CRM automation and sales pipeline setup

What this actually solves

You know that leads are arriving and you know some of them are being lost. What you cannot say is where, or how many. They come in through a website form, a phone call, a Facebook ad, a WhatsApp message and someone walking in, and they end up in five different places — an inbox, a phone, a notebook, a spreadsheet, and one person's memory. Nothing is deliberately dropped. It is just that no single surface holds all of them, so nothing can be counted, and what cannot be counted cannot be fixed.

The second problem is quieter and usually the real reason someone buys. The owner wants to see the state of the business without asking anyone. How many deals are live, which stage each is at, which ones have gone quiet. Right now that information exists only by interrogating people, and asking makes them defensive, and their answers are optimistic. A pipeline that is accurate at a glance removes a daily source of friction from the top of the company. People pay for that calm, and they are right to.

There is a third one that shows up as soon as you hire. Every new salesperson has to be taught the entire process verbally, because it lives in the heads of the people already there. A defined pipeline with defined stages is the difference between onboarding a rep in a week and onboarding them over a quarter.

What we do, and in what order

1. Extract the real sales process. One question does most of the work — what happens between a stranger first appearing and money arriving. We turn the answer into five to seven stages, no more, and every stage is named after work that has been done, not after how the deal feels. Site visit scheduled. Quote sent. Contract sent. Two to four days.
2. Choose the CRM and define the fields. Fields split into mandatory and optional, with mandatory capped at eight, and each one labelled with who fills it in and when. Every field beyond that cap is a field a busy rep will skip, and a half-filled database reports nothing. One day.
3. Connect every lead source. Website forms, Facebook lead ads, WhatsApp, inbound calls, walk-ins. This is the step that turns the CRM from a list someone maintains into a system that catches things by itself. Two to four days.
4. Build the automations. Instant acknowledgement to a new lead, assignment rules, tasks that fire on stage change, a follow-up sequence after a quote goes out, an alert when a deal has been untouched too long, and a defined path for a lost deal so the reason is captured rather than lost. Three to six days.
5. Import and clean the old data. Deduplication, name normalisation, date formats, splitting mixed columns. Budget honestly for this — it is the least interesting and most underestimated part of the project. One day to a week, depending on the state of the file.
6. Build the reporting. A pipeline dashboard for the owner, conversion by stage so you can see which step actually leaks, and activity per rep. Three reports that answer three specific questions, rather than a wall of charts nobody reads. One to two days.
7. Train the team and check adoption. A working session with the people who have to use it, then a review after real use, because the places reps route around the system only become visible once they are under pressure. One day, plus a follow-up review.

Tools we typically use: HubSpot or Pipedrive for the CRM, Make or n8n for the automation layer, Airtable where a full CRM would be overkill, Looker Studio for the owner dashboard, and Google Sheets as the staging area for import cleaning.

What moves the price

Team size. Two salespeople and fifteen salespeople are different projects even with an identical pipeline. Fifteen means permissions, territory or round-robin assignment rules, a real training programme, and reporting that has to hold up when someone disputes it.

The state of your existing data. A clean export imports in an afternoon. A spreadsheet that has been edited by six people over four years can absorb a week before a single automation is built, and that week is real work regardless of how it looks on a quote.

Setup versus adoption. Building the system and being responsible for the team actually using it three months later are two different commitments. The second costs around twice as much, takes longer, and is usually the one that determines whether the money was well spent.

What usually goes wrong

Stages named after optimism instead of work. Hot, Warm, Cold. They feel natural and they are useless, because every rep decides privately what counts as hot, and two people never mean the same thing. The pipeline then reports sentiment rather than fact, and the owner is looking at a mood board. Stages must correspond to an event that either happened or did not — the visit was scheduled, the quote went out, the contract was sent. Then the pipeline is evidence.

Field sprawl at setup. It is tempting to capture everything while you are configuring anyway — industry, source detail, company size, referral notes, preferred contact time. Two weeks later the mandatory fields are being filled with full stops to get past the form, and every report built on those fields is quietly wrong. Fewer fields, each with a named owner and a moment when it gets filled, beats a complete schema nobody honours.

Shipping a system the team was never brought into. A CRM the sales team does not use is money set on fire, and the failure is invisible for about a month because deals keep closing the old way while the dashboard slowly diverges from reality. The people who have to type into it need to see something in it for themselves on day one, and someone needs to be watching in weeks two and three for the workarounds — the deal that never changes stage, the lead logged after it closed, the notes kept in a personal file.