First operations hire: when to make it and who to hire

Hire your first operations person when there is more than a full week of revenue-affecting work going undone, when a named decision-maker will back their calls, and when you can fund the seat for eighteen months. Screen for judgement under ambiguity, willingness to refuse work, and range across systems. Certifications and deep platform specialisation are the wrong filters at this stage.

The short answers

• The trigger is workload, not headcount. Count the hours of revenue-affecting operations work currently undone or done badly by someone whose job it is not. When that exceeds a full week, you have a hire. Every published headcount ratio is a heuristic dressed as a benchmark.
• The best sampled number available puts revenue operations at 6% of go-to-market headcount. ICONIQ’s April 2025 survey of 205 B2B software go-to-market executives found RevOps at 6% of GTM full-time employees, rising to 7% at high-growth companies. That implies the first ops seat around a 16-person revenue team.
• Many companies asking this question should not hire yet. If leadership disagrees about what you sell and to whom, an operations hire will spend a year encoding a disagreement instead of resolving one.
• Hiring late is the more expensive error. Nobody has published a cost for it, but the remediation work is visible: a CRM with two systems of record, commission paid on a number later corrected, and a definition of a qualified opportunity that nobody will say out loud without a caveat.
• Fractional and agency support are real alternatives, and they are cheaper. Published fractional retainers run from $3,000 to $15,000 a month against a fully loaded permanent cost of $143,000 to $250,250. Every one of those pricing figures comes from a vendor’s own page.
• The first hire should be a senior individual contributor with decision rights, not a manager with a team plan. Underlevelling produces a ticket-taker. Overlevelling produces someone with nothing to manage who leaves inside two years.
• Reporting to the VP of Sales is the default and it is usually wrong for a first hire. The first ops person has to arbitrate between sales, marketing and customer success, and cannot do that from inside one of them.

The readiness test for a first operations hire

The question “do we need RevOps yet” is usually answered with a headcount number, and headcount is the wrong variable. Two companies of 40 people can be a year apart on this. The one selling a single product through two sellers on annual contracts has almost no operations work. The one selling three products through partners, self-serve and a sales team, with monthly billing and a renewal motion, has a full-time job going undone.

What follows is a diagnostic. Score section A and section B separately. One point for each statement that is true of your company today. Do not score the company you plan to be in six months.

Section A: is the work there

#	Statement	Point if true
A1	More than one person publishes a revenue number, and their numbers do not match	1
A2	Somebody on the leadership team spends half a day or more each week rebuilding a report by hand	1
A3	The forecast is assembled in a spreadsheet outside the CRM, and that spreadsheet has an owner	1
A4	A new seller cannot reach their first meeting without a person explaining the process, because it is not written down	1
A5	At least one integration between two systems is maintained by hand, by export, or by somebody remembering	1
A6	Lead routing, territory assignment or account ownership caused a named argument in the last quarter	1
A7	A pricing, discounting or commission question took more than a week to answer in the last quarter	1
A8	You bought a tool in the last twelve months that has never been fully implemented	1
A9	The revenue team is twelve or more people across sales, marketing, customer success and support	1
A10	Somebody senior has said “we should fix the CRM” three or more times, and nobody has fixed it	1

Section B: can you support the hire

#	Statement	Point if true
B1	A named person will decide the questions the ops hire escalates, and has thirty minutes a week to do it	1
B2	Leadership agrees on what the company sells and who it sells to	1
B3	Somebody can approve changes to the CRM and to the definition of a qualified opportunity, and will confirm that in writing	1
B4	The fully loaded cost of the seat is funded for a full eighteen months	1
B5	Product market fit is not in active question	1
B6	The hiring manager can describe what success looks like at ninety days without using the word “help”	1

Reading your score

Section A	Section B	Verdict	What to do
7 or more	5 or 6	Hire permanently, now	You are already later than ideal. Open the role this month
7 or more	3 or 4	Hire, but fix section B first	The work is there and the support is not. Fix B1 and B3 before you advertise, or you will hire someone into an unwinnable job
7 or more	2 or fewer	Do not hire permanently	Take a fractional lead for two quarters. A permanent hire will leave, and you will conclude the role does not work
4 to 6	4 or more	Contract a scoped project	Buy six to twelve weeks of contractor time against a named deliverable, then re-score
4 to 6	3 or fewer	Not ready	The support problem is the real problem, and it is a leadership decision before it is a headcount one
3 or fewer	Any	Not ready	The work is not there. Hiring now buys you somebody who will invent work to justify the seat

The uncomfortable outcome is the third row and the last row, and between them they cover a good share of the companies who ask the question. Two failure patterns show up repeatedly in that group.

