Eight Legal Spend Management Software Mistakes General Counsel Regret
Most disappointments with spend tools trace back to selection and setup, not invoice review. These are eight mistakes legal teams regret, from messy matter taxonomies to gaps in VAT and currency handling, with a fix for each.
Where spend platforms fail: before the first invoice arrives
When a legal department says its legal spend management software "isn't working," the invoice-review engine is rarely at fault. The real failure usually happened months earlier. Someone decided how matters would be categorized, which firms got a say, and whether the platform needed to handle billing in six currencies and four tax regimes. Those decisions cost little to get right during selection. They cost a lot to undo after go-live.
General counsel and legal operations leads tend to regret eight mistakes in particular. We've grouped them into four pairs. Each comes with a fix you can apply during evaluation or, if you've already signed, during a configuration reset.
Mistakes 1 and 2: Analytics on top of a messy matter taxonomy
Mistake 1: Buying dashboards before defining what a matter is
Vendors demo legal spend analytics on clean sample data: spend by practice area, business unit and firm, trending neatly by quarter. Your data will only look like that if you settle a few things in advance. What separates "employment litigation" from "employment advisory"? Who owns each category? How is a matter that touches three business units allocated?
Without a taxonomy, every intake form becomes a judgment call. Within a year, a catch-all bucket like "Commercial" holds a large share of matters because it's the safest default, and your benchmark reports tell you nothing.
Do this instead: Before signing, draft a taxonomy with no more than two levels, roughly 8 to 15 top-level categories, and a written definition for each one. Then test it. Have three lawyers classify the same 30 recent matters independently. If they disagree on more than about one in five, rework the categories before any software sees them.
Mistake 2: Migrating historical data as-is
Importing five years of spreadsheet history looks like a quick win. What it actually imports is five years of inconsistent coding, and your trend lines are unreliable from day one.
Do this instead: Migrate 18 to 24 months of history remapped to the new taxonomy, and archive the rest as read-only. Treat remapping as its own workstream with its own owner and budget. It often takes longer than the technical migration.
Mistakes 3 and 4: Ignoring panel firms and unenforceable guidelines
Mistake 3: Assuming outside counsel will simply adopt the platform
Your panel firms submit the invoices, often in LEDES format, through a portal they may use for no other client. Large international firms usually handle this routinely. Boutiques, local counsel in markets where e-billing is uncommon, barristers' chambers and sole practitioners often can't produce LEDES files at all. Some platforms also charge law firms submission or subscription fees, and smaller firms resist them.
Do this instead: During selection, ask your five highest-spend firms and three of your smallest regular firms to review the submission workflow. Put three questions to each vendor: do firms pay anything, can PDF invoices be ingested and converted into line items, and who handles firm onboarding? Then roll out in phases, starting with your top firms by spend and following with the long tail.
Mistake 4: Writing outside counsel guidelines the system can't enforce
Many departments load a 20-page set of outside counsel guidelines into the platform and expect automated enforcement. The system can only check rules that are expressed as data: rate caps by timekeeper level, banned task codes, block-billing detection, travel limits. It can't enforce "use judgment on staffing."
Do this instead: Split the guidelines into machine-checkable rules and reviewer guidance. Map each checkable rule to a named configuration setting. During the demo, submit a test invoice that breaks those rules and confirm the platform flags every violation.
Mistakes 5 and 6: Currency, tax and fee structures treated as edge cases
Mistake 5: Missing multi-currency and VAT/GST handling
If your team has matters in more than one country, multi-currency legal billing is a core requirement. A common failure looks like this: the platform converts everything to a reporting currency but can't show which exchange rate it used, on which date, or how tax was treated. Finance ends up rebuilding the numbers in spreadsheets, and the spend reports lose credibility.
Indirect tax on legal services varies a great deal by jurisdiction, and the software has to cope with that. A few illustrative scenarios, which are general patterns and not tax advice:
| Scenario | Typical tax treatment | What the software must support |
|---|---|---|
| EU firm billing a business client in another EU member state | Usually reverse charge, so the invoice shows no VAT | Zero-VAT lines with a reverse-charge flag and both parties' VAT IDs captured |
| UK firm billing a UK business client | VAT generally charged on fees at the standard rate | Line-level VAT validation against the expected rate |
| Individual advocate or firm of advocates billing a business entity in India | GST generally paid by the recipient under reverse charge | Recording a tax liability that isn't in the invoice total |
| Firms in jurisdictions such as the UAE, Saudi Arabia, Singapore or Australia | VAT or GST at differing local rates. Some Gulf states don't levy VAT at all | Configurable rates per jurisdiction, plus tax registration numbers |
Exchange-rate policy matters just as much. Here's a hypothetical: a GBP invoice is dated in March, approved in April and paid in May. It can produce three different USD figures depending on which date's rate you apply. Neither answer is wrong, but your platform and your general ledger have to use the same one.
