Need a CRM today?
Zoho CRM is free for three people and mature; Pipedrive Lite is $14 a seat and very good at one job. We would rather say that than have you wait a quarter for something that is not built.
CENTILIO RELATE / SELL · EMBER
Six things about CRM, whichever one you end up buying.
Relate is not out yet, so this is not a manual for it. It is what we would tell any growing business about customer records: what a CRM actually is, how to choose one, why so many get abandoned, how to build a pipeline that means something, how to forecast honestly, and how to move off the one you already have. None of it is written to sell you Relate, and some of it will tell you not to buy a CRM at all.
6 guidesuseful before Relate shipsno sign-up needed
Product design preview · Relate has not shipped. All features, screens and prices are proposals.
Relate has not shipped yet.
Relate is not available yet. These are proposed screens, features and prices—not a beta or live CRM. All console actions affect fictional data in this page only.
Guides
The six things people ask first.
These guides explain the category and a proposed migration approach. Validate field mapping, privacy and retention requirements before importing real customer data.
Already on Centilio?
Sign, Drive and Account are live. A quote signed in Sign today will be on the customer record when Relate opens.
Want to shape it?
Tell us how your business actually sells. We are still designing, and the awkward requirements are the useful ones.
What a CRM actually is, in plain words
Strip away the category and there is one idea underneath it.
- It is a shared memory. A current record of every customer and every conversation, so the business remembers what a person would forget — and so the answer to "what did we say to them?" does not depend on who is in the room.
- Everything else is built on that. Pipelines, forecasts, sequences and reports are all downstream of the record. When the record is stale, every one of them is fiction, which is why a beautiful pipeline on bad data is worse than no pipeline.
- The unit is the customer, not the deal. A deal ends. A customer does not. If your tool forgets a customer the moment the deal closes, it is a pipeline tracker wearing a CRM's name.
- You need one when memory stops scaling. Not at a headcount. At the first "I thought you were handling that", or the first time two people email the same customer about different things in the same week.
How to choose a CRM for a small business
Six questions that predict whether you will still be using it in a year.
- What does it cost when we are twice this size?. Read the second tier, not the first. Cheap entry plans are doors; the room is where the money is. Multiply the tier you will actually be on by the seats you will actually have, and add any one-time onboarding fee.
- Who has to type, and how much?. Ask a seller to walk you through logging one real conversation. If it is more than a sentence, the record will be stale within two months and everything built on it will be wrong.
- Can the people who only look, look for free?. The owner, the accountant, the delivery lead. If watching costs the same as selling, either you pay for seats that do nothing or those people stay outside the record — and then the record is not the whole truth.
- How do we get our data back out?. Ask before you buy: every field, every note, every timeline entry, on the lowest plan, without a phone call. The answer tells you what the vendor thinks of its own product.
- Does it fit the channels our customers actually use?. Email everywhere; WhatsApp, marketplaces and phone in India. A CRM that cannot see the channel the leads arrive through is guessing.
- What happens in the first fortnight?. Import, field mapping, and somebody checking it. Most CRM projects fail in week two, not month six, and a guided migration is worth more than a feature list.
Why CRMs get abandoned, and what to do about it
It is almost never the software people say it was.
- The cost was in the typing. Every abandoned CRM has the same autopsy: stale records, because updating them was an unpaid second job. Judge a CRM by what it fills in by itself, not by what it can store.
- It was built for the manager, not the seller. If the only person who gets something back from the tool is the person reading the report, the people entering the data will stop. Make sure a seller gets an answer out of it in the first week.
- Nobody agreed what a stage meant. Write the entry criteria for each stage in one sentence each, in your own words, before anyone moves a deal. "Proposal" means the proposal is sent — not that you talked about sending one.
- The forecast was a performance. When people learn the number is theatre, they stop maintaining the data behind it. Weight stages by what you have actually converted, publish the honest number, and let it be lower.
- Start with one pipeline and five fields. Add nothing for a month. Almost every CRM that survives its first year was small in its first month.
How to build a pipeline that means something
Stages are definitions, not decoration.
- Name stages after evidence, not feelings. Each stage should be a thing that visibly happened: a call took place, a proposal was sent, terms were agreed. "Interested" is not evidence.
- Five stages, at most. More stages give an illusion of precision and cost real time. If two stages always change on the same day, they are one stage.
- Write the exit criterion down. One sentence per stage, agreed by everyone who moves deals. This single page prevents more forecast arguments than any tool.
- Give every deal a next step with a date and a person. A deal without one is not a deal, it is a hope. This is the one rule worth being strict about.
- Close the dead ones. A pipeline nobody prunes is a pipeline nobody believes. Losing is information; leaving it open is not.
How to forecast honestly
Two numbers, and the assumptions behind each.
- Stop using the default percentages. 10, 25, 50, 90 is a convention that came with your software. It has never been measured against your business and it may not fit your business.
- Measure your own conversion by stage. Take the last two years. Of the deals that reached "Proposal", what share were actually won? That number is your Proposal weighting, however uncomfortable it is.
- Publish both numbers. The textbook forecast and the honest one, side by side, every month. Watching the gap teaches a team more than either number alone.
- Weight by age as well as stage. A deal that has not moved in six weeks is not worth what a fresh one at the same stage is worth, whatever the stage says.
- Forecast the date too. Most deals do not die, they drift. Track how long each stage actually takes here, and expect the drift rather than being surprised by it every quarter.
How to move off a CRM you already have
A fortnight of care beats six months of a half-migrated database.
- Export everything before you decide anything. Contacts, companies, deals, notes and activity. If you cannot get all five out easily, you have learned something important about your current tool.
- Bring open deals and everything about people. Every contact and company, and every open deal with its history. Closed deals older than two years can stay in the export — you need them findable, not live.
- Map the fields with a person, not a wizard. Field mapping is where migrations quietly break. Do fifty records, look at them properly, then do the rest.
- Run both for two weeks, then stop. A fortnight of overlap catches what the export missed. Longer than that and people use whichever one they prefer, and you have two half-truths.
- Keep the old export as proposed. Cheap, and the one thing you cannot recreate later.
Questions
Quick ones.
What actually is a CRM?
A shared, current record of every customer and every conversation, so the business remembers what a person would forget. Everything else a CRM sells you — pipelines, forecasts, sequences — is built on that one idea, and fails when the record is out of date.
Do I need one at three people?
Probably not, if all three of you are in every conversation. You need one the first time somebody asks "what did we say to them?" and nobody can answer, or the first time two of you email the same customer about different things.
Why do most CRM projects fail?
Because adoption was treated as a training problem when it was a data-entry problem. If keeping the record current costs a seller twenty minutes a day, it will not stay current, whatever the training says.
Is a spreadsheet enough?
For a while, and there is no shame in it. A spreadsheet fails at three things: two people editing it, remembering what was said, and telling you what changed. When those three start to hurt, move.
Next for you
What people add after Relate.
The tools Relate sits between. Check each product’s current stage while Relate is built.
The proposed shared record connects these products as they launch. Follow each product page for its current stage.
Sign live
The quote on the record, signed in minutes
Bill next wave
The signed quote, invoiced without retyping
Drive live
Every file about this customer, filed
Account live
Who may see which customers, set once
Pulse next wave
Outreach whose replies land on the record
Book next wave
The meeting the next step needs
Be told when Relate opens.
Wave 2 is the next build after the currently available tools. Leave your name and you will hear before the announcement — and tell us what your CRM actually has to do while it is still being designed.