The first is a company where the founders still disagree about the ideal customer. Operations work is the encoding of decisions. If the decision has not been made, the ops hire either encodes one of the competing views and becomes a partisan, or waits, and gets characterised as slow. Neither ends well. What that company needs is two days of leadership argument, not a job advert.

The second is a company with genuine ops work and no appetite to change how anybody behaves. The hire is asked to fix the forecast without being permitted to change what a stage means, or to clean the data without being allowed to make a field required. That is not an operations job. It is a reporting job, and it is cheaper to buy it as a contracted project.

If you scored below the line, the useful next question is whether your next hire should be a seller instead, which is a budget trade worked through in RevOps hire or another AE.

The signals that you waited too long

Hiring an operations person early wastes money. Hiring late compounds. The early mistake costs you one salary for a year. The late mistake costs you a year of remediation on top of the salary, and it is paid by whoever you eventually hire, in their first two quarters, doing archaeology instead of the work you hired them for.

No published study puts a figure on this. What is observable is the shape of the debt, and the shape is consistent enough to list.

Signal	What it actually means	What it costs to unwind
Deals live in the CRM and in a spreadsheet, and the spreadsheet is the one people trust	You have two systems of record for the same object	Four to eight weeks of reconciliation, and a period where both numbers are wrong
Commission has been paid on a figure later corrected	The data feeding pay is not controlled	Clawback conversations, or eating the difference. Both damage seller trust for a year
Nobody will define a qualified opportunity without adding “well, it depends”	The stage gates are advisory	Every historical conversion rate you hold is unusable as a baseline
The CRM has more than a hundred custom fields and none of them are required	Nobody has been permitted to say no to a request	Field-by-field audit with each requester. Two to three months elapsed, mostly waiting on replies
A number in a board pack has been restated	The reporting layer has no owner	Board confidence, which does not recover on a schedule
Automations are running that nobody currently employed built	Institutional knowledge left with somebody	Reverse engineering, usually by disabling things and seeing what breaks
Reps maintain private trackers of their own pipeline	The system does not serve the people entering data into it	Adoption work, which is a behaviour problem and slower than a systems problem
Customer success verifies renewal dates by email because the field is unreliable	The post-sale data was never anybody’s job	A data remediation project across the whole customer base
“How many opportunities did marketing source last quarter” takes more than a day to answer	Attribution was never defined, only assumed	You cannot answer it retrospectively. You can only start counting properly from today
A renewal was auto-paid for a tool nobody uses	Nobody owns the stack or its calendar	Recoverable, and usually the one that pays for the hire in year one

Four or more of those and the first ops hire is not a growth investment. It is remediation, and you should say so in the job advert, because candidates who want a clean build will leave in month four when they find out. That mismatch is one of the recurring reasons ops hires fail inside ninety days.

There is a second cost, harder to see, which is what the absence does to everyone else. Salesforce’s 2023 research, based on 7,775 sales professionals surveyed between 24 August and 30 September 2022, found reps spending 28% of their week actually selling, with nearly 70% saying they were overwhelmed by the number of tools they had to use across an average of ten platforms per deal. That is a global survey of third-party panellists, so treat it as directional. It is consistent with what you see inside a company that has no ops function: the work does not disappear, it gets distributed across people who are worse at it and more expensive to interrupt.

What the first operations hire is actually for

The first operations hire is the person who decides things nobody has decided yet, and then makes the systems reflect those decisions. That is the job. Everything else on the job description follows from it.

This is the part most hiring processes get backwards. The advert asks for five years of Salesforce administration and HubSpot certification, because those are legible and testable. Then the person arrives into a company with no ops team, no documented process, no escalation path and no precedent, and spends their first month working out what the actual question is. Platform skill is necessary. It has almost no predictive power at this stage, because everybody who reaches a final interview has it.

