A page that ranks for "email marketing software" will get you plenty of visitors, and very few of them will be agency owners about to sign up for a platform they plan to resell. Keyword research for SaaS affiliates who promote agency software starts with something much smaller. You're looking for the exact sentence an agency owner types at four in the afternoon, after a client's newsletter has landed in spam for the third week in a row.
Keyword tools make those sentences look worthless. They show zero to 20 searches a month, and sometimes no figure at all. Ignore that. Someone typing "why are client emails going to spam after changing smtp" has a paying client, a problem that's costing money and a reason to try a new platform this month. On a recurring programme, one agency that signs up pays you commission every month it stays. So one of those searchers is worth more than a thousand people skimming a top-ten list.
This guide covers six steps. You'll seed a list with real agency problems, expand it into the phrases people actually type, read the results pages, score each query on buyer fit, group what's left into pages and publish ahead of the busy season. The method has a cost, and you should know it upfront. You'll get less total traffic and slower early growth than you would chasing broad terms. In return, the readers you do get are the ones who sign up.
By the end you will have a scored list of agency pain point queries mapped to pages, and you need four things before you start
Keyword research for SaaS affiliates works best when it ends in a document you can build from. Here that document is a spreadsheet of 30 to 60 queries. Each one has a score and an assigned page type. Each row holds the phrase, the problem behind it, who ranks today, the intent stage, four buyer-fit scores and the page it belongs to. When you finish, you'll know what to write first and which call to action each page carries.
You need four things ready before the first step. None of them costs extra if you already run a site.
- Any keyword research tool that shows search volume and difficulty. A free tier is enough, because you'll treat its numbers as rough hints rather than verdicts.
- Access to your site's search console data, so you can see the queries you already get impressions for, including the ones no tool reports.
- A spreadsheet with the columns above set up in advance, so you're filling in rows and not designing a sheet halfway through.
- The terms of the programme you promote, already read. Attribution length, first or last touch and whether commission recurs all change which keywords are worth the effort. Our guide to the programme terms worth checking before you promote covers what to look for.
Step 1: Seed your list with problems agencies have, never with product categories
Start with a blank column headed "Problem" and write down 15 to 20 things that go wrong on client accounts. Use the agency owner's view, not the software vendor's. You want the moment where a client emails them annoyed, or where they're stuck on a retainer with nothing left to sell.
These seeds come from the questions agency owners and their clients really ask about email. Use them as your starter set and add your own from client calls, forum threads and support inboxes you've read.
- Client marketing emails landing in spam or promotions
- Why companies send from secondary domains instead of their main one
- Whether email list cleaning is legitimate or a gimmick
- Open rates that look too high to be real
- Running your own mail server versus paying a delivery service
- Securing a signup form before a client's site goes live
- Whether promo emails built as a single image still work
- Turning retainer advice into monthly platform revenue
- Warming up a new sending domain for a client
- Bots filling in signup forms with junk addresses
- Importing a messy client spreadsheet without losing contacts
- Scheduling sends for clients with customers in several time zones
- Showing clients a finished campaign design on the first call
- Reporting real clicks and bookings to clients, not vanity numbers
- Re-engaging subscribers who opened but never clicked
- GDPR consent and double opt-in on client lists
What you should see: a list where almost every line mentions a client, money or a deadline. If you read it out to an agency owner, they should nod at most of them, because they've dealt with these problems.
The mistake: seeding with categories like "email marketing platform", "white label email software" or "best email tool". Those seeds just grow into the broad, crowded terms you're trying to get away from. A category tells you what someone might buy. A problem tells you why they're buying it this week.
Step 2: Turn each problem into the phrases agency owners actually type
Every seed now gets expanded into five to eight phrases. You do that with two kinds of modifier. Agency modifiers show the searcher has clients: "for agencies", "for clients", "white label", "resell" and "client portal". Question forms show they're looking for an answer and haven't settled on a product yet: "why do", "how to", "is it worth", "should I".
Write in the trade's own vocabulary. Agency owners say smtp, gdpr, sending, marketing emails and form. They don't say "deliverability optimisation solution". To check a phrase, type the start of it into a search box and see what autocomplete offers. Then read the related questions further down the results page. Both are free and both come from what real people type.
Worked example one turns spam placement into seven phrases
- why are client emails going to spam
- warm up new sending domain for client
- marketing emails going to spam after changing smtp
- how to fix spam placement for agency clients
- is it worth using a secondary domain for client sending
- white label email platform with domain warmup
- why do new domains go to spam
Worked example two turns list cleaning into six phrases
- is email list cleaning legit
- should I clean a client email list before sending
- how to verify emails before a campaign for clients
- list cleaning gdpr compliant
- does list cleaning improve sending reputation
- clean client list before importing to new platform
Worked example three turns retainer revenue into six phrases
- how to turn a retainer into recurring revenue
- resell email marketing to clients
- white label email marketing for agencies
- what to sell retainer clients besides strategy
- agency recurring revenue from software
- client portal email platform for agencies
What you should see: 80 to 140 rough phrases across your seeds. Some will be clumsy. Keep them for now, because Step 3 sorts the real ones from the made-up ones.
