While we can explain the gain in kinetic energy in terms of work = force $\times$ distance, the potential energy interpretation is more general and can be applied where for example the potential energy is chemical and the work energy relationship is not so obvious.
Here is a numeric example to show energy is conserved when taking potential energy into account:
I am using a system where measurements are made relative to an observer at rest with the pericentre of the system and both the object and the Earth are moving, rather than make the approximation that the Earth is not moving.
Quantities relating to the Earth have a E subscript and quantities relating to the block have a B subscript.
Assuming $m_E = 5.97E24\ kg\ $, $m_B = 5 kg\ $, $a_B = 9.807 \ m/s^2\ $, $R= 6.371009 E6\ m \ $, $t = 3 s $
$v_B = a_0 t = 29.421\ m/s$.
Assuming conservation of momentum is valid, then:
$ m_E \ v_E = m_B\ v_B $
where $v_E ,\ v_B $ are the final velocities.
$v_E = v_B \frac{m_B}{m_E} = 14.92814/E24\ m/s$
$a_E = v_E / t = a_B \frac{m_B}{m_E} = 1.64271/E24\ m/s^2$
$d_B = 1/2 \ a_B t^2 = 44.1315 \ m$
$d_E = 1/2 \ \_E t^2 = 7.39221 \ m$
$KE_B = 1/2*m_B*v_B^2 = 432.798\ J$
$PE_B = m_B\ a_B\ d_B = 432.798\ J$
$KE_E = 1/2\ m_E*v_E^2 = 72.4954/E24\ J$
$PE_E = m_E\ a_E\ d_E = 72.4954/E24\ J$
So $\Delta KE_B - \Delta PE_B = 0$
and $\Delta KE_E - \Delta PE_E = 0$
so it would appear energy is conserved.