How Does Segmentify Work?
The right offer, product or campaign, to the right person at the right time.
Process Steps
1. Collect Visitor Data:
Track the behavior of each visitor, such as page views, clicks and past purchases to make 1:1 personalised recommendations.
2. Segment Visitors:
Analyse visitor data to create customer micro-segments and assign them into those segments according to their behavior.
3. Identify and Group Similar Products:
Collect static data about product aspects like brand, category, price range, color and so on. For each product 3 groups of products that are suitable to be recommended are identified, which are Alternative Products, Upsell Products and Cross-sell Products.
4. Personalise:
Match the best performing product and campaigns with appropriate customer segments to personalise online shopping experience and increase conversion.
5. Continuous Learning:
Collect visitors’ responses to recommended products and campaigns as a feedback loop to increase effectiveness.
6. Output as Insights:
Perform detailed analysis of most and least successful products and campaigns, to create insights that will assist e-commerce managers in deciding future strategies.
All Processes in Real-Time
Today, the most sophisticated recommendation and personalisation engines rely on lengthy data collection phases consisting of processing and analysing data, and then calculate targeted recommendations only after the customer is long gone. This classical method is time-consuming and ineffective, mostly because customers are no longer engaged in buying from you after having left your site. Segmentify’s advanced and innovative recommendation engine fills this gap by generating on-the-spot recommendations, while the customers are still online and ready to buy.
On average, a potential customer remains on a website for 2.5 minutes and checks 10 products and 2 campaigns during a visit to an online store. An efficiently functioning recommendation algorithm has to be able to analyse the online behavior pattern of a user in this short — timeframe and combine the most relevant products and campaigns for the user. Otherwise, that visitor would be easily lost to another competitor. This is why having an algorithm that works in real-time is one of the crucial key points of running an online store.
Segmentify’s leading recommendation engine eliminates the time issue by learning the user’s behavior in seconds. This is possible with real-time updates to underlying customer scoring and segmentation system. With every visitor information about page views, clicks, purchases, the Segmentify updates not only segment and score that specific visitor, but they also recalculate the overall weight of each dimension to create recommendation models. These models are based on the average time on page, page views per session, average basket size, etc. This leads to getting higher values from click-stream. Each visitor action updates the customer model and scoring, and Segmentify’s engine returns the recommended products and campaigns for the visitor in real-time
In span of days, Segmentify recommendations can match classical recommendation engines and outrun them with quick feedback loops from visitors. With real-time recommendation model, Segmentify can offer the right products to the right customers at the right time, which results in higher conversion rates and a significant boost in sales.
Cutting Edge Technology
The advanced recommendation algorithm that enables accurate personalisation is based on the PhD studies in machine learning of Murat Soysal – one of the two co-founders and CEO of Segmentify.
Segmentify’s primary differentiation point is replacing current traditional approaches based on bulk and batch processing with a real-time approach, which is responsive to trend changes. Our segmentation algorithms handle each visitor one-on-one and in real-time.
Due to seasonality and the campaign-oriented marketing nature of eCommerce, trend changes are very common. Our real-time personalisation algorithm is responsive to these fluctuating trends surpassing the conventional, one-time-only customer analysis.
Our technology enables online retailers to recognise and engage with each shopper — one at a time, in real time and at scale — to deliver the most relevant experience and make individualised product recommendations.
Excellent Customer Services
We are very keen on providing our customers all the help they need. Our Onboarding and Customer Success teams are always ready to help and enable you to get the most out of the platform.
Super Easy Integration & Onboarding
Just plug and play! Segmentify can be integrated with any online store in just 1 day by adding a single line of JS code — no IT team required. Personalisation campaigns go live on the 2nd day and 10% additional revenue is generated within 2 weeks.
Custom Recommendation Rules
You have full control over the recommendation criteria you would like to run, so that the process suits your business goals. Segmentify enables you to customise the recommendation algorithm and allows you to combine custom rule-based and smart recommendations delivered by machine learning.
Real-time Conversion Analytics
One of our most distinguishing solution is Real-time Conversion Analytics. Taking the term ‘real-time’ by heart, we were not happy with the classical analytics tools, which give out quite restricted reporting. Having access to detailed reporting after a visitor has left your site doesn’t meet the dynamic needs of an online store. With our analytics solution, our clients can get insights on anything that is occurring on their online stores in real-time. This way, it is possible to make informed business decisions based on real-time data, without having to wait for hours or even days in some cases.
Our analytics solution enables users to identify the best performing products, channels and banners, and lets them make the right investments and increase the ad ROI.
Real-time information is the key to a successful online store — that’s why our analytics solution is a separate entity that can feed all CRM platforms and can be used by anyone
Originally published at https://www.segmentify.com.
