Skip to content
NEWLAW.COM
+61 1300 495 281
Let’s talk

Data migration route

Harvest data to SQLite

Harvest is a cloud accounting and billing system, and SQLite is a store of data kept as files the firm holds. This is not a move to a new working system: the route writes out what Harvest holds and keeps it readable and open to queries.

FromHarvestCloud
  1. Matters: Exported in full
  2. Contacts: Exported in full
  3. Notes: Exported in full
  4. Tasks: Exported in full
  5. Time and expenses: Exported in full
  6. Billing: Exported in full
  7. Documents: Confirmed in the analysis
  8. Custom fields: Confirmed in the analysis
  9. Templates: Confirmed in the analysis
ToSQLiteFiles
  • 6 carried
  • 3 to confirm
What the two products hold. The analysis confirms it field by field.

What moves

What moves from Harvest to SQLite

Each row is a kind of record Harvest holds, whether SQLite has a place for it, and what is planned. On this route six kinds of record carry across and three kinds are confirmed in the analysis.

Records in Harvest and their place in SQLite
RecordIn HarvestIn SQLiteWhat is planned
MattersMatters or cases, with their numbers, types, parties and statusHeld in partProjects under clients stand in for matters; there are three project types.Has a placeA matters table keyed by matter number.Exported in full
ContactsClients, other parties and referrersHeld in partClients and client contacts, kept mainly as invoice recipients.Has a placeContacts and their roles on matters as linked tables.Exported in full
DocumentsFiles stored in or linked from the systemTo be confirmedHas a place in partListed in a table; the files themselves stay in folders beside the database.Confirmed in the analysis
NotesFile notes, matter notes and contact notesHeld in partFree-text notes on each time entry.Has a placeA notes table linked to matters.Exported in full
TasksTo-dos and workflow steps assigned to peopleHeld in partTasks are categories of work for time entries, not to-dos.Has a placeA table of tasks and their status at the final date.Exported in full
Time and expensesTime and expense entries, billed and unbilledHeldTimers and timesheets by person, project and task, with approvals.Has a placeA table of time and expense entries.Exported in full
BillingInvoices, payments and debtor balancesHeldInvoices and estimates, with online payment through Stripe or PayPal.Has a placeInvoice and receipt tables; invoice PDFs kept as files.Exported in full
Custom fieldsFields and record types the firm added itselfTo be confirmedHas a placeColumns added to the table they belonged to.Confirmed in the analysis
TemplatesPrecedents, document templates and automationTo be confirmedNo placePrecedent files are kept in folders, not in the database.Confirmed in the analysis

What the plans mean

Exported in full
SQLite has a place for this kind of record. It is loaded there and checked against Harvest.
Confirmed in the analysis
We could not confirm this from the two products’ own material, so it is checked against your data before the plan is fixed.

The table states what the two products hold and what NewLaw plans for. It is not a promise that every record moves: that is settled against your own data in the analysis.

This kind of move

Not a new system. A complete copy the firm can query.

Writing Harvest out to SQLite is not a migration into a new working system. Nobody will open a matter there on Monday morning. The aim is a complete, faithful copy that can be queried and reported on after the old system is gone. So the rules change. Nothing is cleaned or merged unless the firm asks, because an archive should show the records as they were. Every table and field is carried, including those a new system would have had no place for.

What makes an archive usable is the description that goes with it. The export is delivered with a plain account of each table: what it holds, how it joins to the others, what the codes mean. Record counts and financial totals are agreed to the reports in Harvest as at the export date, so the archive can stand as the record. Documents are exported beside the data with a table linking each file to its matter. Access is restricted, since the archive holds everything the live system did.

Harvest is hosted by its maker and SQLite sits on equipment the firm controls. The data leaves a service that was backed up for the firm and arrives somewhere the firm must back up itself, so backup and access are arranged before the export is copied in. What Harvest will release depends on its export tools, and that is confirmed in the analysis.

Where the data is now

Harvest knows about clients, projects, tasks, people, time entries, expenses, invoices and estimates. It does not know about matters or trust money. A firm leaving it for a practice system is turning projects into matters and carrying across time and invoice history, while the firm’s books stay in the accounting package Harvest was feeding.

  1. Client, then project

    A Harvest client record holds a name, an address and a currency, with contacts added mainly so invoices reach the right person. The detail a legal system wants about a client, such as other parties or a responsible lawyer, was never recorded here and is gathered from elsewhere or added after loading.

  2. Tasks are not to-dos

    In Harvest a task is a category of work, such as drafting or attendance, that a time entry is logged against, and it can carry its own billable rate. Tasks map to activity codes in a practice system. They are not a list of things waiting to be done on a matter.

  3. Three ways to set rates

    A Time & Materials project bills at a project rate, a rate per person or a rate per task. The same lawyer can therefore have different rates on different projects. Historical time is loaded with the rate it was recorded at, and the going-forward rate table is built separately in the new system.

  4. Invoices copied to the ledger

    Where the QuickBooks Online or Xero integration is on, invoices and payments were copied across, but time and expenses went only as invoice lines. The accounting package holds the receivables and Harvest holds the time behind them. Both are read, and the invoice numbers are used to match one to the other.

  5. Archived clients and projects

    Clients and projects that are finished can be archived instead of deleted, and archived records are easy to leave out of an export. The analysis counts active and archived records separately, and the firm chooses whether closed work comes across as closed matters or stays behind in the Harvest account.

More about moving off Harvest

Where it is going

A SQLite archive suits a firm that wants the old system’s records kept, seldom consulted, and free of any server or subscription. Everything structured goes into one file. That simplicity moves the responsibility elsewhere: the file has no security of its own and no administrator watching it, so where it is stored and who can reach it are the real decisions.

