How AutoModel works
How dotAI AutoModel turns an order file into a call list
One upload is enough for the first run. AutoModel reads the order-history export your team already produces and returns a ranked, explained list of dealers who need a field conversation this week. No complex IT project is needed for the pilot, and integration can remove the upload work later.
From raw file to action list
Step 1
Raw order file
Start with CSV, Excel, a billing export, or an ERP extract carrying dealer, date, value, quantity, SKU, region, and owner. No special template is required, and you do not need to change how your team works today.
Step 2
Field mapping
Dealer and owner fields are confirmed before analysis. Missing or inconsistent columns are surfaced early, so you know the list is built on data your team recognises.
Step 3
Local normal
Each dealer is compared with its own history and its local peers, so a broad seasonal dip does not dominate the list and every ranking is relative to that account’s normal.
Step 4
Combined signals
Weak movements are read together, not as one threshold. AutoModel checks 150+ signals across the dealer relationship — buying rhythm, revenue movement, product behaviour, seasonality, field engagement, service friction, payment stress, returns, and delivery.
Step 5
Every dealer arrives with a reason
The output is not a score without context. Each row states who to contact, why the account was flagged, and what the account manager should check first.
Step 6
Action file
You get an editable list for account managers, downloadable as CSV and shareable through WhatsApp or SMS. Reps record what they found; the next list reflects it.
What the first run needs
These are the fields AutoModel confirms before ranking. Partial data is workable: order date, count, quantity, or value are all acceptable, and if your export uses different column names the mapping is handled during the run.
- Dealer code and name
- Order date, count, quantity, or value
- Product SKU or appliance category
- State, sales region, and account owner
More context, sharper ranking
The order file is enough for a first run. When you also upload or connect field visits, loyalty, complaints, payment, returns, or delivery data, AutoModel reads those signals too, so the ranking reflects the whole dealer relationship rather than orders alone.
What you get back
- A ranked list of dealers, ordered by possible loss.
- A plain-language reason for each flag.
- The next field check for the account manager.
- An editable CSV you can assign and download.
See it on your own order file
The fastest way to judge whether the signal is useful is one run on your own data. Uploads, runs, and reports auto-delete after the session.