AI-driven Competitive Intelligence

These days businesses have great visibility internally, into their sales, marketing, product metrics and strategies, but lack visibility externally.

To outdo competition businesses should implement Competitive Intelligence services that convert data into intelligence that can provide a real-time view of your competitive landscape. We’ve found great value in machine algorithms that can help identify trends and insights in vast streams of data and make faster decisions that potentially position businesses to be competitive in real-time or even ahead of time.

By competitive intelligence, we mean the company’s efforts to gather and analyze information about its industry, business environment, competitors, and competitive products and services. The gathering of the information and the analysis will support a company’s strategy as well as identify competitive gaps.

In this blog, I talk about three main areas of competitive intelligence in which we believe there is great value in applying AI. I will talk about the best competitive intelligence sources and the role of web scraper agents.
Although AI-driven competitive intelligence provides a greater competitive advantage, there are limitations in the web scrapers capabilities that we (Skim Technologies) fixed by building our own agent; the Skim Engine™. More on web scrapers will be explained later in the blog.

Unstructured & external data

Firstly, the best thing to do is to systematically find the best competitive intelligence sources, based on what your goals are, to create competitors profiles. Also, processing real-time data and up-to-date data is crucial to gain accurate insights.

Competitive Intelligence from online, external sources include::

• Company websites for insights into target audience, shifts in strategy, pricing product, and so on.
• Company press releases for new product, personnel changes, or expansion news.
• Social media postings, especially if the company is introducing a new product or service that hasn’t yet been launched.
• Job postings, types and number of positions available indicating a potential new category development or a new product.

AI-powered tools are well positioned to harvest data from millions of webpages to surface valuable business insights. These tools are able to track a competitor’s complete digital footprint, both on and off their website. The most critical updates are those that aren’t announced in a press release but rather hidden in websites or in a customer review. It’s down to the web scrapers ability to provide high-quality data by unearthing these gems.

Web scraping agents are already popular and used to access unstructured data sets from the web. Once the data is collected, structured and organised, an AI product can automatically categorize and prioritize the data to enable users to review it. As scrapers are able to extract a vast amount of intelligence data, it’s critical to surface the highest priority updates and act on them before its too late.

Competitive Intelligence

At Skim, we’ve identified three areas of competitive intelligence where we can harness the power of external unstructured data:

1. Monitor job roles

A company can monitor news, competitors’ pages and social media for trends of staff from the interested employment category (i.e., Chief Executives, Data Experts, AI developers), joining or leaving or moving to and from competitor companies. This would provide valuable intelligence by signalling that a competitor may be in trouble, takeover, moving into a different area or implementing new technologies.

2. Monitor price changes

In a fast-changing online market, businesses are aiming to remain competitive and certainly with e-commerce customers are often price driven. Using price data collected from millions of e-commerce sites across the market you can gather competitor price trend data 24/7.

Price Monitoring could also alert you when your competitors prices change, it can also tell you a lot about your competitors pricing strategy (i.e., Christmas discounts, sales, etc). Through additional sentiment analysis towards your competitors products, you can also track how it affects price over time. It will put you in control by learning from your competitors mistakes and wins.

3. Monitor Competitors Product or Brand Hype

Using the latest Narrative Extraction techniques using Natural Language Processing, you can monitor specific underlying social processes that indicate hype surrounding a brand or product through the narrative within social and news data. This will provide you with a clear indication of when a product or brand is beginning to pick up virality or buying trends, so you can counter the movement with your own product launches or announcements.

Web scrapers limitations and SKim Engine solutions

Rule-based web scrapers are limited on their ability to work independently. Web scraping tools are usually made to handle simple websites if they use dynamic elements a scraper tool might not be able to extract data. Not only, as websites are updated quite frequently, many of these changes can render the scraper tool useless. In the latter case, you would lose the data and will be forced to update the tool to make it work with the new changes on the target pages, very onerous.

For example, if the goal is to extract prices from a website, the scraper would be trained to find that information in a certain position of the site. If this website changes the scraper won’t be able to get the information you’re looking for and/or it will provide you with incorrect data.

We are constantly updating and adding new features to our own web scraper agent; the Skim Engine™. The Skim Engine™ is a rules-free agent that extracts and organises unstructured data in a machine-readable way, so changes to websites are a thing of the past, and you can gather reliable, accurate competitor data with ease.

Originally published at https://www.skimtechnologies.com on June 5, 2019.