We have again very simple spectrum with signals in the aromatic region.

Let's start by zooming into the aromatic region and integrating the signals.

From the integrals we may deduce that we have benzene with one substituent. Let's create a spin system with three chemical shifts. The spin particle type should be '1*2*1' for the signals with integral of 2.

From the roofing effect of the chemical shift '2' we can deduce that it's coupled strongly with the chemical shift '1'. The integral of the shift '2' also means that it is in para position in respect of the substituent. Also the shift '3' needs to be coupled strongly with the shift '1'. Add also initial guesses for the long-range couplings and simulate.

The shift '2' seems to be quite alright but the another two seem to miss some couplings. Note that we didn't have any other signals in the spectrum so the missing coupling is heteronuclear coupling. From the intensity of splitting patterns we can deduce that the abundance of heteronucleus is 100 % or close to it. Also the spin quantum number needs to be 1/2. With these criteria there are only two NMR-active nuclei: ¹⁹F and ³¹P. In this case the correct nucleus is ¹⁹F. Let's add new chemical shift for it from the 'Chemical Shift' tab:

Choose ¹⁹F-nucleus and set the 'Calculated' value large enough that it doesn't show in the spectrum f.e 100 ppm. This also prevents second order effects which aren't observed in heteronuclear couplings.

Next we need to add the couplings which should be very similar with proton homonuclear couplings:

Now the initial guesses aren't close enough:

Now we can try new feature 'Total Line Shape Fitting (With Broadening)':

'Total Line Shape Fitting (With Broadening)' is used when the initial guesses aren't good enough and/or there are overlapping signals. Most probably the normal 'Total Line Shape Fitting' will converge to correct values. Anyway, after the 'Total Line Shape Fitting (With Broadening)' the spectrum should look like this:

Let's continue with the normal 'Total Line Shape Fitting' and the spectrum should look like this after iterating twice:

If the J-couplings needs to be published the iteration process should be repeated until RMS value converges or almost converges:

The calculated spectrum should now look like a perfect match with the observed one and the correct molecule is fluorobenzene.