Entirely automates the pricing of all items in the webshop hourly or daily without any manual intervention ever. 

Our mathematical model understands micro and macro trends deeply and reacts to market dynamics quickly to maximize the interests of both the seller and the buyer by approaching a Nash equilibrium. It maximizes profit for the seller but also keeps customers coming back long term by constantly converging toward the optimal prices including every possibility. 

Copying the price and undercutting it does not guarantee maximum profit. It also decreases the profit hugely for the whole sector by cutting each other’s prices continuously. The optimum is hidden from human thinking.

The trends keep changing with no stop and there is a delay in every action and reaction. For the time the owner decreases the price, the customer may already lose the interest, or it would have already bought the item at a higher price.

Also the number of factors affecting the market are always much greater than those fixed numbers of competitors that one can keep monitoring. You cannot monitor the whole market, so it's always partial information. And you cannot know whether the price of the competitor is an optimum either that may be based on a human’s unprecise decision.

Our solution however is able to consider all factors affecting the market by making decisions from the most valuable information constantly. This is the reaction of customers for the prices. And since the decisions of the customers are affected by the whole market at their own levels in the given sector, and our model is able to consider all these vibrations computationally, therefore our optimization mathematics is continuously able to target the very customer base in the most efficient way that the web shop can expect.

Even if competitors keep cutting prices, our model will react to it with optimum.

Because classical machine learning algorithms need much and unnoisy data to be able to provide good decisions instead of random noise. In the case of sales events, there may be only a couple or even no events registered for an item.

Also these historical data are usually very noisy with random peaks. No correlation between items or within periods of the same item. No forecasting or regression models work well enough.

The other issue is that new prices must be offered and discovered continuously to follow trends because the market keeps changing fast. A so-called time dimension must be considered which is a hard task for the algorithms when there is not enough data.

The owner has 2 options: either never change the price or change it. The first causes bankruptcy for never being optimal and the second brings up the very question of when and how to do it. Since classical algorithms must depend on historical data, an off-the-shelf method cannot work. Highly specialized solution must be developed that we have carried out with several years of research.

Some of the apps in the same app store, targeting the same dynamic price optimization category. But as we see, they are much more difficult to use. Not even entirely automatic like ours. Manual administration is needed to create rules and set limits per item, define time points for the automation to change the price with specific percent and similar. This is troublesome, especially for many items and in a recurring way.

Yes, we can open our API to give access to our mathematical computations for other platforms or external systems.

Bringing top tier automation with high level mathematics for such a hard task that is a heavy burden for many businesses every day or even every hour.

The targeted platform of Shopify has more than 5 million active web shops operating already that are in need of such a novel solution.

Everybody can understand the problem and none seem to be able to give such a convenient answer.

We are a professional team from Hungary and Slovakia, Europe, doing mathematical research and developments with several decades of experience.

See more at honama.dev/pricing