Solutions

In addition to delivering the curriculum, the training system is also actively involved in managing processes. This means that it now undertakes not only educational, but also HR tasks.

An online training platform is indispensable for ongoing internal training, but it is equally useful for business development: it can be used to easily introduce new products or services.

You can find more details about our services on uzleti-oktatas.hu.

Regardless of the size of your business, it is important to see how your operations and business opportunities are affected by the rise of AI-based solutions.

As we are involved in developing such solutions ourselves, we can help you assess the risks to your business performance based on our experience.

However, it is not only threats, but also new opportunities that can likewise be found in new technology. Planimeter also has the capacity to assist in this regard.

• The algorithm for assigning calls continuously improves over time (thanks to the integrated learning algorithm), and increasingly subtly differentiated criteria can also be incorporated.
• Statistical calculations for identical call types make it easy to compare the activities of different operators and measure the performance of employees against specified KPIs. Reports are displayed on the dashboard created for the system.
• Comparisons also include a statistical toolbox, so they can be configured to highlight and analyse only, for example, conversation duration and voice pitch, that significantly differs from average.
• Integrated learning algorithms can be used to also forecast anticipated capacity needs. The forecast can be made for any horizon (for example, six months) and becomes more accurate every day.

This method can detect patterns of incoming and outgoing cashflow. Patterns are determined by taking into account the effects that are specific to the time series (seasonal effect, trend, cyclical or irregular movements).

Cashflow forecasting is used to better understand the future financial situation of a company, helping businesses determine the financial resources they will need in upcoming months, as well as identifying potential liquidity difficulties and financial risks.

Cashflow forecast modelling is be implemented as a web-based application. While creating the application, input data sources are defined and the data connectors are created. After that, modelling is completed in the “black box”, and the results are made available on a dashboard that complies with a user needs (for example, in the Power BI application).

As concerns modelling, even daily frequency can be incorporated, which mainly plays a role in reducing preliminary estimation errors.

In the report, the derivative results (such as a table) are generated at runtime, meaning that after any change in input data, the report can be updated easily and quickly (in one or two minutes) by running the report generator again.

In addition to its speed, this procedure stands out with its quality: since there is no manual data input, typos are entirely out of the question (notwithstanding the possibility that the error was already present in the input data, of course).

Conditional statements can also be formulated during processing. If, for instance, you want to use the verb form “increased” for a positive change and “decreased” for a negative change, then this choice can be integrated into the program, and the appropriate verb form will be entered in the report without any human intervention whatsoever. Similar statements can be made in connection with formatting needs: for example, a negative value can be shown in red (even in a table or a chart) and a positive value in green.

Reproducibility is another advantage. If a report is later found to be incorrect, the input data can be reconstructed and used as the basis to determine whether the error came from the source data or if the generator program may need to be modified. In the latter case, the bug can be fixed, usually with one or two hours of work, and all the reports thought to be incorrect regenerated in a matter of seconds.

Designing a personalised GA dashboard is no easy feat, even for IT specialists, which is why a dedicated team of specialists is tasked with building the interface according to customer needs. Furthermore, there is the risk that your painstakingly designed dashboard falls victim to arbitrary changes made by Google.

Once a customised GA database is created for a business, it is no longer impossible to create a truly customised dashboard supported by the GA database, but t can be achieved independently of the particular environmentt. To that end, all you have to do is connect directly to the Google Analytics database, and take your plans as the basis for creating the automatically updated dashboard that best serves your goals, even embedding it in your own reporting environment.

A dashboard that is created using your own data and tailored to your needs means that the information needed by sales/marketing, manufacturing or staff working on other planning can always be accessed in the same format. In addition to the above, there is also the option to derive integrated KPIs by including data outside Google Analytics, which would also be impossible to solve on Google Analytics’ own interface.

If the modelling environment can be linked to the webshop, then real-time recommendations can be made for the most common pairings. For example, if you have observed that buying baby formula often accompanies purchases of nappies, you can offer the frequently purchased product pair, even at a discount, when one of the products is added to the cart.

