Guides
Translation agency growth compared: fixing response time versus adding spend
Response time and added spend are the two growth levers for a translation agency, and the first is cheaper to pull, so fix reply speed before buying more inquiries.
What to take away
- Translation agency growth comparedfixing response time versus adding spend comes down to two levers. Answer the inquiries you already get, or buy more inquiries. Fix reply speed first.
- Added spend means Google Ads, Microsoft Advertising, LinkedIn Ads, ProZ.com and TranslatorsCafe. It also means the ATA directory, MultiLingual, Slator, LinkedIn Sales Navigator and Apollo.io. Every one buys inquiries you still have to answer.
- Deals die at five points, and only one of them is price.
- Log every loss in the buyer's own words, not your reading of them.
- Check the bench before you chase volume. A won project with no available linguist is a scheduling risk you took on the client's behalf.
Fix response time versus adding spend, side by side
Fixing response time costs process, not budget. Adding spend costs money, and it multiplies whatever your intake already does. If your intake converts one inquiry in ten, doubling inquiries doubles your quoting load and your losses with it.
Fix response time vs add spend
Fix response time
- What it is
- Answer inquiries faster
- Cost
- Staff attention
- Time to effect
- Days
- Main risk
- Discipline only
- Best when
- Losing inquiries quietly
Add spend
- What it is
- Buy more inquiries
- Cost
- Media budget
- Time to effect
- Weeks to months
- Main risk
- Unanswered inquiries
- Best when
- Intake converts well
Fix response time versus adding spend
Fix response time
- What it is
- Answer inbound inquiries faster and better
- Cost
- Staff attention and a written rule
- Time to effect
- Days
- Main risk
- None beyond the discipline
- Best when
- You already receive inquiries and lose them quietly
- Named tools
- Shared inbox, HubSpot Free CRM, Google Workspace, Calendly, Slack
- More channels
- Ruby or AnswerConnect phone answering, Zoom Phone, Aircall, Notion, Trello
- Typical price signals
- HubSpot Free CRM $0; Google Workspace $6 to $18 per user/month; Calendly $10 to $16 per seat/month; Ruby plans typically $235 to $500/month; Slack $8 to $15 per user/month
- Cost per won client
- C divided by R, plus staff hours already paid
- First project test
- Margin on the first project must cover the staff time you redirect
Add spend
- What it is
- Buy more inquiries through paid search, directories, trade ads, outbound
- Cost
- Media budget, plus the same staff attention
- Time to effect
- Weeks to months
- Main risk
- Paying for inquiries you cannot answer or staff
- Best when
- Intake converts well and the bench is deep
- Named tools
- Google Ads, Microsoft Advertising, LinkedIn Ads, ProZ.com, TranslatorsCafe
- More channels
- ATA directory, MultiLingual, Slator, LinkedIn Sales Navigator, Apollo.io
- Typical price signals
- Google Ads search clicks typically $1 to $6; LinkedIn Ads often $8 to $20 per click; ProZ.com membership typically $120 to $200/year; ATA membership typically $200 to $300/year; MultiLingual or Slator ad placements typically $500 to $3,000
- Cost per won client
- C divided by R, plus media budget and the same staff hours
- First project test
- Margin on the first project must clear C divided by R, or the client must renew
For most small agencies the left column wins, because the inquiries are already arriving and going unanswered. Spend is the right lever only after intake converts.
Run a loss review first
Take the last twenty inquiries. For each, write down what happened and where it stopped. Not why you think it stopped. Where.
Five stages where inquiries are lost
- Inquiry never answered fast enough
- Qualification never happened
- Quote buyer could not evaluate
- Silence after the quote
- Loss on price to a buyer who wanted something else
Sort them into five stages. The stage with the most losses is your problem, and it is rarely the one you assumed.
Stage 1: the inquiry never answered fast enough. A content owner with a deadline contacts three vendors. The first useful reply usually wins, and useful means a real question about their file, not an automated acknowledgment.
Stage 2: the qualification that never happened. You quoted without seeing the file, without knowing the deadline, or without knowing who signs. Those quotes convert badly and consume the same effort as the good ones.
Stage 3: the quote the buyer could not evaluate. A total with no breakdown gives a buyer nothing to compare except the number, so they compare the number.
Stage 4: the silence after the quote. Nobody followed up. This is the cheapest fix on the list.
