SoftwareHow-tosPublished

Questions to ask a vendor that the demo will not answer

Demos are built to succeed. These are the questions that find out what happens on a bad Tuesday, and what it costs to leave.

  • 9 min read
  • 1,405 words
Questions to ask a vendor that the demo will not answer — software tech at a print shop
Software9 min readA1printsoft EditorialPhoto: 1981 Digital / Unsplash
On this page5 sections

The short answer

  • Ask for a demo using your artwork, your catalogue and your worst order
  • Get onboarding, training, migration and integration fees in writing before you sign
  • Ask what your data export looks like — request an actual sample file
  • Ask for a customer who has been on the platform for three years, not a new one
  • Ask what they removed from the product in the last two years, and why

A software demo is a performance. That is not an accusation of dishonesty — it is simply what a demo is. It uses clean sample data, a design chosen because it renders well, an order chosen because it flows through without exceptions, and a presenter who has run the same thirty minutes several hundred times. Everything works because everything was arranged to work.

Your shop is not clean sample data. Your shop has a customer whose logo arrived as a photograph of a business card, an order with fourteen sizes across three garment styles, a supplier whose stock feed times out twice a week, and someone on the front counter who needs to find last February's job while a customer waits on the phone. The gap between the demo and that reality is where buyer's remorse comes from, and it is entirely findable in advance if you ask the right questions.

Why do demos mislead even when nobody is lying?

Because they are optimised for the happy path, and the happy path is the part of the software you will spend the least time in. Every product looks capable when the inputs are perfect. The differences between products only become visible when something is wrong — low-resolution artwork, a size going out of stock mid-order, a customer changing their mind after approval, an integration going down.

There is also a selection effect in what gets shown. Demos highlight the features the vendor has invested in most recently, because those are the ones that look impressive. That tells you something about the product's direction and very little about the parts you will touch every day, which are usually older and less frequently updated. The fix is not scepticism. It is specificity.

What should you ask about a bad day?

Ask these, and watch how they answer

  • "Here is a real customer logo at 72dpi. Show me exactly what your designer does with it."
  • "A customer orders 14 different sizes across 3 styles. Walk me through that order screen."
  • "A size goes out of stock after the order is placed. What does the customer see, and what do I see?"
  • "Show me the screen where I find an order from eighteen months ago."
  • "A customer wants to change the artwork after approving the proof. What happens?"
  • "Your supplier feed is down. What does my store show?"

The answers matter less than the manner of answering. A vendor who has thought about failure will show you the screen, because the screen exists and they are not embarrassed by it. A vendor who has not will move to a different topic, offer to follow up by email, or describe the intended behaviour in the abstract without demonstrating it. That difference is the most reliable signal available to you, and it costs nothing to test.

Pay particular attention to the low-resolution artwork question. Every platform in this market claims a design tool, and those tools vary enormously in how they handle imperfect input — which is to say, in how they handle almost all real input. Watching one struggle with a genuine customer file tells you more in two minutes than an hour of feature comparison, and it shows you where your unbilled art time will actually go.

The supplier feed question is the other one worth pressing. A platform that keeps serving a cached catalogue when the feed fails will happily sell a blank that was discontinued months ago. Ask to see what the storefront does, not what it is supposed to do.

Any product looks good on the happy path. You are buying the exception handling.

What does it really cost?

The monthly subscription is rarely the whole number, and the additional items are rarely volunteered. Ask for each of them explicitly and ask for the answer in writing, because a figure given verbally on a sales call has a way of changing by the time the invoice arrives.

Ask aboutWhy it matters
Onboarding / setup feeOften a four-figure one-off, sometimes mandatory
Training, and whether it is includedSome charge per session after the first
Data migrationMoving your customers and past orders is rarely free
Per-store or per-user feesMultiplies fast if you run team stores
Transaction or order feesTurns a fixed cost into a variable one
Integration feesEach supplier or storefront connection may be extra
Contract length and price-rise termsAsk what the last two years of increases were

The per-store line catches people out. If you run seasonal team stores, a per-store fee that looks trivial in the demo becomes a real number across a school year with a dozen programmes. Ask how a store that closes is billed, and whether an archived store still counts.

The last row is the one people forget, and it has caused the most public complaint about platforms in this market. Ask directly: "What price increases have existing customers seen in the last two years?" A vendor with a stable record will tell you a number. A vendor without one will explain their pricing philosophy, which is not an answer.

What happens if you want to leave?

Ask this during the sales process, when you have the most leverage and they have the most incentive to be accommodating. Asking after you have signed is asking a question nobody has a reason to answer well.

The exit questions

  • "If I cancel, what exactly can I export? Customers, orders, artwork files, designs?"
  • "What format? Show me a real sample export file, not a description of one."
  • "How long do I keep access after cancelling?"
  • "Do I own the artwork files my customers uploaded?"
  • "Are the designs my customers built portable, or only viewable inside your system?"

Request the sample export file and actually open it. This is the single most revealing thing you can do during an evaluation. A clean, complete CSV with your orders, line items, customers and file references means a vendor who expects you might leave and is comfortable with that. A partial export that omits artwork, or a "contact support and we will arrange it" answer, tells you the cost of leaving is deliberately high.

The artwork question deserves emphasis. Years of customer-uploaded logos and saved designs represent real accumulated value. If those live only inside the platform in a proprietary form, you are not renting software — you are gradually depositing your business into somebody else's building.

Who should you actually talk to?

Not the reference customer they offer you, or at least not only them. That customer was selected, is probably recent, and is often still inside their first year, which is the honeymoon period for any platform. Ask instead for someone who has been on the system three years or more and runs a shop roughly your size. If the vendor cannot produce one, that is itself informative.

What to ask another shop

  • "What took longest to learn, and what would you have done differently?"
  • "What do you still do outside the system, in a spreadsheet or on paper?"
  • "How long does support take to reply when something is actually broken?"
  • "Has anything you relied on been changed or removed?"
  • "Knowing what you know now, would you choose it again?"

We would rather you asked us all of these than took our word for it — that is what our demo is for, run on a job you bring. The second question is the most useful on that list. Every shop keeps something outside its main system, and what that something is tells you precisely where the product's real boundary sits — far more accurately than any feature list. If three different shops all tell you they still do quoting in a spreadsheet, you have learned something the demo would never have shown you, and you should go in knowing your cost model will stay on paper. The rest of the software archive assumes you are evaluating rather than hoping.

Filed underSoftwareHow-tos
Share
Written byA1printsoft Editorial
175 articles

The print production desk at A1Printsoft. Writes the practical side of running a decorating shop: what things cost, how long they take and what to charge, checked against published trade figures.

Keep reading

More on software

All Software articles

Set up in an afternoon. From what you already do.

Free for 14 days · no card · pick what you make and we preload the methods, products and price tables.