Who we are

About

A small company building a CRM for organisations of two to two hundred people, in the gap between tools that cannot hold a relationship and tools that arrive with a six month project.

Why

The gap this exists for

Almost every organisation between two and two hundred people ends up with the same arrangement, and none of them chose it. A spreadsheet that started as a list and became the database. A shared inbox holding what was actually agreed. A calendar no one can read across. And the most valuable thing the business owns, which is what it knows about the people it works with, living in the memory of whoever happens to remember it.

This is not a discipline failure and it is not fixed by asking people to be more organised. It is what happens when the tools that are easy to start with have no way to hold a relationship, and the tools that can hold a relationship are priced and shaped for organisations ten times the size.

Consonas is for that gap: a real system of record that a small business can be running properly by the end of an afternoon.

And one technical reason

There is a second reason, which is less obvious from a pricing page and matters more over time. Almost every system of this kind keeps every customer's records in one shared database and relies on each query carrying the right condition to separate them.

It is a reasonable way to build software and almost everyone does it. It has exactly one failure mode, and the failure mode is that a query written without that condition returns one customer's records to another, silently, with nothing about the request looking unusual while it happens.

We thought that was worth doing differently, and doing it differently is the kind of decision that has to be made at the beginning or not at all.

How we decide

Six principles you can hold us to

Each of these is visible in the product and checkable on this site rather than being a statement of values.

Describe what exists
Every page on this site describes the product that is running instead of the one we intend to build. The roadmap names what is not built, including the things decided against. Describing intentions as though they were features produces signups that leave within a fortnight.
Separate by architecture
Each organisation gets a database of its own instead of a share of one. It is the most consequential decision in the product, it is the one that cannot be changed later, and it is why a query written wrongly returns the wrong answer to you rather than somebody else's records to a stranger.
Obligations are not features
The audit trail, consent records, subject access tooling and two factor sign in are on every plan including the free one, permanently. An organisation that cannot afford a paid plan does not deserve worse evidence when something goes wrong.
Leaving must work
Export is complete, free, on every plan, at any time, without asking, including after you stop paying. A portability promise you can only test after paying is not a promise.
Refuse rather than guess
Where a mistake is expensive, the product refuses. It will not create an organisation in a region it cannot honour, will not accept a credit note larger than what remains uncredited, and will not let one person approve their own contract.
Measure rather than look
Contrast is computed, touch targets are measured, and claims are checked by tests that run on every build. Several defects on this site were invisible to review and obvious to arithmetic.

A buyer who discovers a limitation on a website has learned something in thirty seconds. A buyer who discovers it after a migration has learned the same thing at enormous cost, and has also learned the website is not reliable.

Which is why every page here names its limitations, and why we accept losing the enquiries that costs.

Being small

What that means for you, in both directions

What it means in your favour

The person answering the address is somebody who can change the product. A request that describes a real problem genuinely moves the order of work, and several things now built arrived that way. There is no sales team, no qualification call and no demonstration to sit through before you can use it.

There are also fewer people with access to anything, which is a real security property instead of a marketing one.

Where an argument needs more room than a page here gives it, it goes on the blog, which is occasional instead of a schedule.

What it means against us

We do not hold SOC 2 or ISO 27001. There is no published penetration test. There is no marketplace, no partner network and no consultant in your city who already knows the product. The list of systems we connect to is short, and the customers page still has no one on it. If your procurement requires a certification, we are not the right supplier yet and would rather tell you now.

There are also fewer people to notice something, which is the other side of the security property above, and the security page treats it as both rather than only the flattering half.

The question worth asking, and the answer

What happens if we stop. It is the right question to ask any supplier of our size and it deserves a straight answer rather than reassurance.

Your data is exportable, completely, free, on every plan, at any time, without asking, in a format that opens without our software. Run one on your first afternoon so you know what it produces, and periodically thereafter so a current copy exists somewhere we do not control. That is the honest answer, it is entirely in your hands, and we would give the same advice about any small supplier including ourselves.

The bill

What a database each actually costs, and who pays which part

The separation described above is the decision we are most confident about and the one with the largest bill attached. This is the bill, itemised, including the parts of it you pay rather than us.

We cannot ask a question of everyone at once

There is no table anywhere holding every customer's records, which is the entire point, and it means there is no table anywhere we could count. To find out how many organisations use a particular part of the product we would have to open every database in turn and ask each one on its own.

So we do not have the dashboard most software companies run on. No cohort chart, no funnel, no aggregate view of what everybody is doing, and no analytics in the product to build one from even if the shape allowed it. That is also the plain reason this site carries no statistics: any number we published about usage would be one we had made up, and once you have made one up you have to keep making them up to stay consistent.

Every change has to happen many times

A change to the shape of the data is not one piece of work applied once. It has to reach every database separately, which means there is a period during which two shapes of the same thing exist and the code has to be correct against both of them.

The compensation is real: when something goes wrong it goes wrong in one place, to one organisation, rather than to everybody in the same second. We would take that trade again. It is still a permanent tax on every change, and it is one of the reasons this product has fewer features than the things it competes with and will go on having fewer.

What you pay: two organisations are two organisations

If you run a group and put each subsidiary in an organisation of its own, nothing in the product joins them. No report across both, no shared list of people, nothing. It is the same wall that keeps a stranger's records away from you, and it cannot tell the difference between a stranger and your own parent company. Each organisation also carries a plan of its own, priced on the pricing page.

A supplier who has never told you to buy something else has either never been the wrong answer, or is not telling you when they are.

Which is why the section below names the buyers we would turn away, and what to buy instead.

