Part 3 in this series of tracking the development of a new app for a small business in the business of residential lettings and property management.
Good progress! The easy phase is nearly over – in other words, importing the existing datasets from our previous system and then relating them together. So the app is now (or will very soon be) displaying an impressive series of connections and relationships between the following datasets:
- Property Addresses
- Property Particulars
- Property Utilities
- People
- Log records
- Tenancies
- Inspections
- Snags
This has been a great opportunity to weed out unnecessary data – with my previous system (Filemaker) it was very easy to add fields to an existing dataset – consequently I ‘bolted on’ extra features and functions throughout the last 9 years or so. Whilst that might seem flexible – actually, it created lots of unused data that we don’t need. With Appsheet, in order to add “fields” you need to return to the source table and create another column, then go through a ‘schema regeneration’ process. This is actually good I think – it encourages correct table/data design in the first place. Thankfully, having already managed databases I have a good idea of what we do and don’t need.
Another lesson is breaking down the datasets – previously all our property data was in one table. Now, our property info is spread across 3 tables (addresses, utilities and particulars). This is because not every address is going to be in our management portfolio – eg, we don’t need information on utilities for an address that is a supplier.
The process has also cemented some design of information flow. The app is not going to be public – it’s only for staff. But we still need clients, tenants etc to be able to submit data. So this is achieved by integrations between our website forms and the same table data that the app accesses.
Also, the app is not going to store files – previously, I didn’t really design document management as a separate system from our database – all our records were within the database and retrieved in PDF format on demand. However, this is vulnerable.
With the new app, it will automate appropriate records, documents and PDFs to be sent to clients etc as required – however. It will also save copies to a separate cloud storage area. That way, records and data can be deleted from the app but, for audit trail purposes, documents will still be available in a long-term archive.
Ok… next is building in functionality!