Consider what a normal Tuesday looks like for a company’s only ops person. The VP of Sales wants a new pipeline stage added before the board meeting on Thursday. Marketing wants lead scoring changed because a campaign underperformed. Customer success has found forty accounts with the wrong renewal date. The CEO has asked, in passing, why last month’s number moved after it was reported. There is no manager to prioritise this. There is no ticket queue with an SLA. There is one person deciding which of those four things matters most, doing it, and being able to explain afterwards why the other three waited.

Judgement under ambiguity is the capability being purchased. Autonomy is the working condition it requires. A company that hires for judgement and then requires approval for every field change has bought the capability and disabled it.

The four things worth screening for

The screening problem is that judgement does not appear on a CV and does not survive a competency interview, because competency interviews reward rehearsed answers. Four properties are observable if you build the assessment to show them.

Range. Somebody who has worked across reporting, workflow design, data hygiene and at least one integration will be more useful in year one than somebody with six years of depth in a single platform. The first ops hire meets problems in the order the business generates them, which will not be the order of their expertise. A candidate whose entire history is marketing automation will do the marketing automation superbly and stall on the commission data.

Will they refuse work? An operations function that ships every request accretes complexity for two years and becomes the reason nothing can change afterwards. Ask a candidate for a specific request they declined, who asked for it, what they said, and what happened next. The answer separates people who understand that ops is a governance function from people who understand it as a service desk. Candidates who cannot produce an example have either never been given the standing to refuse, or have never used it.

Watch how they handle having been wrong. The question is not whether they made a mistake. Everybody has. The question is how fast they noticed, who they told, and whether they changed anything structural afterwards. Operations mistakes are public and quantitative: a routing rule that sent two weeks of leads to the wrong team, a report that overstated pipeline, a merge that lost account history. A candidate who describes a mistake in the passive voice, with no named consequence, is describing a story instead of an event.

They can explain the work to people who do not care about the mechanics. The first ops hire spends more time talking to a sales leader about why the forecast changed than they spend building anything. If they cannot explain a data model decision to somebody who does not want to hear about data models, the work will be correct and it will be ignored. Test this directly: ask them to explain something technical they built to somebody in the room who has no operations background, and watch whether they simplify or just slow down.

What not to screen for at this stage

Certifications tell you somebody sat an exam. They correlate with configuration competence and with nothing else that matters in a first ops seat. In a company with an established ops team and a change control process, a certification is a reasonable proxy for whether somebody can be trusted in the org on day one. In a company with no ops team, the constraint is not configuration skill.

Deep specialisation is the more expensive error, because it looks like seniority. A candidate with eight years exclusively in marketing operations at large companies has genuine depth, and that depth was built inside an environment with defined boundaries, a manager, a request process and other functions covering everything outside their remit. None of that exists in the seat you are filling. What you need is somebody comfortable being the whole function badly for a year before they are the whole function well.

Two other filters get overweighted. Industry background matters far less than hiring managers expect, because the systems and the failure modes are common across B2B software. Brand-name employers matter less than the size of the team the candidate sat in: somebody who was one of forty ops people at a large company may never have made a decision without an approver. Ask what they owned outright, and ask who could overrule them.

Screening for judgement requires seeing somebody work through a problem that has no clean answer. The design of that assessment is covered in the questions that surface operations judgement, and the wider process it sits inside is covered in how to hire a RevOps manager. Which of the five job titles you are actually hiring, once you have decided you are hiring, is a separate question answered in the role comparison.

Generalist or specialist for the first operations hire

For a first hire, generalist, almost without exception. The interesting question is when that stops being true, and the answer sits between 15 and 25 people on the revenue team.

The reasoning is arithmetical before it is philosophical. If revenue operations runs at 6% of go-to-market headcount, as the ICONIQ survey found across 205 companies, then a 16-person revenue team supports one ops person and a 33-person revenue team supports two. The Pedowitz Group’s staffing guide reaches a compatible shape from a different direction, recommending one generalist FTE up to 25 go-to-market staff and then one FTE per 30 to 40 as you grow, though it labels its own figures “TPG Planning Heuristics” and says explicitly that they are starting points and not rules.

Below 15, no sub-function generates a full week of work. A marketing operations specialist in a company with two marketers will spend three days a week doing sales reporting badly. Above 25, the generalist becomes the bottleneck, because every request in the company arrives at one calendar, and the second hire has to be a specialist or the first one leaves.

