Why do quotes take so long to go out?

Quoting starts out easy. In a new business, one person knows the materials, the labour, the margins and the customer, so a quote is a short job done in an afternoon. Nobody builds a process for it because nobody needs one. Then the business grows, and the same task starts taking a week.

Why did quoting get complicated?

Two things grow at once. The products and services get more complex, and the range of them gets wider. Together they pull more people into every quote.

Someone takes the enquiry and works out what is actually being asked for. Someone goes to site and measures. Someone prices the materials. Someone else owns the specialist section that only they understand. Each one is good at their part, and each one has to wait for the person before them.

That waiting is the cost. Not the thinking, not the expertise, just the gap between one person finishing and the next one starting. And here is the part worth sitting with: almost everything those people hand over is a data point. A measurement, a rate, a quantity, a unit price. Genuine judgement is a small share of what a quote is made of, but it is buried inside a pile of fetching and typing that hides it.

Why does every quote need re-pricing?

Because prices will not sit still. Materials, carriage, energy, subcontract labour: they move, and they move quickly enough that a rate you used last month is not safe to reuse this month.

That adds a task to every single quote. Before anyone can price their section, they have to go and find out what the number is today. Multiply that by the number of people involved and you have added days to something that used to take an afternoon.

Then the quote goes out and the customer goes quiet. Three months later they come back, keen. The quote has expired, so the whole pricing exercise runs again from the start, with the same people, for the same job. Options make it worse. Every alternative specification you offer is another set of numbers to gather, check and keep current.

Should I just buy quoting software?

Sometimes, yes. If you map your process out and a package covers it comfortably, buy the package. It is cheaper and simpler than building anything, and there is no prize for the harder route.

The reason people go looking for quoting software in the first place is usually that their process has outgrown a spreadsheet. That same complexity is what makes packages awkward. A quoting package wants all the information inside it, which means moving your data in and reshaping how your teams work to suit the tool. If your quoting is genuinely complicated, the tool is often a little too generic to hold it.

The alternative is to leave your systems where they are and put a layer over the top that connects them. The pricing stays where it lives, the customer records stay where they live, and the automation reaches into each one and pulls what it needs. Nobody has to move house.

Which of those is right depends on how complex your quoting actually is, and you cannot answer that until you have written the process down. If you are searching for quoting software, the odds are it is complex. Prove it either way before you spend.

What can the machine actually do?

Gather, assemble and remind. That is the honest scope, and it covers more of the week than it sounds like.

That last stage is the one people underrate. When a quote finally goes out, everyone involved is relieved and immediately starts the next one. Asking them to also remember that a particular quote turns seven days old on Thursday is asking a lot. Businesses solve it by giving someone the job of chasing quotes, which works and costs a salary. A reminder triggered by the send is the same outcome for none of the attention.

The accuracy question is the one that decides whether any of this is worth doing, and it is answered by testing rather than by assurances. Run known jobs through, compare the output against what your team would have produced by hand, and count the differences. Once that error rate is proven low and stays low, the sign-off stage changes character. It stops being a full recalculation and becomes a sample check.

Where does your pricing data end up?

A quoting automation handles some of the most commercially sensitive information you own: supplier rates, margins, who gets which discount. Before anything is built, know where that data travels and set the rules. Send a system only the fields it genuinely needs rather than whole documents. Where a general purpose model is involved, use a business tier with a written agreement that your data is not used for training. Where the information is too sensitive for that, run a model on your own server so nothing leaves your control, and pin processing to the United Kingdom or Europe. This is the same argument as self-hosted against cloud automation, and quoting data is exactly the case that makes it concrete.

What must a person still own?

Final approval, and the send itself.

Unless the quoting is genuinely simple, no quote should be produced by a system and emailed to the customer with nobody in between. Someone looks it over, confirms it reads right, and sends it themselves with the attachment. That last part matters more than it appears to. A quote that arrives from a person is a different object to one that arrives from a machine.

The second thing a person owns is the judgement that was never written down. Loyalty rates, goodwill on a job that ran long, a discount because of how the relationship has gone this year. None of that lives in a rules table, and it should not be forced into one. The system produces the quote as the rules say it should be. The person applies what they know about that customer, checks the whole thing, and sends it.

That division is the point. The machine takes the fetching and the arithmetic. The person keeps the relationship and the final word.

How do I start on Monday?

Map the quoting process. Not a diagram of how it is supposed to work, a record of how it actually works.

Take one real quote and follow it from the enquiry to the customer’s inbox. Write down every person who touches it, in order. For each one, record what they do, what they produce, where they get their numbers from, what checks exist to make sure they got it right, and what risk each check is there to catch. Then add two numbers to every stage: how many people are involved, and how long it takes.

Those last two numbers are what let you prove anything later. If you know a quote currently takes eleven days and four people, you can measure what it takes afterwards and put a figure on the saving. Skip that step and you will be arguing about whether it feels quicker. There is more on why the measurement comes first in audit the process before you automate it and how you know the automation actually saved anything.

Once the map exists, sorting it is quick. Stages that just move data go on one pile. Stages that apply a fixed rule that rarely changes go on the same pile. Stages where someone has to think about something that is not a number stay exactly where they are. The first pile is where the week goes, and it is the pile a system can take.