Skip to content
BSENAvailable
Menu
BSENAvailable
Tell us about your project

How we decide

Three options, and what each depends on.

Every proposal has a fastest, an optimal and a best option, each with its price and what has to be true in two years for it to pay off. You choose.

Tell us about your projectSee the case studies
threshold.reviewlive
Contract value / year$100K
Recommendation
Negotiate with the current vendor

Below the threshold, building costs more than it saves: about twice the price in our 2026 review.

scenarios, five-year cost each5
date to revisit the decisionset
assumptions ranked by sensitivity✓
01 / 04

Three options for a campaign and email system.

We built this system for a marketplace. The client chose the best option: the company had been running for fifteen years and knew it wanted a system rather than a platform.

FastestAn account on a hosted service, its templates, users imported from a CSV.The business depends on a service you have no say over.
OptimalWe send the email ourselves, through an external service.More control, but the data still sits in someone else’s dashboard.
BestOur own system that builds templates, sends, and tracks bounces and opens, visible per user in the admin.One system covers campaigns, renewals and bulk sends.✓ Chosen

You own the logic and data, and pay for sending.

Sending email is a commodity, so we didn’t build that part from scratch. We used SES and built the system around it: 55 templates, a message queue and per-user tracking that shows on each user’s profile in the admin.

Search cluster5%

of the capacity it was sized for: a five-year roadmap whose features were never built.

Time-series databaseReports

Planned for predictive analytics. Used for ordinary reports.

02 / 04

Where “best” was wrong.

The technical choices were sound. Each time, the mistake was financial: the system was built too early, for a plan that then didn’t happen.

So every option we propose now has one more line: what has to be true in two years for it to pay off.

03 / 04

We work out the break-even point, and you decide.

A typical proposal says what will be built. Ours also says under what conditions it pays off.

If an assumption turns out wrong and the break-even point isn’t reached, the decision can be revisited.

04 / 04

What a review looks like.

In 2026 we were asked to make the case for replacing a BI vendor with an in-house build. We worked out the cost: building came out at about twice the price, so we recommended negotiating with the vendor.

The document itself stays private: it names the client’s customers.

01Five scenariosEach with its cost over five years.
02Where the answer flipsA table showing the contract value at which the recommendation changes.
03A threshold for buildingA contract worth about $200K a year. Below that, negotiating costs less.
04Dated checkpointsEach with the criteria for stopping there.
05An assumption registerEvery assumption marked by how sensitive the answer is to it.
06Bias named both waysWhere the model leans for and against building, left uncorrected when it can’t be estimated fairly.
Established companyWe plan at least ten years ahead.
StartupWe do the opposite and start with the fastest, cheapest option, so fixed costs stay low and there is time to reach revenue. The long-term plan comes after the MVP.
Services47 services, each with three options to choose from.See the services →Case studiesEighteen systems and the decisions behind each one.See the case studies →

How we decide

Facing a decision? We’ll prepare three options and the break-even point.

The first call takes 30 minutes, and you’re under no obligation.

Tell us about your projectcontact@microtech.consulting