ESD testing, to be meaningful beyond "yep, blowed up
real good!", has to be traceable to a common standard
in the electrical details. There are only a few "models"
(old HBM, CDM, MM and some newer ones I don't know
details, of). Key are source resistance, source capacitance
and source voltage. Depending on how the waveform is
generated / applied you might also care about standard-
compliant risetimes.
A transmission line pulser is fairly easy and low cost to
build, aside from the power supply. 50-100' of 50-ohm
coax, mercury relay, high value charging resistor, low
value series resistor (for rough output Zmatch). This
can let you capture pin current/voltage waveforms at
ESD-realistic currents (everybody talks about voltage,
but current loop is key) if you want to do things like
characterize your ESD protection elements to enable
explicit design (often we are only provided rules of
thumb, by people who know nothing about ESD or
circuit design, and often are not even provided ESD
protection element models that actually -work- (MOS
device breakdown being kind of a forgotten feature
of older compact models, and modeling group being
more interested in forward-active circuit simulation
accuracy than any "abnormal conditions").
Measuring a few-kV waveform is not going to be easy
or cheap, and accuracy will be kind of on you to assess.
But low voltage ICs can be tested current-mode with
no real issues (at least for HBM where the kV and kOhms
make the threat look pretty much like a pulsed current
source).