A Yagi-Uda antenna, even poor constructed will in most cases have a clear directivity. At 2.4 GHz is directivity in need of free space. Indoor environment creates to much reflections so that any directivity is more or less random.
Most antenna design found at internet are theoretical antennas, that not will function well in real life. Calculated by people which not are able to verify result due to that it not is within their knowledge how to do such things or lack of even simplest measurement tools.
Gain and other predicted data for an ideal antenna can be hard to realize.
If the design not provides clearly measured result can you assume that given data is wishes, not what to expect.
As for all electronics, you need tools to measure that everything works as expected. Antennas are not excluded, it is opposite, good tools are especially required to achieve expected antenna result.
Antennas can be very hard to tune without tools.
If a flashlight is dead, how to find the problem, if it is the battery, switch or lamp that not is working? Without tools? It is very simple with a voltmeter.
Tuning an antenna is way more complicated then fix a broken flashlight. Doing that without tools is impossible. An alternative is to assume that what you have found at a webpage is a reliable and verified design that you have copied in every detail.
Basic antenna tools are not expensive. Most simple tools for antenna design is a diode-detector to check if a modification of an antenna increases or decreses signal level at a distance of a meter or two. A such tool is very simple to build and cost almost nothing.
VSWR meter is another relative cheap tool and can easily be built if you increase current mechanical and solder skills with an factor 10 or more. Your current soldering looks way off from what can be accepted and coaxial cable did really look like it was opened with a chain saw.
I am sure that you can do much better.
A good an reliable soldering is just as when you glue two surfaces against each other. A little glue/solder as possible makes the strongest joint and smallest losses. A lot of tin and it becomes a mini-antenna of solder which is as weak as your solder.
As a reference, pictures of some of my junk-antennas are attached so that you can get an idea how they are assembled.
It is antennas that have been used as temporary references when I have done some kind of measurement.
Soldering at these antennas are of good quality, wet and soldered at right temperature. Cables are semi rigid with low losses at 2.4GHz.
Iron wire or copper-traces at PCB or CU-tape is what I often use for these temporary antennas. When using PCB must dielectric constant be taken in account which can be a bit tricky but advantage is that it is very mechanical stable designs.
Also notice that coaxial cables have very clean cuts. I use a standard knife or an exacto and use it as a saw, sawing carefully. Sawing only almost through isolation. Or else will it easily make cuts in wires. Even minor scratches from a knife and such wires and they do then more easily become broken.