  1. One file, kept with the archive

    Matters, contacts, time, invoices and trust transactions become tables in a single file. It is stored with the firm’s closed-matter records and can be copied like any document. A checksum recorded when the file is finalised lets anyone confirm, years later, that the copy they hold has not been altered, and the firm decides whether to keep one.

  2. No logins inside the file

    SQLite has no users or passwords. Anyone who can open the file can read every client’s ledger. Protection comes from where it is kept, such as a restricted folder or an encrypted volume, and that choice is the firm’s. It can suit an archive consulted by two people, and it is said plainly so nobody assumes otherwise.

  3. Built for reading, not sharing

    Only one program can write to the file at a time, and SQLite’s authors point people to a client/server database where many users share data over a network. A finished archive never meets that limit. It is the reason SQLite is used here for records that are closed and not for anything staff still update.

  4. Documents stay outside it

    Documents remain ordinary files in matter folders, and a table in the database lists each one with its matter, title, dates and path. Keeping them as files means they open with a double-click and can be handed to a client. The table is how a file is found by matter number instead of by guessing its name.

  5. Someone has to open it

    SQLite publishes its own command-line program for opening a database file, and other tools read the format too. None of that helps if nobody remembers how. Saved queries for the usual look-ups and a short note on which tool to use can sit in the same folder as the file, and the firm decides what it needs.

More about moving into SQLite

How NewLaw runs it

Four steps, in this order.

This route produces an archive, not a working system, so the steps are about completeness and proof.

  1. Analysis

    We read what Harvest holds, table by table, and agree with you what the copy in SQLite must be able to answer later. Everything Harvest holds is in scope unless you take it out.

  2. Clean-up, if you want it

    An archive should show the records as they were, so nothing is merged or corrected unless you ask. Where the copy is also meant for reporting, duplicate and inconsistent entries are tidied in a working copy and the original values are kept beside them.

  3. Trial export

    The data is written to SQLite as a trial and checked against Harvest: record counts, totals where Harvest holds money, the links between records, and a sample of documents opened from the export. Exceptions are listed for you to decide.

  4. Final export

    After the last day of work in Harvest, the export is run a final time and checked again. There is nothing to go live. Harvest is kept read only until you are satisfied the copy answers what you need, and it is handed over with a written description of its contents.

Fixed price, agreed after the analysis. How NewLaw moves legal data

Start

Start this migration

Tell us what is in Harvest and roughly how much of it. The first step is the analysis, and the price is fixed once it is done.

Step 1 of 5

Step 1 of 5: Start with the route

Moving fromHarvest
Moving toSQLite

The route comes from this page. Four short steps follow. Nothing you send commits your firm to anything.

What happens after you send it

  1. A free first call

    We go through your answers with you and ask about anything they leave open.

  2. An analysis of what Harvest holds

    What is in it, how its records link, and where each piece would sit in SQLite. Anything with no place there comes across as a note or stays behind, and you see which before the move.

  3. A fixed price in writing

    Set after the analysis, so it reflects what is in your data. You have it before any work starts.

Beyond the data

Structure, permissions and handover

SQLite is a place to keep data, not a system staff work in each day. What is agreed is how the data is laid out, who can reach it, and what is handed over with it.

There is nothing to install on a server and nothing to license. The work is in making the file usable and safe to keep: naming it, documenting its tables, storing it where the firm’s records policy says archives go, and checking that a member of staff can open it and find a matter. Whether a second copy is held somewhere separate is the firm’s call.

What is agreed and set up

  • File name, storage location and whether a second copy exists
  • Whether a checksum is recorded for later verification
  • Saved queries for matter, ledger and document look-ups
  • Which tool staff will use to open the file
  • Who may open the folder the file sits in
  • Retention period and destruction date noted beside the file

Questions

Harvest to SQLite: what firms ask

Harvest has projects, not matters. How does that convert?

Each project becomes a matter under the same client, keeping its name and any code as the matter reference. Firms that used one long-running project per client, with all work logged to it, decide during clean-up whether to split it into separate matters or carry it as a single general matter.

Will our old time entries keep their notes?

Time entries are read with their date, person, project, task, hours, billable flag and the notes typed on them. In the new system the notes become the narrative on each entry. Entries already invoiced are linked to their invoice where the destination supports that, and listed as billed where it does not.

We have no trust records in Harvest. Is that a problem?

It is common. Harvest has no trust accounting, so client money has been tracked in the accounting package or a spreadsheet. Those records become the source for opening trust balances in the new system, and each one is agreed to the trust bank statement before it is loaded.

Does the copy in SQLite hold billing records from Harvest?

Yes. Invoice and receipt tables; invoice PDFs kept as files. Billing records are written out with the rest, and counts and totals are agreed to the reports in Harvest as at the export date.

Does everything in Harvest have a place in SQLite?

Most of it does on paper. Documents, custom fields and templates could not be confirmed from the two products’ own material, so those are checked in the analysis. Anything that turns out to have no place in SQLite is kept as a readable archive, and you see which before the move.

What has to be decided before Harvest is written out to SQLite?

Documents, custom fields and templates are confirmed in the analysis. The firm also decides who may open the archive and how long it is kept.

Where to start

Before and after the move.

Choosing the system comes before a migration. Setting it up and putting AI to work come after.

How we do itDiscovery to rollout, step by step

Harvest and SQLite are the names and marks of their owners, used here only to identify the two systems.

Moving from Harvest to SQLite?

Tell us what you are moving. You get a plan and a fixed price after the analysis.

We reply within one business day.