Do this instead: Have finance and your tax advisers define the exchange-rate policy (invoice date, approval date or month-end rate) and list the tax scenarios you actually encounter. Then give each shortlisted vendor three real, anonymized invoices to process live in front of you. Confirm the treatment of specific transactions with a qualified tax professional in each jurisdiction.
Mistake 6: Configuring everything as hourly
If your strategy includes alternative fee arrangements (fixed fees per phase, capped fees, portfolio retainers or success components), the platform has to track them natively. If it can't, AFA matters show up as anomalies, and you lose the ability to compare their cost with hourly equivalents. That comparison is usually the whole case for using AFAs.
Do this instead: Ask for a demo of two scenarios: a capped-fee matter that overruns, and a phased fixed-fee matter billed in milestones. Make sure the demo covers how accruals are reported for both.
Mistakes 7 and 8: Under-scoped integrations and the wrong pricing metric
Mistake 7: Treating ERP and matter-management integration as phase two
Approved invoices have to reach accounts payable, and matters have to line up with whatever intake system your lawyers already use. If integration is deferred, someone ends up rekeying data with no end date. Common targets include SAP, Oracle, Microsoft Dynamics, NetSuite and in-house matter tools. Phase two has a habit of never being funded once the launch budget is spent.
Do this instead: Include the AP integration in the initial statement of work, with written acceptance criteria and a fixed price.
Mistake 8: Comparing quotes on license fee alone
Pricing models vary. Vendors may charge per user, by annual spend under management, or by invoice volume, and some offset part of the cost through law firm fees. The license fee is often less than half of the first-year cost. The table below shows indicative 2026 market ranges. They vary widely by contract, region and scope, and local-currency pricing can differ from USD list prices.
| Cost item | Mid-size team (under $10M external spend) | Large team ($50M+ external spend) |
|---|---|---|
| Annual subscription | Roughly $25,000 to $90,000 | Roughly $150,000 to $500,000 or more |
| Implementation and configuration | Roughly $15,000 to $60,000 | Roughly $75,000 to $250,000 |
| ERP integration | Roughly $10,000 to $40,000 | Roughly $40,000 to $150,000 |
| Internal admin time | About 0.25 to 0.5 FTE | About 1 to 3 FTE |
Do this instead: Build a three-year total cost model. Include expected growth in external spend if pricing scales with volume, and get local-currency quotes from each vendor before you compare them.
A seven-point pre-signature check
Here's a hypothetical. A 25-lawyer in-house team has about $18M in annual external spend across the UK, Germany, India and Singapore, and it's down to two vendors, say Brightflag and Onit, or LawVu and a Wolters Kluwer ELM Solutions product. Before it signs either contract, it should be able to answer yes to each of these:
- Our matter taxonomy is written down and has been tested on 30 real matters.
- At least eight panel firms, including small ones, have reviewed the submission workflow.
- Every checkable guideline rule maps to a named configuration setting.
- The vendor processed our sample invoices in at least three currencies with the tax treatment our advisers expect.
- One fixed-fee matter and one capped matter were demonstrated end to end.
- AP integration is in the initial scope at a fixed price.
- We have a three-year cost model in our reporting currency.
Items 1, 4 and 6 are non-negotiable, because they're the hardest to fix after go-live. If any of those three is missing, or you can't tick at least five of the seven, push signature back by a month and close the gaps. A short delay costs far less than a reimplementation. Start this week with the taxonomy, because every other item depends on it.
Quick answers
How long does a legal spend management software implementation usually take?
Timelines vary with scope. A mid-size department with a few dozen panel firms often goes live in roughly three to six months. Multinational rollouts with ERP integration and several tax jurisdictions commonly take six to twelve months. What usually sets the pace is taxonomy design and firm onboarding, not software configuration.
Should we require LEDES invoices from every outside firm?
Require LEDES from high-volume firms that already support it. Don't make it a hard requirement for small firms or local counsel in markets where it's uncommon. Pick a platform that can ingest PDF invoices and convert them into structured line items. Otherwise the long tail of firms stays outside the system, and so does a share of your spend.
Can spend software decide the correct VAT or GST treatment for us?
No. The software can check invoices against rules you configure and flag the ones that don't match, but it can't determine your tax position. Have tax advisers in each relevant jurisdiction define the rules, then configure the platform and test it against real invoices before go-live.
Which pricing model is best for legal spend software?
No model is cheapest across the board. Per-user pricing suits teams with a small number of reviewers and high spend. Pricing based on spend under management can get expensive as your external spend grows. Models that charge law firms lower your direct cost but can slow firm adoption. Compare models on a three-year total cost basis using your own projected volumes.
Written by the Juris Tdesk Pro editorial team. Prices, features and rules change often - confirm details with the vendor or a licensed professional before you decide.