Choosing lettings software is a decision you live with for years, made under time pressure, from demos designed to look identical. This framework is the one we would use ourselves — including the questions that are awkward for vendors. (Yes, we make LintelCRM. The framework works even if it leads you elsewhere.)
Step 1: write the Tuesday list, not the wish list
Skip "requirements workshops". Instead, write down what your team actually did last Tuesday: the certificates chased, the arrears calls, the landlord who wanted a statement re-sent, the viewing feedback typed twice into two systems. Twelve to twenty concrete moments.
That list is your evaluation script. In every demo, ask to see those moments, in the vendor's product, end to end. Feature matrices tell you what exists; the Tuesday list tells you what it costs in clicks.
Step 2: pressure-test the four differentiators
Most platforms overlap on the middle 80%. The differences cluster in four places:
Compliance automation. Not "do you track certificates" (everyone does) but: who notices when one is 30 days from expiry — the software or a human running a report? What happens automatically at 30 days, at 7, on the day? Can a landlord see status without calling?
Money and arrears. Is arrears detection daily and automatic? Is there a ladder with configurable steps, and does it pause itself when payment lands? Can the tenancy produce a court-ready statement of account in one export?
Landlord experience. Statements: generated from the ledger or assembled by hand? Portal: self-serve and branded as your agency, or "we can email PDFs"? At fee-review time, this is your retention argument.
The record model. One record per property/person/tenancy that every module shares — or modules stitched together where the same tenant exists three times? Re-typing is where errors are born; ask to follow one tenancy from applicant to renewal without re-entering anything.
Step 3: interrogate migration before you commit
Migration is where software decisions go to die. Before any contract:
- What exactly imports? Properties and people are easy; tenancy histories, certificates with expiry dates, and document files are the differentiators.
- Who does the mapping — you or them? "We provide a CSV template" means you are doing the migration.
- What does it cost? (LintelCRM's answer: our team imports your export free, inside the trial. Whatever vendor you choose, get their answer in writing.)
- Run the trial on your real data. A demo on sample data evaluates their demo; a trial on your book evaluates the product.
Step 4: read the pricing shape, not just the price
- Per unit, per user, or both? Model your cost at today's book and at 1.5×. Per-user pricing quietly punishes growth in exactly the teams that need everyone in the system.
- What is inside the tier? Seats, storage, emails/SMS, portals — extras have a way of doubling headline prices.
- Setup fees and minimum terms. Both are negotiable exactly once: before you sign.
Step 5: check the exits before you enter
The most revealing question in the process: "If we leave in two years, what do we get and what does it cost?"
You want: a full export (data and documents) in usable formats, no exit fee, no "data held hostage during notice period" clauses. A vendor confident in their product answers this instantly. Monthly terms with free export (our model, for what it's worth) is the strongest signal a vendor has to earn your renewal every month.
The scorecard
Rate each shortlisted platform 1–5 against: Tuesday-list coverage · compliance automation depth · arrears automation · landlord portal quality · single-record model · migration (scope, owner, cost) · pricing at 1.5× growth · exit terms · support responsiveness during the trial (test it deliberately with a real question at 5pm).
Weight them for your business — a compliance-heavy HMO book weights differently from a sales-led agency — and let the scorecard argue with the demo glow.
Frequently asked questions
How long should an evaluation take? Two to four weeks: one to shortlist (three platforms is plenty), two for parallel trials on real data, one to check references and terms. Longer evaluations don't produce better decisions — they produce fatigue purchases.
Should the whole team be involved? The people who will live in it daily should run the Tuesday list themselves during trials. Software chosen by directors alone gets quietly worked around within a quarter.
What is the single biggest red flag? A vendor who cannot show your Tuesday list live and offers a follow-up video instead. If it were smooth, they would show you.