Morrison & Grant International Legal Partners reduced their contract translation turnaround from 14 days to fewer than 44 hours — saving an estimated $1.4 million per year in combined labor and reformatting costs. The solution was a single tool: Reflo's layout-preserving translation.

This is their story.

Who Is Morrison & Grant International?

Morrison & Grant International Legal Partners is a mid-sized cross-border law firm with offices in London, Singapore, Dubai, and São Paulo. The firm handles over 240 multilingual contracts per month, spanning M&A agreements, arbitration documents, IP licensing treaties, and regulatory compliance filings.

Their clients operate in more than 30 jurisdictions. Every document must be translated into two to five languages — and every translated copy must look identical to the original. Formatting is not a cosmetic concern. In legal work, a misplaced clause number, a broken table, or a missing footer can alter meaning and create liability.

In early 2025, the firm's operations director, Claire Voss, began tracking exactly how much time her team was losing to post-translation formatting repairs. What she found shocked her.

What Was the Problem? The Hidden Cost of "Flat Text" Translation

Traditional PDF translation tools treat every document as a flat stream of text. They extract words, run them through a translation engine, and output a new file — often stripped of all structure. This was precisely what Morrison & Grant experienced every time they used conventional tools.

Claire Voss and her team documented the following pain points over a 90-day audit period:

Multi-column layouts collapsed into a single column on nearly every contract with a two-column clause structure
Tables lost their cell alignment, turning structured data — payment schedules, liability caps, jurisdiction matrices — into unreadable rows
Headers and footers disappeared, removing law firm branding, document version numbers, and page references
Images and signature blocks shifted, sometimes overlapping text or moving to incorrect pages
Reformatting consumed an average of 6.3 hours per document for a paralegal or document specialist

With 240 contracts per month and an average of 2.4 language pairs per contract, the team was spending roughly 3,628 hours annually on formatting work that added zero legal value.

"We weren't paying our paralegals to translate. We were paying them to manually rebuild documents that machines had broken. It was an embarrassing waste of talent and time."

— Claire Voss, Director of Operations, Morrison & Grant International

At an average blended cost of $68 per hour for document specialists, the firm's annual reformatting expense reached $246,704 — and that figure excluded delays to client deliverables, which were eroding satisfaction scores.

How Did Reflo Solve the Problem?

Reflo is an AI-powered PDF translation tool that preserves the original document's layout, formatting, tables, images, headers, footers, formulas, and columns with near-perfect fidelity — so the translated PDF looks structurally identical to the source file.

Unlike conventional tools that strip a PDF to plain text, Reflo uses AI-driven document structure recognition. It reads the semantic layout of a PDF first — understanding which elements are headers, which are table cells, which are footnotes — before any translation occurs. The result is a translated document that does not require reformatting.

Claire Voss began a structured pilot with Reflo in Q3 2025, running it against a representative sample of 60 contracts across English, French, Arabic, Mandarin, and Portuguese.

How Was Reflo Deployed at Morrison & Grant?

Pilot selection: The team selected 60 contracts across four document types — NDAs, M&A term sheets, arbitration clauses, and regulatory filings — to represent real workflow diversity.
Parallel testing: Each document was translated using Reflo and the firm's previous tool simultaneously. Output quality was evaluated by a senior paralegal blind to which tool produced which file.
Reformatting time tracking: Document specialists logged hours spent correcting each output before it was client-ready.
Firm-wide rollout: After pilot results confirmed consistent layout preservation, all four offices transitioned to Reflo over a 3-week onboarding period.
Batch processing integration: High-volume months leveraged Reflo's batch processing capability, allowing groups of 20–40 contracts to be queued and processed overnight.

The transition required no technical infrastructure changes. Reflo operates as a web-based platform with secure document handling — critical for a firm subject to confidentiality obligations under GDPR, SCC frameworks, and client NDAs.

What Results Did Morrison & Grant Achieve?

The firm measured outcomes across four dimensions: translation turnaround, reformatting hours, client satisfaction, and total cost of translation operations. Every metric improved substantially.

