Everything can be both tech and not-tech if you want to argue semantics (ie. If you use a computer to do your job, is it tech?). That's not useful.
But I find the zero marginal cost qualification useful. These companies are fundamentally different than companies that are further out on the marginal cost spectrum.
Take for example CME, NYSE, Nasdaq, etc. Pure zero marginal cost. Virtually unassailable positions for the marketplace niches they dominate.
Then compare that to JPM, Morgan Stanley, Goldman (farther down the spectrum). While they do offer software products in trading, retail, etc...they also offer lots of high touch, high human cost investment banking services. Not the same thing.
It doesn't really matter if you IPO with JPM or Goldman. The bank will make lots of fees but it will be proportional to their man hours worked. No obvious winner.
By this definition, you could say they do some tech, some old school professional services.
As a "tech worker," I find this useful when examining which employers to work for. It determines how much leverage I can have in affecting my employers business. Far and away, the best jobs I've had are at companies on the extreme end of the zero marginal cost spectrum (pure tech, by this definition).
To use banking as an example, you can go from zero to $1B quite quickly in the software (zero marginal "tech") side of the business. Look at SoFi or Plaid.
Meanwhile, on the old school professional services side, imagine how hard it would be to get to $1B there.
