Does anyone actually track which media package was on the properties that sat versus the ones that moved fast

Four deals is not a sample you can conclude from, and you already know that. The harder problem is that even with a larger dataset, you are trying to isolate a single variable inside a system where price, condition, agent network, list day, and seasonality all interact. Days on market is an output; media is one of many inputs. The attribution problem does not get clean with more deals, it just gets less noisy.

The assumption doing the most work in your framing is that media affects absorption rate directly. It probably affects something upstream: click-through on the listing, showing requests, first impressions at showings. If you are not pulling listing views and saves from the MLS or whatever portal data your agent shares, you are skipping the mechanism and jumping straight to the outcome. That is where I would start. Compare views-per-day in the first week across your deals. If the $800 package is not generating more early traffic, the spending case collapses regardless of what eventually happened to DOM.

The risk you did not name: your LP is asking this because they are forming a view about how you make budget decisions, not just about photography. "I don't have a good answer" on a recurring line item is a governance signal to a capital partner. It is worth having a framework even if the data is thin. Something like: full package on anything listed above $X or in a competitive submarket, stills-only on distressed product where condition is the story. That at least shows the decision is structured.

On the 47-day deal: what did the listing views look like in the first week relative to the nine-day deal? That single comparison, if you can pull it, would tell you more than anything else in your current dataset.

The LP question is the right frame to be working from, because it separates marketing cost from marketing signal. The deal that sat 47 days with full media probably had a pricing or condition problem that better photography made more visible, not less. Three-dimensional tours in particular extend time on market for overpriced properties because they let buyers rule the listing out from their couch before you get a showing to make the case in person.

The related strategy that changes the picture here is absorption rate segmentation. If you back out your four deals by price band and run them against the absorption rate in that band during the week they listed, the media variable starts to isolate. A property in a sub-30-day absorption band that still sits almost always has a price or condition explanation, and that means the $800 package was not the failure and the $220 package was not the win.

What I would track going forward is click-through rate from the listing to the showing request, not days on market, because DOM conflates your pricing decision with your media decision. Most MLS-integrated platforms log this, and your agent should be able to pull it per listing. That gives you something you can actually show an LP as a cost-per-showing number tied to media spend.

Which price band did the 47-day property fall into compared to your three-week comps?