Revenue team size	Ops shape that works	The failure mode if you get it wrong
Under 12	No ops hire. Contracted projects, or a founder owning it explicitly	Hiring produces somebody who invents work to fill a week
12 to 25	One generalist, senior, broad, with decision rights	Hiring a specialist produces excellence in one area and neglect in three
25 to 50	Generalist lead plus one systems or analytics specialist	Keeping one generalist produces a queue and a resignation
50 plus	Function splits along sales, marketing and systems lines	Keeping a flat team means definitions stop being arbitrated

Published ratios disagree with each other by a factor of three, and the reasons are set out in full in the ops headcount ratio comparison. The short version is that most of the widely quoted figures are practitioner heuristics offered as advice, and their authors say so in the source documents. The ICONIQ percentage is one of the few numbers in this field that comes from a survey with a stated sample and date, which is why it anchors the table above.

Permanent hire, fractional lead, contractor or agency

Pretending that a permanent hire is the only serious option damages the argument, because it is not true, and any founder who has looked at the market knows it is not true. Three alternatives are genuinely competitive at the first-hire stage, and each buys something different.

Option	Published cost	What you are buying	Ramp	When it is the right answer	When it is not
Permanent senior IC	$143,000 to $250,250 fully loaded	Continuity, standing, and someone who can say no	30 to 90 days	Section A is 7 or more and section B is 5 or 6	You cannot fund eighteen months, or nobody will back their decisions
Fractional RevOps lead	$3,000 to $15,000 a month, commonly $4,000 to $7,000 for a first engagement	Senior judgement, part time, immediately	Week one	You have the work and not the support structure, or you need decisions before you need capacity	The problem is volume of build work. Fractional leads do not have the hours
Contractor on day rate	Vendor hourly rates of $200 to $400 for comparable systems work; no sampled US day-rate data	Hands on a named deliverable	Days	A scoped project with a defined end: a migration, an implementation, a data clean	You need somebody to decide what the deliverable should be
Agency retainer	$3,000 to $27,000 a month by tier. Projects $8,000 to $150,000	Bench depth, multiple skill sets, no hiring risk	Two to four weeks	Several build projects at once and no internal owner needed yet	You need one accountable person. Agencies rotate staff and they own nothing

Read those cost figures with the source in mind. The permanent numbers apply Rareix’s published bands, which sit at $100,000 to $175,000 base for a RevOps manager, plus employer costs of 43%. That loading is real: in the BLS Employer Costs for Employee Compensation release for March 2026, benefits were 30.1 per cent of private-industry employer costs, which is 43 per cent on top of wages, before equipment, tooling and any recruitment fee. The full band detail sits in the salary benchmarks.

The contractor figure is the weakest in the table. Rareix does not publish a sampled day-rate dataset, and the $200 to $400 an hour is a vendor’s list price rather than survey data, so treat it as an indication of the order of cost rather than a benchmark. Two days a week for six months at the bottom of that range is about 52 days, or roughly $83,000 at eight hours a day.

Every fractional and agency figure above comes from a vendor’s own pricing page. MergeYourData publishes retainer tiers from $3,000 to $27,000 a month with hourly consulting at $200 to $400. OpsEthic’s April 2026 cost guide sets four tiers from $2,500 to $15,000 a month against 8 to 60 hours. Mountainise publishes a compatible ladder at $4,000 to $15,000. None of the three cites a survey, a sample or a source. They agree with each other, which is worth something, and they are all selling the service being priced, which is worth remembering. Ask any provider for a rate card and two references at your stage before you treat the published range as your range.

Three things are true about the alternatives that their advocates rarely say.

A fractional lead cannot be the person who refuses your VP of Sales. They are a supplier, invoiced monthly, and the person who signs the invoice sits next to the person being refused. Fractional works beautifully for diagnosis, sequencing and building the first version of the definitions. It works badly for governance, because governance requires somebody who is still there in eighteen months and whose standing does not depend on renewal.

Agencies own outputs and not outcomes. That is not a criticism, it is the contract. If you need a Salesforce instance rebuilt, an agency will do it faster and more cheaply than a first hire learning on the job. If you need somebody to notice in March that the pipeline definition adopted in January is producing a distorted forecast, an agency will not, because nobody there is watching your numbers on a Monday morning.

