Key takeaways
- A dashboard is for decisions: keep it to a dozen numbers, each with a definition, an owner and a review rhythm.
- Use outside benchmarks for context: a typical Propertymark sales branch agreed 9.5 sales in August 2026, and a letting branch 9.92 new tenancies[1].
- Track the age of the sales pipeline, not just its size: around one in three transactions fails, and completion averages 120 days after an accepted offer[2].
- Put service and risk on the same page as revenue, because complaints and compliance misses cost more than most owners expect[3].
What an owner's dashboard is for
Most CRMs can produce dozens of reports, and most owners look at none of them regularly. The usual reason is not lack of interest but overload: too many numbers, defined differently in each report, with no clear link to a decision. A useful dashboard does the opposite. It shows a small set of numbers that tell the owner whether the business is on track this week, which branch or team needs attention, and what to ask about in the Monday meeting.
Before building anything, write down the decisions the dashboard should support. Typical ones are: do we need more valuations this month, is the sales pipeline at risk, are lettings voids creeping up, is any team overloaded, and are we missing compliance deadlines or upsetting clients. Every number on the page should help with one of those.
A dozen numbers that cover most agencies
| Number | Definition | Review |
|---|---|---|
| Valuations booked | Market appraisals booked in the period, by branch and source | Weekly |
| Instruction win rate | Instructions won divided by valuations carried out | Monthly |
| Stock on the market | Properties available for sale or to let at period end | Weekly |
| Sales agreed | Offers accepted in the period | Weekly |
| Pipeline age | Sales agreed but not exchanged, grouped by weeks since acceptance | Weekly |
| Fall-through rate | Sales agreed that fell through, divided by sales agreed, rolling three months | Monthly |
| Lets agreed | New tenancies agreed in the period | Weekly |
| Void days | Days between tenancies for managed properties | Monthly |
| Enquiry response time | Median time from first enquiry to first human reply | Weekly |
| Missed calls | Calls not answered and not returned within a working hour | Weekly |
| Compliance due | Certificates and checks due in the next 60 days, and any overdue | Weekly |
| Complaints | New complaints and open complaints by age | Monthly |
Write the definitions into a one-page document and keep them fixed. If 'sales agreed' means memorandum issued in one branch and verbal acceptance in another, the comparison is meaningless.
Using outside benchmarks for context
Your own trend matters more than any national figure, but benchmarks help you notice when something is unusual. In August 2026 a typical Propertymark sales branch placed about 10.8 homes on the market and agreed 9.5 sales, while a typical letting branch agreed 9.92 new tenancies and took 2.97 new fully managed instructions[1]. Letting agents also reported an average void period of almost four weeks[1].
Survey figures come from a sample of member agents and move month to month, so use them as a rough guide, not a target. A branch agreeing half the benchmark is worth a conversation; a branch agreeing slightly fewer in a quiet month is not.
Watch the age of the pipeline
A sales pipeline can look healthy while quietly going stale. The government says around one in three property transactions fails and that completion takes an average of 120 days once an offer is accepted[2]. In August 2026, 31.8% of Propertymark member agents said most of their agreed sales were taking more than 17 weeks[1]. Pipeline value alone hides that risk.
- Group sales agreed by weeks since acceptance: 0 to 4, 5 to 8, 9 to 16 and over 16.
- Flag any sale with no recorded progress update in the last 14 days.
- Show the expected exchange month next to each band, so cash flow forecasts reflect reality.
- Track the fall-through rate on a rolling three months, with the main reason recorded for each one.
Leading and lagging numbers
Fee income and completions are lagging numbers: by the time they fall, the cause happened months earlier. Valuations booked, instructions won, enquiry response time and viewings per property are leading numbers that warn you early. A good dashboard puts the leading numbers at the top, where the owner looks first, and the lagging ones underneath for confirmation.
Market context belongs on the page too, in one line. Rightmove reported sales agreed in October 2026 down 8% on a year earlier[4], while a net balance of 6% of RICS surveyors reported more new instructions in September[5]. If your numbers move with the market, the response is different from when they move against it.
Service and risk on the same page
Complaints are expensive in time, money and reputation. The Property Ombudsman resolved 7,681 disputes in 2025, 54% more than in 2024, and 61% of the cases it decided went in the consumer's favour[3]. Owners who see complaints only when one escalates miss the pattern behind them. Put open complaints, their age and the main themes on the dashboard, next to compliance items due and overdue.
