How to Save Thousands by Replacing Bloated SaaS Tools

Sick of paying heaps for your software tools? Same. That's why we rebuilt our quoting tool, it took 2 weeks and will save us thousands of dollars every year.

Jun 21, 2026

The view from High Street Auckland, watercolour by Steve Dunlop

We turned off PandaDocs this week. If you have never used PandaDocs then it is similar to Docusign and helps you with quotes and documents.

Screenshot of entering your information in the highlighted fields

PandaDocs cost $49 USD per seat per month.

4 seats is $200 USD or $380 NZD a month, $4,560 a year.
We technically need 12 seats on that which is $12,552.

That is a heck of a lot of money for a tool that helps you to make quotes. Then you start to add up the cost of all the other software tools that you have and it’s getting ridiculously expensive.

I’ve talked to so many businesses that are struggling with rising software costs and also managers of divisions where you have a fixed budget for the year and these costs are really starting to bite.

All up this quoting product took us about 2 weeks to make and we made that in between making products for our clients.

Not only did it take 2 weeks, it’s now way better than PandaDocs as you can generate quotes with AI.

If you fill out that form it basically one shots the quote which is awesome. It’s trained on 50 of our existing quotes as well so it is in our tone of voice.

PandaDocs doesn’t have this feature.

Or if we want to upload a meeting transcript from Fireflies then we can do that to get it to help generate the draft.

For all 12 users in our company it will probably cost us around $80 NZD a month, including the AI.

It’s not exaggeration to say that the product is 10 times better and it’s 10 times cheaper.

Cool story Chris, so you are just way cooler than PandaDocs?

My point of this story isn’t to be like I’m so cool, it’s more to point out how fundamentally broken SaaS products are at the moment and particularly if you live in a non USD earning country, how the cost benefit of these systems just don’t add up.

Let me ask you 2 questions.

When was the last time your favourite SaaS product built a feature that actually benefited you?
When was the last time your favourite SaaS product upped it’s price?

Here is how most people answer this

Never
Every single year

If you answered that as well then doesn’t that tell you something is fundamentally broken?

I want to show you that given the right circumstances you absolutely can build an internal tool for yourself and that also it’s not as wild and scary as everyone makes it out to be.

I think that everyone has a form of Stockholm syndrome, where you get used to loving your crappy tool because that is your only choice. The only exception to this is Sharepoint, it seems that nearly everyone hates Sharepoint. I think even the creators of Sharepoint hate Sharepoint.

Elon Musk and the Idiot Index

Elon Musk has this concept of the idiot index.

Cost of finished thing ÷ cost of raw materials

So for rockets:

A rocket part costs $100,000 to buy from a supplier.

But when SpaceX breaks it down, the raw aluminium, titanium, wiring, valves, fasteners, etc. only cost about $2,000.

So the idiot index is:

$100,000 ÷ $2,000 = 50x

Meaning the part costs 50 times more than the raw stuff inside it.

Musk’s view would be:

“That is insane. Either the manufacturing process is wildly inefficient, the supply chain is bloated, or we should make it ourselves.”

And no, as much as people make it out to be, building B2B SaaS software is not rocket science.

You are paying for Salesforces marketing

Did you know that 35 - 37% of Salesforce total revenue is allocated to marketing! That’s insane! That means that 35-37% of the price you are paying is subsidising them getting on other new customers and not really helping you.

This is pretty normal for most of these companies, they all have to grow and they all have huge sales teams.

So let’s do some idiot index examples for products.

Let’s start with the thing that I think businesses waste the most money on, CRMS.

The raw cost of sending 100,000 emails a month is $34.95 on SendGrid.

Let’s compare the cost using raw Amazon to Salesforce for 1.5 million emails a month.

Cool so in this case Salesforce is 100x more expensive than the raw cost. It’s just insane. This is why Salesforce has a tower and you don’t.

But if you do this for pretty much every product you will find that the idiot index is just growing and growing as time goes on.

Products are just so much more expensive than the raw costs.

The thing is that before AI was invented these products had you in a bind. They could extort you like this because what are you going to do about it?

You don’t know how to build a big complicated system.

But guess what, your AI sure does.

AI has got to the point where it is now possible to copy an existing system

If you write the above prompt it gets you started, but the AI already knows what these systems are and can help you cut straight to the point.

But but but - these products have so much, and you can’t build all that yourself

Here is the trick! You don’t need to build everything they have. Sure they have a lot of stuff in them but 90% of it is complete and utter junk and not built for you. Remember the golden rule of large scale SaaS is that most of the features aren’t built for you.