Contractors inside the UK bring a compliance question that founders often meet for the first time here. Under the off-payroll working rules, medium and large private sector clients determine the contractor’s employment status and issue a status determination statement, and where the engagement is inside IR35 the deemed employer deducts income tax and employee National Insurance and pays employer National Insurance on top. Small companies are outside that obligation and the contractor’s own intermediary decides. Either way, price the engagement on gross cost.

What the first operations hire owns on day one, and what they do not

The commonest scoping error is to give a first ops hire everything with the word “revenue” near it and then be surprised when nothing gets finished. The second commonest is to give them nothing with a decision in it and then be surprised when they leave.

Area	Day one status	Why
Definition of a qualified opportunity and the stage gates	Owns	This is the arbitration the role exists for. If somebody else owns it, the hire is an analyst
CRM field architecture, required fields, validation	Owns	Nobody else will refuse a field request, and refusing field requests is most of data quality
Lead routing and account assignment rules	Owns	Mechanical, contested, and the thing that generates the arguments in section A6
The weekly pipeline and forecast pack	Owns	Owning the number’s production is what makes the definitions stick
Tool inventory, contracts and renewal calendar	Owns	Cheap, unglamorous, and usually recovers a meaningful share of the salary in year one
Data quality standards for accounts and opportunities	Owns	Standards without an owner are documentation
Territory design and quota setting	Advises	Sales leadership decides. Ops models the options and holds the data
Commission plan design	Advises, administers	Designing the plan is a leadership decision. Calculating it accurately is an ops one
Sales process and methodology	Advises	Ops owns the instrumentation of the process. The sales leader owns the process
Pricing and discount policy	Not theirs	Finance and leadership. Ops enforces the approval path once it exists
Marketing budget allocation	Not theirs	Attribution is an ops output. Spending decisions are not
Headcount planning	Not theirs	Ops supplies capacity models. It does not set the plan
Building the ops team	Not in year one	A first hire with a hiring plan is a first hire not doing the job
The board narrative	Not theirs	Ops produces the numbers. The CEO explains them

Write that table, or one like it, before the role is advertised, and countersign it with whoever the hire reports to. It takes an afternoon. Skipping it is the single most reliable predictor of a bad first ninety days, which is worked through in why ops hires fail early.

Who the first operations hire should report to

There is no survey answering this. Forrester’s study of B2B revenue operations, based on 316 respondents, measured whether the function was centralised and found centralised models rising from 15% to 40% across two years. Centralisation is not the same question as the reporting line, and the published guidance on the reporting line is practitioner advice with no sample behind it. What follows is reasoning about what each line optimises for, applied to a first hire specifically.

CEO, COO or chief of staff. The right answer in most cases at this stage. It makes the arbitration mandate real, it gives the hire an escalation path that resolves in days, and it signals to sales and marketing that ops rulings are not partisan. The cost is the CEO’s attention, and the failure mode is a CEO who takes the reporting line and then never takes the meeting.

CFO. Works when the pain that triggered the hire is the forecast, the board pack or the billing data. It brings discipline and a natural home for definitions. The risk is drift: ops reporting to finance tends to become a reporting function, and the process work that produced the mess in the first place never gets touched.

VP of Sales or CRO. The default, because the budget already exists there and the loudest pain is usually in the pipeline. It is usually the wrong answer for a first hire, for two reasons that compound. The first is arbitration. A person who reports to sales cannot credibly settle a dispute between sales and marketing, and marketing will read every ruling as partisan whether or not it is. The second is capacity, and it is the one specific to a first hire. A VP of Sales with sole call on the only ops person in the company will consume the whole role inside a fortnight on pipeline reporting and CRM requests. The marketing data, the renewal data and the tool sprawl, which is to say most of the work that justified the hire, will never be reached. If this is the line you choose, ring-fence a fixed share of the hire’s week in writing for work the VP of Sales does not set.

Head of Marketing. Rare, and it inherits the same partisanship problem in the opposite direction. It appears in product-led companies where marketing owns most of the funnel, and it works there for as long as that remains true.

The test to apply is simple. Name the two functions most likely to disagree about a definition in the next six months. If the ops hire reports into either of them, you have chosen a line that guarantees the first serious dispute escalates anyway.

