How to benchmark in-house team performance

Key takeaways

To benchmark in-house team performance, measure five things for each process: volume, time per task, cost per task, error rate and turnaround. Compare against your own history first, then against an outside quote for the same scope. Without a benchmark, an in-house team can be slow and expensive for years and nobody will know.

• 5 measures: volume, time per task, cost per task, error rate, turnaround.
• Benchmark against your own last quarter before anyone else.
• An outside quote for the same scope is a free market benchmark.
• Measure the process, not the person.

Why is in-house performance rarely measured?

In-house performance is rarely measured because nothing forces it. A supplier is judged against a contract. An internal team is judged on whether problems reach the owner. As long as work seems to get done, cost per task and error rate stay unknown.

• The hidden costs of doing everything in-house
• Outsourced vs in-house quality: how to compare fairly

What should you measure?

Measure a small set of numbers for each process, weekly, with the same definitions each time.

• Volume: tasks completed per week.
• Time per task: total hours divided by volume.
• Cost per task: time per task x loaded hourly cost.
• Error rate: corrections per 100 tasks.
• Turnaround: time from request to completion.

• Work out your cost per task before you outsource
• BPO scorecard and KPIs

What do you compare against?

Compare first against your own trend, because definitions match. Then ask one or two providers to quote for the same written scope and service levels. The quote is a market benchmark even if you never buy. As indicative ranges, managed offshore support costs UK businesses around £9–£18 per hour and onshore £20–£35.

• Activity based costing for small businesses

How do you keep the comparison fair?

Keep it fair by comparing like with like. Use the loaded cost of in-house hours, not salary. Include management and correction time on both sides. Match quality standards and hours of cover. A cheaper option that answers in two days is not equivalent to one that answers in two hours.

When should you not benchmark?

Do not benchmark creative, strategic or relationship work by cost per task; the measures do not fit. And do not use benchmarks to rank individuals in a small team. The purpose is to see whether the way the work is organised is efficient.

• Core vs non-core activities: what to keep in-house

How do you collect the numbers without a new system?

Use what you already have. Most accounting, CRM and helpdesk tools can report how many items were processed in a period. For time, ask the team to note hours on the process for two weeks; a rough figure is enough. For errors, count corrections, credit notes or reopened items. A simple spreadsheet with one row per week is all the system this needs.

What do the numbers usually reveal?

First measurements tend to show the same few things.

• Cost per task is higher than anyone guessed, because of interruptions.
• Volume varies a lot from week to week, while staffing does not.
• A small number of error types cause most corrections.
• Turnaround is fine on average and poor at month end or during leave.
• Several people each do part of the process, and nobody owns it.

What should you do if the benchmark is unfavourable?

Start by improving the process, not by replacing the team. Simplify steps, give the process one owner and fix the top error cause, then measure again after a month. If cost per task is still well above an outside quote for the same standard, consider moving the routine volume and keeping the exceptions and customer-facing parts in-house.

• The cost of poor quality in a small business

What if the benchmark is favourable?

Then you have evidence that in-house is the right model for that process, and a baseline to protect. Keep measuring quarterly. A good result can erode as volume grows or people change, and the numbers will show it long before complaints do.

What does this look like in practice?

A pattern we see when firms first measure: invoice processing turns out to cost several pounds per invoice in-house, mostly through interruptions and corrections. The team is not slow. The process is spread across three people who each do it part-time.

Across our client work, back-office cost per task has fallen by 42% once the process was written down and run by a managed team.

• See client work
• BPO Solutions for UK businesses

Benchmarking checklist

Run this for one process over four weeks.

• Choose one high-volume process.
• Define the five measures.
• Record them weekly for a month.
• Calculate cost per task at loaded rates.
• Request an outside quote for the same scope.
• Decide: improve, keep or move.

• The hidden cost audit: score your business in an hour
• How to run a cost review in a small business

Next step

Send us one process and its weekly volume. We will help you calculate cost per task and show how it compares with market ranges in a 30-minute call.