The data is the raw material. What you build with it is the interesting part.
Every day we collect, clean and match resale pricing across nine sources and 122 phone models, down to the exact storage, condition and carrier lock. Some companies want that as an API. Some want a file. The ones we enjoy working with most turn up with an idea we had not considered.
We built it to price our own inventory. Then somebody much larger came asking.
Andover started as a tool for our own phone reselling, because guessing was costing us money and nothing on the market answered the question properly. It stayed that way until one of the largest and most influential platforms in recommerce approached us, wanting the same data for something we had not built it for. They licensed it white-labeled, and it has been feeding decisions at a scale we did not originally design for.
That is the reason this is also a $39 product for a solo flipper. If the biggest operators in the industry find this data worth paying for, the person buying four phones a week off Swappa deserves the same numbers rather than a worse version of them. The sophistication gap in this market is not about who is smarter. It is about who has the data, and that is a thing we can simply fix.
Four ways this tends to start.
None of these are packages with a price list. They are the arrangements that have made sense so far, and they get scoped to what you actually need.
API access
Query the same pricing that runs the platform, on your schedule, matched on exact model, storage, condition and carrier lock.
Custom data feeds
A daily file shaped the way your systems already expect it, with the fields you care about and none of the ones you do not.
White label
Our pricing intelligence inside your product, under your brand, at the volume a large platform needs. It is how this data found its second life.
Teams and exports
Seats for your analysts, exports for your models. Export is an enterprise capability and stays that way.
Priced daily, matched exactly, and auditable back to the listing.
Nine sources, one grain
Marketplaces and direct buyers, every observation matched on model, storage, condition and carrier lock. No blending across tiers to fill a gap.
Cleaned before it counts
Listings more than 25 percent from the peer median are excluded before any figure is calculated, and dealer bids are never pooled with marketplace asking prices.
Traceable, not asserted
Every published number records the source rows it came from, with timestamps. If a figure cannot be reproduced from real observations, we do not publish it.
Now tell us the one we have not thought of.
We built this for phone resellers, and the most interesting conversations we have had were with people who wanted it for something else entirely. A few directions we find plausible, mostly to show the range rather than to sell you one of them:
If your idea is on that list, good, we can move quickly. If it is not, that is the email we would rather get. Bring us the problem, tell us what shape the answer needs to be, and we will tell you honestly whether our data can carry it. Sometimes the answer is no, and you will get that answer fast rather than after a procurement cycle.
Three steps, no procurement theatre.
Email us the idea
What you are trying to do, and what you would need from the data to do it. A paragraph is plenty.
We look at whether it works
Against the real dataset, not a brochure. You get a straight answer about coverage, history depth and the gaps.
Scope something small first
A sample extract or a narrow pilot before anyone signs anything long. If it does not prove out, we both learned it cheaply.
What would you do with a daily map of the resale market?
We are a small team and we read every one of these. Tell us the idea, however half-formed, and we will tell you whether it holds.
Start the conversation