Short answer
The real cost is the subscription or license plus implementation, internal staff time, backfill, data conversion, integrations, training, ongoing support and the price increases written into renewals. Agencies that win on cost model all of those for ten years before they sign, and negotiate the renewal terms as hard as the first-year price.
When I look at an ERP budget, the number that worries me is almost never the software price on the proposal, because the costs that hurt are the ones nobody wrote down, like the hours your best finance and HR people spend on the project, the temporary staff who cover their day jobs, and the integrations with your permitting, utility billing or timekeeping systems that turn out to be more complicated than anyone said. I would build a ten-year model that puts all of those side by side for every finalist, including what happens at each renewal, what the vendor charges for additional modules or users, and what you will pay for support after go-live.
Subscription pricing deserves special attention, since the first term is often attractive and the escalator in years four through ten is where the real money is, so I push agencies to cap annual increases in the contract and to lock pricing for modules they may add later. I would also include the cost of keeping your legacy system available for reporting and audits for a few years, which is easy to forget and almost always needed.
Done this way, a ten-year model usually changes the conversation, because the cheapest first-year bid is not always the lowest total cost and your council will appreciate seeing the full picture before they approve anything. The Start at Zero Navigator walks through how to build that model, and it is free.