Stage 5: the loss on price to a buyer who wanted something else. Some are real price losses. Many are scope losses wearing a price costume, where a cheaper competitor quoted less work and the buyer did not notice.
Fix them in this order
- Cut first-response time.Name who answers inbound inquiries and by when. Reply with one specific question about their content.
- Write a five-question qualification.What is the file, what is it for, which locales, what is the deadline, and who approves. Ask all five before quoting.
- Show the breakdown on every quote.Give the buyer something to understand other than the total.
- Follow up twice, on a schedule.Once with a useful addition, once to close the loop. Then stop.
- Record the loss reason verbatim."We went with someone who had done device manuals before" is a positioning instruction. Paraphrasing it into "lost on price" throws it away.
When adding spend is the right lever
Spend earns its place when three things are true: intake answers within the hour, the qualification questions get answered, and the bench covers the pairs you sell. Miss any one and the budget buys quoting work, not revenue.
When adding spend is the right lever
Intake answers within the hour?
qualification questions get answered
budget buys quoting work, not revenue
Decide the budget as arithmetic you can check. If an inquiry costs C and converts at rate R, your cost per won client is C divided by R.
Compare that with the margin on a first project for that client, which is revenue minus linguist cost, minus review, minus DTP.
If the first project does not clear its own acquisition cost, the client has to renew before the spend pays.
Paid search, directory listings and trade advertising all sell the same thing: attention.
| Channel | What it does | Typical price signal |
|---|---|---|
| Google Ads | Search ads on queries such as "certified Spanish translation" | Clicks typically $1 to $6 in language services |
| Microsoft Advertising | Search ads on Bing and partner sites | Clicks typically $0.50 to $4 |
| LinkedIn Ads | Job-title and industry targeting for B2B localization buyers | Often $8 to $20 per click |
| ProZ.com | Directory profile and job notifications for translators and agencies | Membership typically $120 to $200/year |
| TranslatorsCafe | Directory and forum exposure to translators and buyers | Free profile; paid upgrades are listed on the site |
| ATA directory | Listing in the American Translators Association directory | Membership typically $200 to $300/year |
| MultiLingual | Trade magazine ads and sponsored content | Placements typically $500 to $3,000 |
| Slator | Industry news ads and sponsored content | Placements typically $500 to $3,000 |
| LinkedIn Sales Navigator | Prospecting filters for outbound sales | About $99/month |
| Apollo.io | Contact data and email sequencing | About $49 to $99 per user/month |
The Google Business Profile guidance on improving local ranking lists relevance, distance and prominence. It points to complete business information, accurate hours and reviews. Treat that as a vendor describing its own product, which is the one thing vendor documentation is authoritative about.
Reviews deserve care wherever they appear. The FTC guidance for marketers on soliciting and paying for reviews covers honest reviews, disclosing material connections, and not suppressing critical ones. Asking only your happiest clients is the pattern to avoid.
The inquiries worth having
The goal is more of the right inquiries. Two levers change the mix.
The inquiries worth having
- Say what you specialize in
- Name a pair and a subject
- Say what you do not do
- Publish the boundary as a filter
Say what you specialize in. A page that names a pair and a subject brings fewer inquiries and converts more of them. It also spares you the individual with one page to translate who needs an hour of explanation.
Say what you do not do. Publishing the boundary is a filter that works while you sleep.
Positioning and buyer-type work sits in the marketing and growth guide.
Do not win work you cannot deliver
A won project in a pair with one available linguist is a scheduling risk you have taken on behalf of a client who does not know about it. Check the bench before you push demand, which is the discipline described in building depth in the pairs you sell.
What a quote should contain and how a proposal is structured belongs to the sales and proposal guide. The number in it comes from how pricing and margin are built. Keeping a won client is a delivery problem addressed in how a new account is onboarded and run.
Common questions
How many follow-ups are too many?
Two after the quote, spaced, each adding something. After that you are training the buyer to ignore you.
Should I quote every inquiry?
No. Quoting work you cannot staff, or work from a buyer who will not answer the five questions, costs you the time you needed for a real opportunity.
What if I keep losing on price?
Check whether you are losing on price or on scope. Ask the buyer what the other quote included. Often it included less, and the buyer did not know.
Does cold outreach work in this trade?
Sometimes, and only when it is specific: a named subject, a named pair, and a genuine observation about the recipient's own published content. Generic outreach performs badly and damages the sender.