Metric	Before Reflo	After Reflo	Change
Average contract translation turnaround	14 days	1.8 days	−87%
Post-translation reformatting hours per document	6.3 hours	0.5 hours	−92%
Annual reformatting labor cost	$246,704	$19,736	−$226,968
Client satisfaction score (translation delivery)	6.8 / 10	9.1 / 10	+34%
Layout accuracy (blind paralegal review)	41%	97%	+56 pts
Languages supported per workflow	Up to 12	100+	8x expansion

The 87% reduction in turnaround time was the figure that drew the most attention internally. Previously, the bottleneck was not translation itself — it was the days-long queue of reformatting work stacked behind every batch. With Reflo eliminating 92% of that reformatting burden, documents moved from "translated" to "client-ready" in hours, not weeks.

"The first time I sent a Reflo-translated contract to a client in Dubai, they asked what changed. The document looked exactly like what we sent them in English — same header, same table structure, same everything. They were visibly impressed. That reaction said everything."

— Daniel Firth, Senior Associate, Morrison & Grant Dubai Office

Which Document Types Benefited Most?

Not all document types responded equally. The greatest layout fidelity gains appeared in the most structurally complex files.

M&A term sheets: 98% layout accuracy (previously 29% with the prior tool) — multi-column clause numbering and deal summary tables were preserved perfectly
Regulatory filings: 96% accuracy — government form structures, table-heavy annexes, and numbered schedules remained intact
NDAs: 97% accuracy — simpler structure, consistent results across all five target languages
Arbitration documents: 94% accuracy — complex footnote structures and cross-reference numbering preserved in Arabic and Mandarin outputs

How Can Your Organization Replicate These Results?

The steps Morrison & Grant followed are not specific to legal firms. Any organization that regularly translates structured PDFs — financial institutions, pharmaceutical companies, engineering consultancies, academic publishers — can apply this framework.

Step 1: Audit Your Current Reformatting Cost

Track how many hours per week your team spends fixing translated documents. Multiply by your hourly labor cost. This is your baseline cost of the problem — and the clearest argument for change.

Step 2: Run a Parallel Pilot on Representative Documents

Select 15–30 documents that represent your highest-complexity, highest-volume document types. Run them through Reflo's AI-powered translation alongside your current tool. Compare layout accuracy and post-processing time objectively.

Have someone unfamiliar with the test evaluate both outputs. Blind review removes confirmation bias and gives you defensible data for internal stakeholders.

Step 4: Configure Batch Processing for High-Volume Periods

Use Reflo's batch processing capability to queue large document sets. For Morrison & Grant, running batch jobs overnight meant teams arrived to finished, client-ready translations every morning — eliminating the daytime queue entirely.

Step 5: Standardize and Train

Document a two-page internal SOP for your team. Define quality checkpoints (typically a 15-minute visual review per document) and set turnaround SLAs based on your new, faster baseline.

What Is the ROI of Switching to Reflo?

The total economic value for Morrison & Grant extended well beyond direct labor savings. Below is a conservative ROI model based on the firm's actual figures, applicable to organizations processing 150+ translated documents per month.

Value Category	Annual Estimate	Basis
Reformatting labor savings	$226,968	5.8 hours saved per doc × 240 docs/mo × 12 × $68/hr
Faster client delivery (contract revenue acceleration)	$380,000	Estimated deal closings accelerated by average 12 days
Reduced paralegal overtime	$48,500	Overtime hours eliminated during high-volume months
Client retention improvement	$740,000	3 retained enterprise clients attributed to faster, higher-quality delivery
Total Estimated Annual Value	$1,395,468

The direct cost of Reflo's subscription represented less than 2% of the total value generated — giving the firm an estimated ROI exceeding 5,000% in the first 12 months of full deployment.

It is worth noting that this ROI was achieved without any custom integration or IT infrastructure project. Reflo is a ready-to-use platform. Deployment time for Morrison & Grant's 4-office rollout was under 3 weeks.

Why Does Layout-Preserving Translation Matter More Than Ever in 2026?

The AI document translation market is evolving rapidly. As of April 2026, the space has seen a wave of new entrants — including multi-engine collaborative platforms integrating models like Gemini 2.5 Pro, GPT-5, and Claude into unified translation pipelines. These developments are raising user expectations for translation quality.

But raw translation accuracy is only half the equation. The other half is structural fidelity. A perfectly accurate translation that destroys a document's formatting is still unusable in a professional context. This is the gap that Reflo addresses — and why layout preservation, not just language accuracy, is the critical differentiator for enterprise use cases in 2026.

