Best Practice

Analysing online Menus

with RPA and Web Scraping

Contacts and leads in the consumer goods industry

To sell products within the consumer goods industry in a targeted way, quickly generating new leads and establishing new and relevant customer contacts is important. The underlying CRM system, in which existing and potential customers are maintained, is the basis for this. The search for new customers is often not an automated process however, but rather is done manually by the responsible sales representative on the internet. From research to establishing contacts the effort involved is great, and has to be done alongside the day-to-day business.

Automating manual research with RPA

For a manufacturer of consumer goods in the catering sector restaurants, pubs and similar establishments are the main source of potential customers. General information, such as name, address and telephone number, must be made available to the sales department. Extra information, such as menu extracts, are useful in gaining an impression of the brands on offer.

Automated research, analysable data

4brands Reply, in collaboration with Macros Reply, has developed a more efficient and smarter solution which reduces the manual effort of research. Thanks to Robotic Process Automation (RPA), it was possible to relieve the field sales staff enormously in the research process. 4brands Reply and Macros Reply supported the company by using RPA to make profitable use of unstructured web data, thus creating a competitive advantage.

Bot research of keywords and images using Google

Using RPA, the research work normally done by the field sales employee over several hours can be handled by a bot which works much faster. The bot does its research in the same way a human does: For example by making a search request via Google. A list of results is worked through sequentially. The bot filters for previously defined information automatically.

The bot is not only able to extract standard information such as name, address and phone number, but can also read menu content stored on Google maps. If there are no menu links stored, the bot searches the images available. As soon as the bot finds a photo with a menu tag within Google images, OCR (Optical Character Recognition) is used to convert the photo and it is saved ready for the field sales employee.

Reducing the effort

The employee can start establishing contact directly and only has to add the newly acquired contacts to the CRM system. This scenario can be used not only for a general Google query, but can also be implemented on Facebook, Tripadvisor, foursquare or Twitter.

Structuring external web data for analysis

For the unstructured web data compiled by the bot to be subsequently usable by the CRM or ERP system, and thereby also for trend potential or turnover analyses, it must first be structured. This was made possible by 4brands Reply using web scraping, a process that pulls data from a website and presents it in table format. Information is thereby available in a structured format within a short timeframe. Existing internal data can then be enriched with the newly acquired data, and used for exploration and analysis.

Picture

web scrapping

The benefits at a glance

Picture
Data is current

Thanks to RPA, company data can always be kept up to date without spending a lot of time on research

Picture
Increased efficency

The smart search creation reduces manual effort and saves the employee a lot of time, which they can spend on the customer instead

Picture
Analysis

The more structured processing of the data means it can be used effectively in exploration and analysis

Picture
Extra hit rates

The use of OCR increases the hit rate, so additional information, on menu content for example, can be extracted from tagged images

Picture
Further potential

The bot can be used not only for lead management, but also for updating existing customers' contact details