Tides are important in many fields, such as port planning, nautical charting, and bathymetry surveys. However, the distribution of tide observation stations in Indonesia is still uneven, so simple tidal data collection still uses tide palms to observe the movement of sea levels. The manual tidal observation method has limitations because visually the ability of the observer to see the tide reading is different, so it is subjective. Therefore, the solution to this problem is a tidal instrument that is easy to make, provides real-time data, and produces accurate information, saving energy and time. However, using automatic tide gauge instruments is still not affordable in the Indonesian market. This study creates a low-cost pressure tide gauge by integrating a pressure sensor with an Arduino microcontroller. In addition, the data monitoring process uses the Internet of Things system. Data from the instrument measurements were uploaded to the Thingspeak website live, allowing observers to monitor in real time. Instrument’s data with noise and anomalies will go through a data filtering process with the moving average and moving median methods. Testing the instrument’s error rate compares the data generated by the instrument with tidal data from the Indonesian Geospatial Agency (BIG). Both tidal data went through calculating Root Mean Square Error, or RMSE. Based on the instrument error test results, the RMSE value for moving average data against BIG is 27.963 cm and moving median data against BIG is 27.787 cm.