The seniority question, and why first ops hires get underlevelled

Companies underlevel the first operations hire because the budget conversation happens before the scoping conversation. The role gets posted as an analyst or a coordinator, at analyst money, and the responsibilities describe a job that involves telling two vice presidents that their numbers are wrong.

The costs of that are predictable. An underlevelled hire cannot refuse a request from a VP, so the request gets built, and the accretion described in the section on saying no begins on week three. Every decision escalates to a founder who does not have time for it, so decisions queue, so the hire looks slow. The work drifts towards the tasks the person can complete without permission, which is reporting, and reporting is the part of operations that produces the least durable value. Fourteen months later they leave, the CRM debt is worse than when they arrived, and the company concludes that the role does not work. The search restarts at a higher level, and the true cost includes the elapsed time, which is set out in how long a RevOps search takes.

The counter-argument deserves to be taken seriously, because it is also true. Hire a Director or VP of Revenue Operations into a company with no ops team and no near-term plan to build one, and you have bought a manager with nothing to manage. They will do the work capably for a year, discover that the next rung does not exist, and go somewhere with a team attached. The academic literature on this is about perception more than title: Erdogan and Bauer’s 2021 review of the overqualification research in the Annual Review of Organizational Psychology and Organizational Behavior finds consistent negative associations between perceived overqualification and job satisfaction, with turnover a documented consequence, and it identifies autonomy and empowerment as substantial moderators of whether the effect is negative at all.

That moderator is the resolution. A senior person in a first ops seat stays if the job is genuinely decision-rich and leaves if it is a queue. So the level to hire is a senior individual contributor with real decision rights, priced accordingly, with no team and no promise of one. Say that explicitly in the process. Candidates who want a team will select themselves out, which is the correct outcome for both parties, and candidates who want to own a function without managing anybody are a larger population than most hiring managers expect.

One practical marker: if the person you are hiring would need approval to make a field required in the CRM, you have levelled the role below the work.

How to test judgement instead of tool knowledge

Everything above is a scoping exercise. The selection problem remains, and it is the harder half, because judgement is invisible on a CV and rehearsed in interviews.

The mechanism that works is to put the candidate in front of a problem that is genuinely underspecified and watch the order in which they attack it. Give them a messy pipeline export, a conflicting set of definitions from two functions, and a question a CEO would actually ask. The signal is not the answer. It is whether they establish what problem they are solving before they start, whether they say out loud what they are choosing not to do, and whether they can explain the trade-off afterwards to somebody who does not care about the mechanics.

That is the assessment Rareix runs, in one recorded sitting, and the reason it is recorded is that the reasoning sequence is the output. A written answer hides the order of operations. The full argument for that method, and the design of the questions inside it, sit in the interview questions that surface judgement and how to hire a RevOps manager. If the role you land on turns out to be a builder rather than an arbitrator, the distinct capability set is described in what a GTM engineer actually does.

Hire the person who can decide, or do not hire yet

The readiness test at the top of this page exists because the most expensive version of this decision is not hiring early or hiring late. It is hiring correctly into a company that has not decided who gets to make decisions.

Score section A honestly. If the work is there, score section B more honestly still, because that is the section founders inflate. Seven or more in A with two or fewer in B is a real result and it has a real answer, which is a fractional lead for two quarters while you fix the support structure, then a permanent hire into conditions that will hold them.

When you do hire, screen for the four properties, level the role at senior individual contributor with decision rights, report it outside the sales organisation, and write down what the person owns before the advert goes out. None of that is expensive. All of it is easier to do in the week before you hire than in the quarter after.

Rareix pauses a meaningful number of first ops searches at the scoping stage for exactly this reason. The conversation that surfaces the disagreement is worth more to the company than the shortlist would have been.

Related reading

• Which of the five ops roles you are actually hiring, once you have decided that you are hiring one.
• RevOps hire or another AE, for the budget trade between ops headcount and sales headcount.
• How to hire a RevOps manager, for the search process itself.
• RevOps interview questions, for the questions that surface judgement instead of recall.
• Why ops hires fail in 90 days, for the onboarding plan this hire needs on arrival.
• RevOps salary benchmarks, for the bands behind the cost comparison on this page.
• What a GTM engineer actually does, if the constraint is execution volume rather than definitions.
• Why your RevOps job description attracts the wrong applicants, for turning the ownership table into an advert.
• How long a RevOps search takes, for planning around the elapsed time.