You can store your data as json in the databin and import it after reading it from the databin.

In[22]:= bin = CreateDatabin[];

In[23]:= DatabinAdd[
  bin, <|"myjson" -> 
    ExportString[<|"a" -> <|"a1" -> 1, "a2" -> RandomInteger[100]|>|>,
      "RawJSON"]|>];
DatabinAdd[
  bin, <|"myjson" -> 
    ExportString[<|"a" -> <|"a1" -> 2, "a2" -> RandomInteger[100]|>|>,
      "RawJSON"]|>];

In[25]:= data = Values[bin]

Out[25]= <|"myjson" -> {"{
    \"a\":{
        \"a1\":1,
        \"a2\":100
    }
   }", "{
    \"a\":{
        \"a1\":2,
        \"a2\":60
    }
   }"}|>

In[26]:= ImportString[#, "RawJSON"] & /@ data["myjson"]

Out[26]= {<|"a" -> <|"a1" -> 1, "a2" -> 100|>|>, <|"a" -> <|"a1" -> 2,
     "a2" -> 60|>|>}

Neil, You can send json data directly to a databin. Here is an example:

In[1]:= bin = CreateDatabin[]

Out[1]= Databin["....."]

In[2]:= json = 
 ExportString[{"x" -> 11, "y" -> "foo", "z" -> True}, "JSON"]

Out[2]= "{
    \"x\": 11,
    \"y\": \"foo\",
    \"z\": true
}"

In[3]:= URLFetch[
 "https://datadrop.wolframcloud.com/api/v1.0/Add?Bin=" <> bin["UUID"],
  "Body" -> json, "Headers" -> {"Content-Type" -> "application/json"},
  "Method" -> "POST"]

Out[3]= "<|\"Message\" -> \"The data was successfully added.\", \"Bin\
\" -> \"DD56f1e2eac88-e905-4ee5-8b18-7c4bd6072f3c\", \"Data\" -> \
<|\"x\" -> 11, \"y\" -> \"foo\", \"z\" -> True|>, \"Timestamp\" -> \
{2015, 7, 27, 13, 34, 52.135154`8.46970563570117}, \"DataID\" -> \
\"data0bcbcb32-146b-44f5-bcc2-caf713ac9b7d\", \"Information\" -> \
{\"EntryCount\" -> 1, \"LatestTimestamp\" -> 3646992892, \"Size\" -> \
512}|>"

In[4]:= Values[bin]

A couple of things to keep in mind:

• Make sure the "Content-Type" header in your request is set to "application/json"
• Data Drop will use the first level of parameters in the json submission. It will not separate nested