Buy something else

Five buyers we would send elsewhere, and roughly where

Not modesty. Each of these has an outcome we can predict, and the predictable version is that you spend a month discovering it and we spend the same month helping you discover it.

Your procurement requires a certification

Buy from someone certified. We would genuinely rather that than watch you run an exception process on our behalf, because an exception granted reluctantly gets revisited at the worst possible moment, usually once the person who granted it has left. Come back if we ever hold one, and the security page will say so before anyone here says it to you in an email.

You need the accounts and the payroll in the same system

We do not build a ledger, and the roadmap says so under decisions made rather than work outstanding. A customer system with a mediocre ledger inside it is worse than two adequate systems that each do one job, and the mediocre one is the one that gets audited.

So buy the accounting package your accountant already argues with, and keep the customer relationship somewhere else. If the entire point of the purchase is one system for everything, buy a suite from someone who sells suites, and accept the configuration project that arrives with it. That is a real answer and it is sometimes the right one.

Your requirement names the United Kingdom

The regions are the European Union and the United States. There is no United Kingdom region.

If your obligation is about adequacy and lawful transfer, the European Union is very likely fine, and the data processing agreement is the place to check rather than this paragraph, with the Information Commissioner's Office behind that for anything the agreement does not settle. If your obligation actually says the data sits in the United Kingdom, in those words, because a regulator or one of your own customer contracts says so, then we cannot meet it. Better said in the first five minutes than in the fifth week.

The plan is to hand it to someone to implement

The product is built so that whoever runs the business can set it up in an afternoon, which is a genuine answer for a small organisation and no answer at all if your buying process assumes a pair of hands you have budgeted for. Buy something with an ecosystem around it. The ecosystem is a real thing you are paying for and it is worth the money on the day you need it.

The spreadsheet is still working

If you are one person, the list is short, and no one else needs to read it, a spreadsheet is the correct tool, and we are not going to pretend otherwise to win a free signup that goes quiet within a fortnight.

If you do go elsewhere, tell us why

A decline is worth more to us than a signup and a great deal more than silence, because it names in one sentence the thing that made the answer no. We have no chart to infer it from, for the reason given further up. Write to hello@consonas.com. No one will try to talk you back around, partly because there is no sales team to do it and partly because a supplier arguing with your reason for leaving is the clearest possible evidence they were not listening to it.

If something happens to us

Five questions that come after the blunt one

The blunt one is answered above: your records leave whenever you decide they should. These are the five that follow it, and none of them is answered with reassurance.

Someone is ill, or away, for a fortnight

The product does not need any of us to be awake. It runs on Cloudflare, where each organisation's database is a Durable Object of its own. Requests are served whether or not anyone is at a desk, and nothing about an ordinary day requires a person here to press anything.

What stops is everything else. Replies get slower, fixes wait, and a question needing someone to read code waits for that somebody. The response targets on the contact page are what we are prepared to commit to rather than what usually happens, and the gap between the two is deliberately where a fortnight of illness goes.

The company is bought

The terms allow us to transfer the agreement to a company acquiring the business. If that happens we will tell you, you may leave, and the acquirer takes the terms exactly as written, which means changing them materially still costs thirty days' notice.

The price goes up

At least thirty days' notice by email before any increase, and it is in the terms rather than in a blog post. The free plan is a plan rather than a trial: it does not expire and does not turn into a paid one through inaction.

The company stops altogether

There is no source code escrow, no undertaking to publish the source if we stop, and no commitment to a wind down period of any particular length. We could write all three sentences this afternoon and none of them would be worth what it appeared to be worth, because a promise about a company's final month is made by the version of that company least able to keep it.

You outgrow us, or decide in month two that this was wrong

Leaving looks the same whether it happens in month two or in year four. Export, then close the organisation from inside the product, which begins a real deletion instead of setting a flag on a row. It asks twice and then holds everything for thirty days, in which you can still export it or simply reopen the organisation, because a closure somebody regrets on the Tuesday should not be an email to us on the Wednesday. We would far rather that happened in week one than in month six. The version of this we find genuinely painful is not the customer who leaves: it is the one who stays quietly for a year without the thing they needed and never mentions it.

Size and money

Asked about the company

Where is the company based?

The United Kingdom. Customer data is held in the European Union or the United States, chosen by each organisation when it is created and fixed from then on. UK organisations should choose the European Union.

How does the company make money?

Paid plans and modules, both priced on the pricing page, and nothing else. No advertising, no data sold or shared, no referral margin paid by anyone we connect to, and no analytics script anywhere to gather the raw material for the first of those. Where a supplier's revenue comes from is worth establishing before you buy rather than afterwards.

Do you use our data to improve the product?

We cannot. There is no view across organisations to learn from, no analytics in the product, and no one here can open your records to answer a question unless you ask for it, for a stated reason, with the access recorded. What shapes the product is correspondence, which is slower and narrower than a chart and considerably more specific.

Why is support on the paid plans when the obligations are on every plan?

Because the two cost differently. An audit trail costs us almost nothing per organisation, and an hour of a person's time costs the same regardless of what you pay, so a support target on the free plan would be a promise we could not hold. The free plan carries the things that are your legal exposure instead of the things that consume our day. The help centre is open to everyone without signing in, and something that is a genuine fault is our problem whatever you are paying.

What would make you change one of the six principles?

Evidence that one of them leaves customers worse off. Not an argument that one of them leaves us poorer, which is a different thing and is the one that will actually be put to us, usually as a proposal to move an obligation up a tier because the modelling looks good. If a principle does change, it will be written on this page with what changed our mind rather than quietly edited out.

Start without talking to anybody

There is no sales team and no demonstration to sit through. Create a free organisation and see whether it suits you.

Three people, a thousand relationships, no card and no time limit.