The mistake: stacking modifiers until a phrase reads like nobody would ever type it, such as "best white label resell smtp gdpr platform for agencies clients". One agency modifier and one problem word per phrase is the limit. Also open your search console data now. Queries you already get impressions for, even three a month, are phrases real people typed, and they belong on the list.
Step 3: Read the results page before you trust the volume number
Search each candidate phrase in a private browser window and look at the first page of results. You don't need a paid tool for this. Your own eyes and ten minutes per batch of phrases will do. For each one, note who ranks. Is it forum threads, vendor help pages, thin listicles or a strong, recent guide?
Then give each phrase one of three intent labels in your spreadsheet. These labels decide the page type later, so be strict about them.
| Intent stage | What the searcher knows | Example phrase | What ranks today |
|---|---|---|---|
| Problem-aware | Something is wrong, not what fixes it | why are client emails going to spam | Forum threads, old help articles |
| Solution-aware | The fix exists, not which product | white label email marketing for agencies | Listicles, comparison pages |
| Product-aware | A shortlist of platforms | resell email platform with client portal | Review pages, vendor pages |
What you should see: forum threads and help articles from a few years ago on the first page for a good share of your phrases. That's the best sign you'll get. The need is real, because people keep asking, and the answer is weak, because no one has written a proper one. A strong, current guide in the top three tells you the spot is taken. Lower that phrase's rank score or drop it.
Look at who the ranking pages are written for as well. If the top results are aimed at shop owners running their own newsletter, the searcher is probably one too. If they're agency forums and threads about managing client accounts, you've found your reader. Our guide on targeting agency owners searching for client marketing tools goes further on reading those signals.
Low and zero volume figures undercount B2B searches
Keyword tools estimate volume from samples of search behaviour, and small samples are weakest for narrow business queries. An agency owner might type the same problem six different ways. The tool splits those six into separate rows of near-zero, or doesn't record them at all. Then the figures get rounded into buckets, so a phrase searched 15 times a month can show up as zero. Your search console data often shows impressions for phrases your keyword tool says nobody searches. When the two disagree, believe the search console.
There's another reason. A single page that ranks for one problem-aware phrase usually ranks for dozens of variants as well. Ten phrases at "zero" each can add up to a few hundred searches a month from people who have clients and a reason to buy.
The mistake: deleting phrases because the tool shows 0 to 20 searches a month. Those are the ones that convert. Cut a phrase only when the results page tells you to, because the intent is wrong, the searcher isn't an agency or a strong guide already owns the spot.
Step 4: Score every query on buyer fit, since keyword research for SaaS affiliates pays on who searches, not how many
Add four columns to your sheet and score every surviving phrase from 1 to 3 on each one. A 3 means yes, clearly. A 1 means no, or barely. Don't overthink it. Two minutes a phrase is plenty, and you'll get faster as you go.
- Agency fit. Is the searcher likely an agency owner or a freelancer with clients? The word "client" or "resell" usually earns a 3.
- Platform path. Does a full, honest answer lead naturally to a platform? Fixing spam placement does. Writing a better subject line mostly doesn't.
- Rank chance. Can your site realistically get onto page one? Forum threads and old help articles ranking earn a 3. A strong, current guide in the top three earns a 1.
- Buying distance. How close is the searcher to choosing a product this month? Problem-aware is usually a 1 or 2. Product-aware is a 3.
| Phrase | Agency fit | Platform path | Rank chance | Buying distance | Total |
|---|---|---|---|---|---|
| why are client emails going to spam | 3 | 3 | 3 | 2 | 11 |
| resell email marketing to clients | 3 | 3 | 2 | 3 | 11 |
| is email list cleaning legit | 2 | 2 | 3 | 1 | 8 |
| how to track open rates accurately | 1 | 2 | 2 | 1 | 6 |
| best email marketing software | 1 | 2 | 1 | 2 | 6 |
What you should see: a clear split. Phrases scoring 9 or more go straight to Step 5. Phrases at 7 or 8 stay as supporting variants. Anything at 6 or below gets cut unless it's already showing impressions in your search console. Look at the last row. It's the phrase with the biggest volume figure on the whole sheet, and it scores as low as a vague open-rate question.
The reason the scoring leans this hard on agency fit is the way recurring pay works. SendTo's affiliate programme pays you 20% of the partner subscription each referred agency pays us, every month it keeps paying. It's never a share of what that agency's own clients pay the agency. So an agency that stays a year pays you twelve times from one referral. A casual reader who never signs up pays you nothing, however many of them arrive. That's why one buyer beats a thousand skimmers, and why the agency column does more work than the volume figure ever will. For what the subscription costs an agency, send readers to the pricing page rather than quoting a figure that'll date.
