LabelsDB

What is LabelsDB?

LabelsDB is a service which produces millions of professional images and textual attributes for Consumer Packaged Goods (CPG) within reasonable time-frame and cost. Such images and textual attributes are essential to build catalog and search in online store. Needless to say, any substantial e-commerce web site or application must provide product search and high-res images, as users are used to fast searching and seeing a product before making a purchase decision. Moreover, Google, Bing, and Yahoo become more and more image-aware giving more priority to pages with rich content.
With the help of LabelsDB you can present your SKUs online without hassles or infrastructure investments. Visit Zakaz.ua to try out in action one of the sites utilizing data produced by LabelsDB!

Why data is so important?

Consider a problem of opening an e-commerce site trading out from brick-and-mortar supermarket with 50,000 SKUs. Where to get product images and descriptions for all these items? Here are several pitfall ideas which typically lead to crisis in a project:

• let's collect product pictures from producers' web sites, or ask them to send images
• data is less of a problem, let's find an open-source/freeware e-commerce engine/CMS (there are dozens), rapidly build our online site, and then tackle a data issue
• let's disect a problem: let's take merchandise titles from accounting application and push it into an engine, while images – there must be an open database of those, and we are going to find it.
Unfortunately, there is no free way to get precious data in grocery market.
But still, you can get all you need in one place from LabelsDB using it as on-demand service.

F.A.Q.

How product photography process works in LabelsDB?

GVMachines has created a technology for automated product photography. Hardware side of technology is a device we call a "scanner" or "scanbot". Scanbot is essentially a mobile light-cube which allows fast and predefined change of illumination settings based on product category. A software part of scanning process runs on servers and deals with cropping, image sizing, colors, rotations, fingerprinting, image versioning, etc.

Why is photography service scalable?

LabelsDB services imply rapid training of staff in-field. Scanbots are delivered with a trainer and exhaustive operation manual. No special skills or knowledge are required from scanner operators before the training, - thus almost any supermarket worker can successfully fulfil scanning work. Plus we can supply required number of scanners to any part of the World where delivery works.

What CPGs are already available in LabelsDB?

As of now, LabelsDB as database contains mostly grocery products sold in Eastern Europe.

What is the difference between the "scanning" and traditional product photography?

These two processes tackle different problem size. Traditional commercial photography are done majorly for advertising purposes, making fairly small number of objects look attractive. Such approach often requires creative mind of an artist, putting object into a context, creating a composition, etc. The work is hard to make every shot perfect, including long post-production (editing shots on a computer). At the end, there are several high-resolution files ready to be used in print, web, and probably outdoor.
Our product photography has web first in mind: almost everything is automated, photos are more uniform, produced quickly and cost-effeciently. Final shot is accurate enough to attract shopper and provide her/him with all important information.

What's the difference between images of LabelsDB and images at photo-banks?

Photo-bank's images of grocery or naked food are created by designers and have no name or brand on them. These are a no-name images useful only for decoration of e-commerce web site, but not as a product information.

What is so special about text attributes in LabelsDB compared to those in retailer's database?

Retail professionals know, amount of symbols used for product information, such as title or description, had limitations in most of thre retail software for many decades. Whereas, in the online world, the more descriptions you have the better. Full names and attributes such at producer, trademark, nutrition facts are essential for leading e-commerce sites.
LabelsDB service enriches and cleans available attributes. For example, from: "Min water Vittel 0.5l non-sparkl. France", we would produce a complete granular attributes: product{mineral water}, producer{Vittel}, volume{0.5l}, property{non-sparkling}, country{France} with curated translations into several languages.