Contract Express conversion project: who designs the questionnaire now?

Ask a firm scoping a Contract Express conversion project what worries them, and the answer is almost always the markup. Will the fields translate? Will the conditional text survive? Will somebody have to rebuild two hundred templates by hand?

Those are fair questions, and the answers are more reassuring than most people expect. The harder question rarely comes up at all: who, at your firm, decides what the questionnaire asks, and in what order?

On Contract Express, few firms have had to. The questionnaire is generated from the template — one of the better ideas in document assembly — so the sequence a lawyer sees follows from how the document was marked up rather than from a separate design step. Move to a platform where the interview is built explicitly and that design becomes real work, with an owner, a timeline and a sign-off.

So there are two jobs here, not one. This post separates them, so you can scope and resource both rather than discovering the second one halfway through.

The two jobs inside a Contract Express conversion project

The two jobs have different owners, different effort profiles and different definitions of done. It is worth naming them separately before anyone puts a number on the project.

• Translation is mechanical. Your templates already work; the job is making them run somewhere else and proving they produce the same paper. That can be settled with a comparison, same answers in and same document out, which makes it bounded, testable and largely automated.
• Data-capture design is a design exercise. You decide what the interview asks, in what order, what is grouped with what, and what stays hidden until it becomes relevant. Success is a judgement call, usually a lawyer’s, rather than a comparison.

Firms leaving HotDocs Classic meet a version of the same split, though there the second job is usually a review of the precedent library itself. Contract Express firms tend to arrive with tidier libraries and a harder question about data capture.

Why the markup half is the easier half

The free Contract Express converter does the mechanical work, and it starts from an advantage. XpressDox does not insist on a single markup convention: alongside its own delimiters it natively recognises those used by HotDocs, GhostFill, and the curly braces used by Contract Express. A template whose fields contain nothing more than a variable name can often run once it has been saved in XpressDox format, with no change to its contents, which in a typical corporate library covers a fair proportion of the notices, short-form agreements, standard letters and schedules.

Beyond that, the converter translates the syntax inside the delimiters as well as the delimiters themselves, either one document at a time with the author approving each translation, or a whole folder in a single pass. A translation dictionary records each original field alongside the equivalent you accepted, so a field it has already seen is offered as the default next time. That is why the second template in a suite is much faster than the first, and why consistent original naming matters more than template count.

The converter is free, as is the equivalent tool for HotDocs Classic. Both handle basic and moderately complex templates; sophisticated suites are work our implementation partners do routinely. For scoping purposes, the point is that this half behaves itself: bounded, testable, and faster as it goes.

Ken Adams made the wider point some years ago. Document assembly, he wrote, “is a straightforward technology”, and the reason it has underdelivered is that using it well “requires that you immerse yourself in the content”. That is the second job, and the one that tends to be left out of the estimate.

💡 Insight
It is worth auditing how consistently your Contract Express variables were named before you convert anything. The translation dictionary rewards consistency and exposes its absence, so an hour spent on a naming audit often removes a day of repetitive review later. Firms with several authors over several years tend to find more variation than they expect — and a conversion is the one moment when renaming costs almost nothing.

The questionnaire is generated, which is why nobody owns it

This is what makes a Contract Express conversion different from a HotDocs one, and it reflects well on the system firms are leaving.

Contract Express builds its questionnaire from the template. Thomson Reuters’ own documentation describes a “unique and patented questionnaire generation engine which automatically creates the questionnaire forms directly from the information contained within the template”. Fields in curly braces carry the variables; optional text is expressed as spans, with the business rule sitting alongside it; a dictionary holds the definitions and the rules. Mark the document up properly and the data capture appears.

That is an elegant design, and it has a consequence that only shows up on the way out. In practice most firms have taken the generated question order as given, because it arrives working. Very few have had to decide what their interview ought to look like, and so very few have anyone who owns that decision.

XpressDox works the other way round. The interview is built explicitly, from commands in the template, and most authors hold those commands in a code template shared across a suite. So you can group questions the way a lawyer actually works, hide sections until they become relevant, and change the sequence without touching the document. That is more control, and more responsibility, and it has to appear somewhere in the plan.

The direction of the trade is worth stating plainly: you are not losing a feature, you are acquiring a decision. Firms that treat it as an opportunity end up with a better interview than the one they had. Firms that treat it as an afterthought reproduce the old question order faithfully, which is the part of the project most likely to leave a firm better off rather than merely relocated.

💡 Insight
It is worth asking a fee earner who uses the document weekly to sit through the new interview once, before anyone calls the template finished. Firms often find their objections are specific, quick to fix, and completely invisible to the person who authored it. Every firm is different, but the questions that irritate a lawyer in month one tend to be the ones that quietly suppress adoption in year two.

What this means for scoping and resourcing

Both jobs belong in the estimate, and the second is the one that resists estimating. A converted template can be checked in minutes against a document the old system produced. An interview cannot be checked against anything, because there is no previous version to compare it with, only a judgement about whether it is better. Work that cannot be verified is work that is easy to defer indefinitely.

In corporate and commercial practice, where Contract Express is strongest, this bites in a particular way. The documents are fewer and more consequential, the conditional logic is denser, and the person whose opinion matters about question sequence is usually the partner who negotiates the deal: comfortably the least available person in the building.

💡 Insight
It is worth naming one person per practice group who decides question sequence, and settling that before the first template is converted rather than after. Firms often find the decision stalls not because nobody can make it but because nobody has been asked to, and an unowned design question tends to resolve itself as “keep what we had”.

Sorting the library by how much data capture it needs

Most conversion triage sorts templates by complexity. For a Contract Express library it is more useful to sort by how much data capture each document actually needs, because that is what predicts the expensive half of the work.

