Businesses choose software by comparing features against what they need this quarter. Four years later they are migrating, having discovered that the product suited a company half their size.
These are the questions that would have predicted it.
1. What does it cost at three times my size?
Not today's price. The price at the volume you plan to reach.
Per-user pricing is where this bites. A product at fifteen pounds per user per month is reasonable at five users and becomes a serious line item at forty — and the tier you need at forty users is frequently not the tier you are on at five.
Go to the pricing page and calculate three scenarios: today, double, and triple. Then check what changes at each threshold, because the jump between tiers is often larger than the jump within one.
Also ask about annual increases. Subscription pricing rises, and a supplier who will not indicate their historic pattern is telling you something.
2. Can I get my data out, completely?
The most important question and the one least often asked before signing.
"Can I export?" is insufficient. Ask specifically:
- Can I export everything — records, attachments, history, notes, custom fields?
- In what format, and can another system read it?
- Are relationships preserved, so I can still tell which note belongs to which customer?
- Can I do it myself, or must I request it?
- How long does it take?
The answers determine how free you will be later. A product you cannot leave will eventually charge you what it likes, and the file attachments are usually where the trap is — many products export records happily and leave the documents behind.
Test this in the trial. Export, open the file, and see what you actually got. It takes twenty minutes and it is the highest-value twenty minutes of the evaluation. More on this in avoiding vendor lock-in.
3. Does it connect to what I already use?
Very few businesses run on one system. What matters is whether this one talks to the others.
Check for ready-made connections to your accounting package, your email platform and your existing operational systems. Then check what those connections actually sync — many advertised integrations move a fraction of what you would assume.
If there is no ready-made connection, is there an API, and does it cover what you need? "It has an API" is not sufficient, as covered in how systems talk to each other.
A product that cannot connect becomes an island, and islands are where duplicate data entry lives.
4. What are the limits I have not noticed?
Every product has ceilings and they are rarely on the pricing page.
Records. Storage. Automations. Custom fields. Users. API calls per day. Attachments per record. Reports you can save.
Find them in the documentation and compare against three times your current usage. Businesses routinely discover a cap on automation runs eighteen months in, at exactly the point automation had become useful.
Evaluate software against the business you intend to be, not the one filling in the trial form.
Things that matter less than people think
Feature comparison tables. Every product ticks every box on its own comparison page. What matters is whether the three things you do daily are quick, not whether an obscure feature exists.
The interface at first sight. You will adapt within a fortnight. Do not choose a product for a year on a first impression that lasts two weeks.
The demonstration. Demonstrations are performed with clean data by someone who uses the product every day. Insist on a trial with your own data instead.
Things that matter more than people think
Support quality. Send a moderately difficult question during the trial and see how long a real answer takes. That response time is what you are buying for years.
Company stability. How long have they existed, are they profitable, have they been acquired? Acquisition frequently means price rises and a slower roadmap.
The pace of change. A product that has not changed in three years is stagnant. One that changes weekly disrupts your team's habits. Look at the release notes.
Whether your staff will use it. The best-specified system is worthless if the team quietly keeps using the spreadsheet, as covered in getting staff to actually use a system.
A practical evaluation
- Write down the five things you will do most often.
- Trial with real data, doing exactly those five things.
- Calculate the cost at triple your size.
- Perform a full export and open the result.
- Ask support a hard question and time the reply.
- Find the limits in the documentation.
- Have the person who will use it daily do steps two and five.
A day of this saves a migration project. Migrations cost far more than anyone budgets, as moving off a legacy system sets out.
Our consultancy team helps UK businesses evaluate systems against where they are going rather than where they are. Start a conversation.
