Electronics Store Promotion System | FISHKI.UA

Background

Work on the project started in 2016. The task was to launch and scale the online direction and build a multichannel distribution.

A comprehensive solution for creating a marketing, IT, and service e-commerce infrastructure capable of not only interacting but also increasing sales in the offline and B2B directions.

Omnichannel Analytics

Considering the company's multichannel distribution model, which combined sales in a network of retail stores, active sales on marketplaces and wholesale channels, as well as in its own online store, it was important to evaluate the effectiveness of promotion investments based on a multichannel sequence. This allowed us to:

Sales Channel Chains

Based on the understanding of how each segment of buyers makes a decision about the choice of purchase location and interacts with the offer, a system was built to evaluate the effectiveness of each channel and advertising campaign, considering its place in the sales funnel.

With separate remarketing campaigns for each stage and a channel work tactic based on associated profitability and return within the framework of building chains, in addition to standard promotion channels such as Google Ads, social networks, and SEO promotion, we used:

Flexible Shopping Campaigns

Given the large number of assortment items, dynamically changing stock, sales, margin, and prices, a system for setting budgets and bids for each product, as well as a group of search queries, was implemented based on key financial metrics.

Stock and Turnover. The algorithm determined whether it was important for us to sell out the product or, due to limited stock, not to force its sale, considering the higher cost of advertising in Google Shopping than in other channels.

Delivery Time. The influence of delivery time and final conversion, with the distribution of stock among 140 warehouses across Ukraine, helped to determine the click cost for each product depending on the region where the client was located.

Margin. The purchase price for the same product could be different. We took this into account and performed automatic bid correction depending on the margin and the expense limit per order cost for each individual product.

Selling Price and Competition. Considering the prices for similar items from competitors, depending on the difference between the minimum price and ours, the average price, and the price of top players, we determined the probable conversion and, accordingly, the click cost for a specific SKU.

Product Feed. Integration with CRM, 1C, stock, competitor price database, and end-to-end analytics, changing all attributes, bids, and budgets in real-time based on the forecast of profitability and return for each of them.

Geographical Targeting. For each of the 54 sales regions, based on turnover, profit, operational, and marketing expenses, a profitability coefficient was applied when setting bids, budget, and the promoted assortment.