• Low-capture documents. Notices, standard letters, short-form agreements, schedules. A handful of variables, little or no optional text. These convert like-for-like and need almost nothing designed. Usually the largest group by count and the smallest by effort.
• Moderate-capture documents. Templates with real optionality: a dozen or more decisions, several passages that appear only sometimes, a repeating list or two. Expect review rather than a clean pass, and expect a visible return from grouping and sequencing. Modest design effort, good value.
• High-capture suites. The negotiated documents: facility agreements, share purchase agreements, complex commercial contracts — anything whose questionnaire is long enough that lawyers have opinions about it. Few in number, and here the interview is the project rather than a step in it.

Two questions place a template quickly. How many decisions does a lawyer make while producing it, and does anyone currently complain about answering them? A document nobody complains about is a conversion. A document people avoid because the questionnaire is tedious is a redesign, and that is where the visible win is.

Sequencing so the two jobs do not collide

Our consistent recommendation is to keep the jobs as separate stages, with separate sign-offs, in this order.

1. Pick a representative suite — a typical one, including its variations, not the easiest one. A clean template proves the tool works and tells you nothing about your library.

2. Convert it and run it before batching anything. The first pass surfaces whatever is systemic in your markup, and it is far cheaper to learn that on one suite than on two hundred templates.

3. Build a verification set. Take documents Contract Express produced, regenerate them from the converted templates with the same answers, and compare. This is the test a partner will actually ask about — without it, “it works” is an opinion.

4. Convert the remainder, resisting the temptation to improve as you go.

5. Design the interviews as scheduled work, practice group by practice group, with the fee earner in the room.

There are two different clocks here, and that is the whole reason for the order. Your Contract Express licence has a date on it; a better interview does not. Tie the second to the first and the migration ends up waiting on a partner’s view about question order. Keep them apart and the conversion lands when it was supposed to, while the interviews carry on improving afterwards, with lawyers already using them who can tell you exactly what is wrong.

Common mistakes in a Contract Express conversion project

Five patterns account for most of the difficulty, and all five are avoidable.

1. Scoping the markup and inheriting the design. The estimate covers translating templates; the work that turns up also includes deciding what the interview asks. Budget for both, or defer one explicitly.

2. Reproducing the old question order because it is there. It was derived from your document structure, not designed for your lawyers. Reproducing it faithfully preserves a constraint rather than a decision.

3. Batching a whole folder before validating one template. Folder conversion is fast enough to be tempting, and a systemic issue found late is expensive to unwind.

4. Discarding what you know about the questionnaire. The reasoning behind a question order, the traps, the answers people always get wrong — that is institutional knowledge. Capture it while the old system is still running, not after.

5. Treating interview design as polish. It is the part lawyers actually experience, and adoption tracks it more closely than it tracks template accuracy.

Our view: convert the markup, and give the interview an owner

Contract Express is a capable system with a strong following in corporate and commercial practice, and the firms running it built real discipline into their document production. Generating a questionnaire from marked-up text forces a clarity about data capture that free-form templating never demands. That discipline is worth carrying forward deliberately, rather than by copying.

Our position is that a conversion should remove the mechanical barrier and be candid about the rest. It is why both converters are free: the migration cost, rather than any feature comparison, is what keeps firms on a platform they have already outgrown, so removing it is more useful to us than charging for it. What a converter buys you is time, and a working library on Advanced Document Automation — not a redesigned one.

Which is why we would resist the framing that a Contract Express conversion project is a technical exercise. The templates will run. Whether the firm is better off afterwards depends on decisions about question sequence, naming and which workarounds to retire — and those are drafting decisions, made by people who know the documents. The same holds for the language inside the templates: wording that appears in nine places is better held once and governed centrally than maintained nine times. The tooling exists to give your people the time to make those decisions, not to make them on their behalf.

Where to start

Pick three templates: your highest-volume document, your most negotiated suite, and one where people complain about the questionnaire. Convert them, run them, and compare the output against documents Contract Express produced from the same answers. Within a couple of days you will know your real translation ratio, where the design effort concentrates, and whether this is an internal project or a partner engagement.

If you would rather work through it with people who have done this many times, book a discovery call and bring those three templates — particularly the one people complain about. That is usually the conversation that tells us the most.

Frequently asked questions

Is a Contract Express conversion project just a markup translation?

No — that is one of two jobs, and the more mechanical one. The free converter handles basic and moderately complex templates and produces the same documents from the same inputs. The second job is designing the interview, because Contract Express generated your questionnaire from the template while XpressDox builds it explicitly. Most firms need both, scoped and resourced separately.

What happens to our Contract Express questionnaires?

The data behind them carries across: the fields, their names, and the conditions attached to them. The sequence and grouping are rebuilt as an XpressDox interview, usually defined in a code template shared across the suite. It is worth treating that as a design step rather than a translation step — it is the part where a firm can improve on what it had.

Is the Contract Express converter free?

Yes. The Contract Express to XpressDox converter is free, as is the HotDocs Classic converter, and there is no licence charge for either. Both handle basic and moderately complex templates. For more sophisticated suites, our implementation partners provide conversion services.

Will our templates run without any changes?

Some will. XpressDox recognises the curly-brace delimiters Contract Express uses natively, so a template whose fields contain only variable names can often run once it has been saved in XpressDox template format. Templates using conditions, repeats, computations or formatting instructions need their syntax translated, which is what the converter does.

How long does a Contract Express conversion take?

It depends on template complexity and on how consistently the library was authored, because the translation dictionary makes later templates much faster than the first. The reliable approach is to convert one representative suite, measure the translation review time and the interview design time separately, and scale from there rather than working from a template count.