The issue of abandoned carts is another problem in shopping-cart analysis. Unfortunately, a higher percentage of purchases in online shops are interrupted, which is often seen in professional circles as an objective consequence of the shopping circumstances (for example, a purchase started during working hours has to be cut short). Still, since a problem related to user experience lies in the background in many cases (among other things, modifying the product, changing the number of items, or payment can be difficult), the analysis of abandoned carts and the related shopping circumstances, is essential.

Grasping the above phenomenon in a quantitative sense is similar to the method described above (pattern analysis, classification), but, as you would imagine, no recommendations are generated. In this case, the output of the system is an analysis of the frequency of abandoned shopping carts, whether according to cart value, purchase period, or any other factor.

The method is most commonly applied when estimating Customer Lifetime Value (CLV), and to calculate ROI (Return on Investment) when designing the marketing mix.

Machine learning and an artificial intelligence-based model can quickly shed light on the causes of churn. Once you have that information, steps can be taken to reduce or stop customer erosion.

With this indicator, which is also subject to continuous refining, in hand, a direct and well-targeted interaction strategy can be developed.

The purpose of data science-driven churn rate examinations is to understand why users abandon a particular product or service. For instance, in an online store, churn rate examinations can help identify customers who are most likely not to return to the site. The respective company can then work on improving customer experience and reducing churn rates. Churn rate examinations can be used, for example, in banks, mobile phone carriers, and the health sector, as well as at any business where a loyalty card has been introduced.

Then what you need is a balanced scorecard solution. The significance of the balanced scorecard is that it projects individual results, which are measured and scored across many dimensions, onto a single scale, converting them into a single, easily interpretable and ranked score.

The method is based on a tried and true mathematical model, which is why the solution can be considered objective. At the same time, it is not without subjective elements in the sense that each scorecard (for example, a system developed for sales representatives in a sales network) is valid only for the company in question.

The mathematical modelling of the scorecard consists of determining the mathematically optimal weight (multiplier) of each parameter, for example, a sales result expressed in HUF, when deriving the final score. However, the methodology provides the option of using manual weighting or a combination of optimal and manual weighting.

How does the creation of scores become an automated process? Either through a Power BI application or by implementing a simpler web-based application. The collection of input data is usually easy to organise. What is more, output sharing can even follow the organisational structure: sales representatives can only see their own data, regional managers can see their own network, while more senior managers can study individual or group-level statistics.

Before the appearance of ChatGPT, it was clear only to the experts that this methodology could be applied in any business – as well as economic, political and even artistic – sector and that its application would revolutionize that sector. While the algorithmic environment is also revealing entirely new ideas and approaches, the pace of development may even exceed the speed of human receptivity. However, the launch of ChatGPT in November 2022 has quickly led lay business decision makers to this recognition. The interest in AI-based solutions increased and very creative application areas (marketing support, chatbot, publishing, thesis writing, etc.) opened up in no time.

We can be sure that this is only the tip of the iceberg, unless such devices are banned or the pace of their development artificially controlled, as there have already been attempts to do, not entirely unreasonably. Planimeter has many years of experience in various AI-based developments, through which we are able to

• Cash flow forecasting,
• Determine attrition risk,
• Customer basket analysis,
• Customer journey analysis,
• Price optimization,
• Improve manufacturing process / call centre efficiency,
• Automated text analysis / reporting,

to mention a few areas in particular.

What is Power BI for? In brief, creating an enterprise dashboard, also including somewhat limited but essentially well and easily manageable data visualisation.

Power BI easily connects to the usual data sources via MS interfaces, and allows deriving new KPIs by supporting the creation of new variables.

Power BI also supports well-structured, complex and dynamic reporting, making it an ideal choice for the SME sector.

What is Power BI not for? The tool does not include analytics elements, so modelling, forecasting or optimisation cannot be performed directly in Power BI. Nonetheless these options need not be dismissed, since on the one hand there is Excel, in which an analytical toolbox is available and whose results can be imported, while on the other hand Power BI supports the integration of higher level statistical analysis tools (R and Python).