It’s like paying $5 million dollars for a house that has 715 bedrooms. That’s really cool that is has 715 bedrooms but you only need 3.

The bigger they are as well the more confusing they are to be onboarded too as opposed to just building what you want.

I call this the 10% rule. You pretty much usually only need 10% of an email system. Here is the big secret for a CRM system (you don’t really need it to integrate to your emails). As soon as you remove that requirement, you just removed 90% of the complexity and the cost.

Whereas these big SaaS companies push their product onto you and try to make you change to their way of working.

But what about the maintenance

The above is a picture of our timesheet system.

We built that 3 years ago because we didn’t want to pay $600 a month. Over the last 3 years we have saved close to $15,000.

Do you know how many bugs that the system has had in 3 years? Zero.

Do you know the number of times we have had to change anything in 3 years? Zero.

When you strip out all the junk from systems you don’t need to maintain things as much and anyway AI has made maintenance so much easier.

So what do you decide what to build

This table should help.

Start with the above list and yeah you probably want to stay away from trying to rebuild your accounting system to start.

When we built the quoting tool we boiled the 10% of things we needed down to this.

Create quote
Pull client/project info
Generate scope
Add line items
Export/send PDF
Track status

For a CRM if you are like me or most businesses, you might not even want an email integration. I find those confusing as anyway.

So a 10% CRM might be

List of clients
List of leads
Size of opportunity
A little kanban board

You could build the above system in a week.

How to identify bloated SaaS features

Basically just try to build things that you do daily or monthly. That will cut out all the junk.

The good thing is that if you have used a system a lot you will know what to cut out.

Who is the prime audience for this

The prime audience for this is bigger companies, especially companies with 100+ staff.

Why?

Because you are the ones getting owned the hardest.

You are paying massive per-seat licence costs for tools that are not really built around your business. You are renting generic software, paying enterprise prices, and then bending your actual workflows around whatever the tool allows.

That made sense when software was expensive and slow to build.

It makes a lot less sense now.

If you have 100 staff, every per-seat SaaS product becomes painful very quickly. A tool that seems fine at $49 USD per month becomes ridiculous once 40, 80, or 120 people need access. And the worst part is that the product usually does not become more tailored to you as you spend more. You just get a bigger invoice.

Larger companies also have another advantage: they can actually afford to build.

You do not need to hire a permanent product team for every internal tool. You can use external developers, AI, and a very narrow scope to build the 10% of the product your company actually uses.

But this is not only for large companies.

If you are a small business and you have one or two SaaS tools costing thousands of dollars a month, this thinking still applies. You might not build a full internal platform, but you can absolutely build a smaller workflow tool that saves real money.

For me the test is not company size, the test is whether the cost has become stupid. How do you know if a cost is stupid? Well usually it just sort of pisses you off all the time is usually the answer. PandaDocs pissed me off every single month. Use that as a test.

If a tool is expensive, bloated, used constantly, and only doing a narrow job for your business, it is worth asking: Do we need this whole product? Or do we just need the small part of it that actually matches how we work?

How much do these things cost to build?

If you can use Cursor and learn to develop then you can build them by yourself! This is great if you can and want to do that.

If you don’t want to, then here is how much you should budget.

If you want it to be really small then honestly $6,000 - $12,000 NZD can get you what you need. That would be how much we would quote you to build our quoting tool. I suppose if you are interested we could sell you a SaaS version of our quoting tool that we have made for like $50 NZD a month. (How else will I ever buy a tower like Salesforce if I don’t get into the SaaS game.)
Most larger businesses systems can build their first workable version for ~ $25,000 to $50,000 NZD.

In fact sometimes putting a constraint is really good, if you say I have a budget of $50,000 what can you make, then you are forcing yourself into making a focused product. I think this leads to great outcomes.

The biggest advantage you get is that then the product is under your control, you can tailor things exactly how you want them. Just like with our quoting tool. Hopefully this is helpful.

If you are looking to go deeper on what to buy and what to build, then look at this post.

Book Club

The football world cup is on. So some sports books will be good to read.

The Sports Gene by David Epstein (unfortunate last name now) is amazing. This book talks about how the money in sport has affected the gene pool basically and that the more money is, the more the outliers you get from an average person entering the sport with basketball illustrating this the best.

How to Win the Premier League is all about the data in football. Between 2012 and 2023, Ian Graham worked as Liverpool FC's Director of Research. His tenure coincided with the club’s greatest period of success since the 1980s, including winning the Premier League in 2020 – Liverpool’s first league title after three decades.

I’m an Amazon affiliate, I make a commission if you make a purchase.