Both options in every pair below are competent. Each comparison names the list size and channel mix it suits instead of declaring a winner.
Both automate LinkedIn from the cloud and both are good at it. They diverge on what they treat as the unit of work: HeyReach treats a campaign as something distributed across a fleet of senders, Expandi treats a sender as something to be run carefully. That difference decides which one belongs in your stack.
The usual comparison of these two is about deliverability and mailbox counts. If email is going to be one half of a cross channel cadence, the deciding question is different: which one can be the system of record for reply state, and which one is just the faster way to get sending.
Both put email and LinkedIn on one campaign timeline, which is the hard part and the reason either belongs on this site. They part company on breadth. Reply.io will also put calls, SMS and manual tasks on that timeline, and the question is whether that width is capability or clutter for the cadence you are actually running.
Apollo puts the data, the sequence and the prospect record in one place, which removes the hardest problem in multichannel: nobody agreeing on who the prospect is. A specialist stack gives you a better email engine and real LinkedIn automation, and hands you that problem back. This is the trade, stated properly.
This gets argued as a preference and it is not one. Which channel opens is settled by dividing your prospect count by the daily ceiling of the scarce channel and comparing the answer to your sequence window. The argument only stays alive because almost nobody publishes the division.