NLU or NLP (Natural Language Understanding or Natural Language Processing, editor’s note) is very interesting and a lot will happen here.

Speaking of language comprehension: I recently read that it has not yet been possible to teach artificial intelligence humor. Is that right?
Yes, it’s not that easy indeed. When, for example, the customer rating in an online shop says: “The shoe is huge, like a VW van.” Then we understand: “Okay, the shoe is most likely quite big. And it was just a bit more fun to paraphrase it.” But the AI would actually compare the shoe with the size of a VW bus. AI just doesn’t think any further. Another example: Jan goes into his bedroom and gets his ball. Then he goes into the garden and puts the ball on the floor. Where is the ball? For the AI it is not clear that the ball is now in the garden.

I heard from a reliable source that you used to be a Picalike beer ambassador. What is your favorite beer and why?
Clearly Sierra Nevada Torpedo. Ken Grossman is a hero! He revolutionized the art of brewing beer. In the 80s he went to Germany and bought a copper brewery there, which he then took back to California. And from then on the beer became simply unbeatable. They use whole hop cones for the beer, not just hop extract as others do, and produce part of their energy themselves via solar energy. When the big California campfires happened, Sierra Nevada brewed a special beer and donated all the proceeds to the victims of the fire

Interviews with the workshop participants

Interview with Lennart from


Interview with Lennart from Shopping24

Interview with Lennart from Shopping24

What is your position at Shopping24?

I am Search Engine Linguistic Manager.

And what exactly do you do in your job?
I help with the processing of search queries. What do users enter as search terms and I take a look at what, for example, linguistically all around it must be captured in order to output the best possible search results.

Why are you in this workshop?
Since I also deal with product texts in my job, I find it interesting to see how information can be extracted there.

Which topics for further workshops would be interesting for you?
In general, I am interested in product search challenges. For example, insights from other website operators who are also involved in product search would be interesting. What challenges do they have and how do they solve certain problems?

Interview with Sarah from AdSoul

What’s your position at AdSoul?
I am a linguist.

And what exactly do you do in your job?
I break down keywords and try to cluster them. A grammatical processing of keywords so to speak.

Why are you here? What are you interested in the workshop for?
First you have to explain what AdSoul does. AdSoul is active in the field of SEM and takes care of automated search engine marketing. Already at university I was involved in text mining and the preparation of data and texts. The goal of AdSoul is basically to create automated text ads sometime. That’s why data extraction is so interesting for me.

Interview with Marc-Olaf from OGDS

What’s your position on the OGDS?
I am a software developer.

And what exactly do you do in your job?
The OGDS is a Company Builder. We identify new and attractive business ideas and build prototypes for them. We provide the operation, the infrastructure and the architecture for these prototypes and I develop the software for them. So basically we provide a technical solution in the area of e-commerce.

Why are you here? What interests you about the workshop?
I’m interested in extracting from texts and I’m interested in what other people are doing in this area, what new ideas are there in this area.

Did you like the workshop and if so, what exactly?
I was primarily here for the exchange, not so much to educate myself professionally because I already know this topic very well. But I think Timo explained the subject very well and captured the breadth of the topic well. This enabled me to draw out interesting ideas and, in part, new perspectives.

Which topics for further workshops would be interesting for you?
I am always very project-driven. At the moment I am very interested in the topic extraction of data from pictures. Therefore I am also happy if I can exchange myself with Picalike on this topic.

Interview with Erwin from


Interview with Erwin from Shopping24

Interview with Erwin from

What is your position at Shopping24?
I am a Java developer.

And what exactly do you do in your job?
I prepare product data in e-commerce. I take care of the product search at Shopping24 and the support of the back-end systems.

Why are you here? What interests you about the workshop?
On the one hand, I’m here to expand my own knowledge. On the other hand, at Shopping24 we use product feeds. The aim here could be to extract text from external websites without feeds.

Editor’s note: The interviews were recorded in a protocol format.