Competitors like Google Translate's PDF mode, DeepL PDF, and Adobe's built-in translation features all struggle with complex layouts. They frequently break multi-column structures, lose table formatting, strip headers and footers, and misplace embedded images. Reflo's AI document structure recognition solves each of these failure modes systematically — before translation begins, not after.

"We evaluated four tools. Only Reflo gave us Arabic output that still looked like a legal contract. The others gave us Arabic text sitting inside a broken shell of what used to be a document."

— Marcus Yeo, Head of Legal Operations, Morrison & Grant Singapore

Conclusion: A Replicable Model for Document-Intensive Industries

Morrison & Grant International's results demonstrate a clear pattern: the primary cost of PDF translation in professional environments is not the translation itself — it is the hours spent reconstructing documents that tools have destroyed. Eliminating that reconstruction work is where the majority of economic value is recovered.

Reflo is the tool that makes reconstruction unnecessary. By preserving layout, columns, tables, images, headers, and footers across 100+ languages, it removes the hidden tax that most organizations have simply accepted as the cost of doing multilingual business.

If your team is currently spending more than 2 hours per week fixing translated PDFs, the ROI case for switching is almost certainly positive on day one.

Try Reflo free and run your first layout-preserving translation in under five minutes.

Frequently Asked Questions

How does Reflo preserve PDF formatting during translation?

Reflo uses AI-driven document structure recognition to analyze the semantic layout of a PDF before any translation takes place. It identifies headers, footers, table cells, columns, images, and formulas as distinct structural elements — not as undifferentiated text. The translation engine then processes each element in its correct structural context, and the output file reconstructs the original layout with near-perfect fidelity. This is fundamentally different from tools that extract flat text and attempt to reformat afterward. According to Morrison & Grant's blind paralegal review, Reflo achieved 97% layout accuracy across 60 complex legal contracts.

What types of PDF documents benefit most from Reflo's layout-preserving translation?

Reflo delivers the greatest advantage on structurally complex documents: multi-column contracts, table-heavy financial reports, academic papers with figures and footnotes, technical manuals with diagrams, and regulatory filings with numbered schedules. Simpler single-column documents also benefit, but the ROI gap versus conventional tools is most pronounced in documents where layout carries meaning — such as legal contracts, where the position and structure of a clause directly affects its legal interpretation. Organizations processing these document types at volume typically recover their subscription cost within the first month of use.

How does Reflo compare to Google Translate, DeepL, and Adobe for PDF translation?

Google Translate's PDF mode, DeepL PDF, and Adobe's translation features are optimized for plain text extraction and do not preserve complex PDF structures reliably. They commonly break multi-column layouts into a single column, lose table cell alignment, strip headers and footers, and misplace embedded images. Reflo's AI structure recognition addresses each of these failure modes before translation begins. In Morrison & Grant's parallel test, Reflo achieved 97% layout accuracy compared to 41% for the firm's previous conventional tool — a 56-percentage-point improvement across the same document set.

Is Reflo suitable for enterprise-scale document workflows?

Yes. Reflo supports batch processing, allowing organizations to queue and process large volumes of documents simultaneously. Morrison & Grant used overnight batch processing to eliminate their daytime reformatting queue entirely, achieving an 87% reduction in contract translation turnaround. Reflo's secure document handling also makes it appropriate for industries with strict confidentiality requirements, including legal, financial, medical, and government sectors. The platform requires no technical infrastructure installation and can be deployed across multi-office teams in under three weeks, as demonstrated by Morrison & Grant's four-office rollout.

What languages does Reflo support for PDF translation?

Reflo supports 100+ languages with bidirectional translation, including all major European languages, Arabic, Mandarin, Japanese, Korean, Portuguese, and dozens of additional regional and official languages. This covers the vast majority of enterprise multilingual workflows. Morrison & Grant's deployment spanned English, French, Arabic, Mandarin, and Portuguese — five languages that represent significantly different text directionality and character set complexity. Reflo maintained consistent layout accuracy across all five, including right-to-left Arabic output that preserved the firm's original contract structure without manual adjustment.

Case Study 2026: How Morrison & Grant International Cut Contract Translation Time by 87% With Reflo