A modest result can contain a precise answer to a real need.

Small does not mean insufficient. Something that answers one real need exactly is usually better than a larger effort aimed vaguely at everything, since precision is what the person on the receiving end actually experiences. Modest results get dismissed by their makers before anyone else has had a chance to use them. Fit matters more than scale.

Look at whatever you are about to expand and ask whether the smaller version already solves the problem. Release the modest thing and let use tell you whether more is needed.