A useful rule: any red item in compliance or complaints is discussed before revenue in the weekly meeting. It keeps attention on the issues that cause the most damage if ignored.
Building it from the systems you have
Write the definitions
One line per number, including the source system and the exact report or field it comes from.
Check the data
Run each report for the last three months and check it against what managers know happened. Fix data entry habits before building charts.
Start in a spreadsheet
Copy the numbers in weekly for a month. It is quick to change and shows which numbers people actually use.
Automate what survives
Move the numbers people use into the CRM's dashboard or a reporting tool connected to the CRM, phone system and accounts.
Assign an owner per number
The owner explains movements and proposes actions; the dashboard does not explain itself.
Design rules that keep it useful
- One screen or one page; if it needs scrolling, it has too many numbers.
- Show the trend over 13 weeks or 12 months, not just this week's figure.
- Show a target or a normal range so it is obvious when something is off.
- Split by branch and team only where someone will act on the split.
- Avoid vanity numbers such as page views or social followers unless they drive a decision.
- Label any figure that is an estimate or comes from a partial data source.
A review rhythm
- Daily, branch managers: missed calls, new enquiries without a reply, and today's valuations and viewings.
- Weekly, owner and managers: the dozen numbers, pipeline age, compliance due and open complaints.
- Monthly, owner: win rates, fall-through rate, void days, response times and staff workload by person.
- Quarterly, owner: tech and portal spend against results, and whether the dashboard itself still answers the right questions.
Worked example: a Monday dashboard
Example: a fictional two-branch agency's Monday dashboard shows valuations booked down from 14 to 8 week on week at one branch, while the other is steady. Enquiry response time at the same branch has risen from 25 minutes to three hours. The owner asks why; the branch has been one negotiator short for a fortnight and evening calls are going to voicemail.
Example, continued: the pipeline panel shows six sales over 16 weeks old with no update in a fortnight. The owner assigns a progression day to clear them and arranges cover for evening calls. Two weeks later valuations recover to 12 and response time is back under an hour. Without the leading numbers, the problem would have appeared as a fall in completions three months later.
Mistakes that make dashboards useless
- Reporting what the CRM makes easy rather than what the business needs, such as total contacts instead of active applicants.
- Changing definitions quietly, so this year's numbers cannot be compared with last year's.
- Ranking negotiators publicly on a single number, which encourages gaming, such as booking weak valuations to hit a target.
- Building it once and never asking whether anyone uses it; a number nobody discusses for three months should be removed.
- Treating gaps in data as zeros, which makes a branch with poor data entry look like a branch with poor results.
The fix for most of these is the same: a fixed definitions page, an owner for each number and a quarterly review of the dashboard itself.
Getting started this month
Pick six numbers from the table, write their definitions and copy them into a spreadsheet every Monday for four weeks. Discuss them in your weekly meeting and note which ones led to an action. Add the rest only when the first six are trusted. For the AI and software side of the numbers, see measuring AI return in an agency, and for one of the most telling ratios, our guide to the instruction-to-sale rate.
Frequently asked questions
What should be on an estate agency KPI dashboard?
A small set of leading and lagging numbers: valuations booked, instruction win rate, stock, sales and lets agreed, pipeline age, fall-throughs, void days, response times, missed calls, compliance due and complaints.
How often should owners review the dashboard?
Managers should check a few numbers daily, the owner and managers should review the full set weekly, and deeper ratios such as win rates and fall-throughs monthly.
Do we need business intelligence software?
Not to start. A spreadsheet updated weekly is enough to find out which numbers matter. Automate only the ones people use.
Should we compare ourselves with industry benchmarks?
Use them as context, not targets. Survey benchmarks come from samples and change monthly; your own trend over time is the more reliable guide.
References
Propertymark (2026). Housing Insight Report, August 2026 (opens in a new tab) ↑
Ministry of Housing, Communities and Local Government (GOV.UK) (2025). Home buying and selling reform (consultation) (opens in a new tab) ↑
The Property Ombudsman (2025). Annual Review 2025 (opens in a new tab) ↑
Rightmove (2026). House Price Index, October 2026 (opens in a new tab) ↑
Royal Institution of Chartered Surveyors (RICS) (2026). UK Residential Market Survey, September 2026 (opens in a new tab) ↑