This demonstrates typical pitfalls when you make a datamodel in Go and expect it work smoothly with mgo or mongo-driver, but it does not.

data.go

package data
import(
  "gopkg.in/mgo.v2/bson"
)
type User struct {
  ID bson.ObjectId `bson:"_id"`
  Name string `bson:"name"`
  Email string `bson:"email"`
}
// for seeding purposes, testing environmen
func seed_data(){
  users := []User{
    User{Name:"JohnDoe", Email:"jdoe@somedomain.com"},
    User{Name:"ParryJean", Email:"parryJ@somedomain.com"},
  }
  // get your connection up here 
  var coll *mgo.Collection 
  for _, u := range users {
    err :=coll.Insert(u) 
    if err !=nil{
      fmt.Errorf("Error inserting user into the database")
    }
  }
}

explanation.md

The above code snippet is actually vanilla and for a cursory look everything seems to be fine. Only when you run the code you realise the error is related to the USer.Id field for being empty. You, just like me when I did this for the first time were expecting the mgo framework to generate the ID if empty. For that to happen you are missing a minute detail

type User struct {
  ID bson.ObjectId `bson:"_id, omitempty"`
  Name string `bson:"name"`
  Email string `bson:"email"`
}

Without specifying omitempty you are essentially indicating to mgo that you would be generating the objectId before insertion. Hence mgo steps aside to handover control. When you dont supply the ID (that you were committing to) Mongo throws back. Contrast to that when omitempty mgo understands its empty and you no longer intend to supply a new ID. Mgo then auto generats the same.