Pricing automation system
Managing a vast product range can be a pain, especially when it comes to imposing a pricing policy. Keeping tabs on competition pricing manually becomes impossible at a certain scale. One of our oldest clients has (happily) crossed that scale threshold and asked for our help with this new challenge.
Understanding the process
We build a lot of process automation tools at Laskowski.pro. Every minute saved by our clients and their employees on repetitive tasks is a minute that can be better spent on creative tasks. Firstly we always attempt to examine and understand the process that is to be automated to asses which steps can be automated. Starting with the theory we met with our client to discuss the theoretical details of their pricing strategy then, to confirm if the theory is applied in practice and what kind of exceptions occur (and they always do!) we witnessed first-hand how they determine new prices for some of their offer. At the end of the day, we have established that our aim was to enable our client to pursue a highly competitive and aggressive pricing strategy to maximize incomes.
By analyzing the current manual process of price adjusting we distilled three main factors that affect price change. Suggested retail price, product margin and prices in competitive prices.
Making it work
Luckily each product in the e-commerce system database already had suggested retail prices assigned. We also agreed that margins are rather homogenous across each brand and do not vary greatly between different products. Therefore we created an interactive table where our client can input and edit maximum allowed percentage price drop (margin drop) from the suggested retail price for each brand.
The last information required is product prices in competitive shops. The good news is that all prices on the Internet are visible publicly so we just had to collect them. The bad news is that there is no easy way to do this. Our aim was to collect as many prices as possible while avoiding false positives. In the end, we build a complex price monitoring service fed by a custom web scraper that harvested prices from selected competitive stores.
Once we had all the information we could implement new price calculation formula which basically boiled down to trying to outbid the lowest price on the market but only if it didn’t mean sacrificing more margin than agreed for this particular brand. Prices in e-commerce system are updated daily during the night by a batch script.
The price monitoring system has been working since 2016 and collects thousands of prices every day, we are all very pleased with the results and the priceless data it provides. As an extra feature, we generate full reports 3 times a week that include a list of anomalies – products much more expensive than competition average and vice versa, and a table with all collected prices for each product which gives a unique insight into the market.
Automatic pricing integration followed shortly with a few safeguards added and works uninterrupted in the background giving our client an edge over his competitors.