The mistake here is adding a fifth column for search volume and letting it outvote the other four. Volume already did its job in Step 3, as a rough hint. Bring it back now and the broad terms creep back to the top.
Step 5: Group the survivors into clusters and give each cluster one page
Sort your sheet by problem, then merge any phrases a single answer would satisfy. "Why are client emails going to spam", "marketing emails going to spam after changing smtp" and "why do new domains go to spam" all get answered by one good article on sending reputation and warmup. That's one cluster. Pick the highest-scoring phrase as its primary query and list the rest underneath as variants to cover in subheadings.
Then give each cluster a page type based on its intent stage. Problem-aware clusters become how-to articles. Solution-aware clusters become comparisons that set platform approaches side by side. Product-aware clusters become reviews or case studies showing an agency running clients on the platform.
What you should see: 12 to 25 clusters from your 30 to 60 scored phrases. Not 200 pages. A dozen strong pages that each rank for twenty variants will out-earn a hundred thin ones fighting each other.
The mistake: one page per phrase. Write separate articles for "is email list cleaning legit" and "should I clean a client email list before sending" and they'll compete for the same results page. Search engines pick one, often the weaker one, and both stall on page two.
Match the call to action to the stage of the cluster
A problem-aware reader came to fix a spam issue today. Put an affiliate link in your second paragraph and you'll look like you skipped the answer to sell them something, and they'll leave. These pages earn an email signup, or a soft mention of the platform once the fix is explained, such as a line noting that warmup and per-domain sending speeds are handled automatically on SendTo. The 90-day attribution window means the click still counts if they come back weeks later.
A product-aware reader is different. They've already got a shortlist and want to know if this one fits. Make them hunt for the link and you lose them to a page that didn't. On reviews and case studies, put the affiliate link high, near the verdict, and again after the details. Same link, opposite placement, because the reader's question is different.
Step 6: Publish on the calendar agency buyers search on
Public interest in email marketing peaks in January, November and December and hits its low in June. Agencies plan client campaigns for the festive season in autumn and set up new tools for the new year. It's late July 2025 now, so you're in the right window. Pages published in August and September have two to three months to get indexed and climb before November.
| Week of | Cluster | Page type | Buyer-fit score |
|---|---|---|---|
| 4 August | Client emails going to spam | How-to | 11 |
| 11 August | Reselling email marketing to clients | Review | 11 |
| 25 August | Secondary domains for client sending | How-to | 10 |
| 8 September | White label platforms for agencies | Comparison | 10 |
| 22 September | Is list cleaning legit | How-to | 8 |
| 6 October | Retainer to recurring revenue | Case study | 9 |
What you should see: a dated calendar with your highest buyer-fit clusters first. They need the longest run-up and they earn the most if they rank. Lower scorers fill the later slots.
The mistake: writing seasonal content in November, when the buyers are already searching. A page published mid-November is still getting indexed when the peak ends. You'd be writing for next year without knowing it.
Check it worked by watching impressions, clicks to the offer and referred agencies, in that order
Wait six to eight weeks after a page goes live, then check three numbers in order. First, search console impressions for each cluster's primary query and its variants. Second, outbound clicks on your affiliate links from that page. Third, referred agency sign-ups in your affiliate dashboard. Each number only means something once the one before it is moving.
Expect the revenue figure to lag well behind the traffic figure. SendTo attribution lasts 90 days from the first click, and first touch wins. An agency owner who clicks in September, trials something else, then signs up in November is still credited to you. Commission is paid monthly by PayPal, so the money from that sign-up shows up a month or more after the traffic that caused it. Judge a cluster on revenue too early and you'll kill pages that were working. Our guide to recurring commissions covers how to plan a content strategy around that delay.
About the cap. Narrow queries do limit your traffic, and that's the trade you made in the intro. They don't cap earnings the same way. Every agency that stays adds another monthly payment on top of the ones already coming in, so income from this method stacks up over time rather than resetting each month.
If it did not work, find which of three stages broke
No impressions after eight weeks means the phrasing is wrong or the competition is stronger than you read it. Go back to Step 2 and check your primary query against autocomplete and your own search console data. Then repeat Step 3 for that results page. If a fresh, strong guide has appeared since you looked, move on to the next cluster.
Impressions but no clicks means your title doesn't match the problem. A searcher who typed "why are client emails going to spam" scrolls past "The Complete Deliverability Handbook". Rewrite the title in their words, so it repeats the question they asked, and check again in three weeks.
Clicks but no agencies means you attracted the wrong reader. Shop owners and in-house marketers clicked through, found a platform built for agencies to resell, and left. Tighten the agency modifiers in your headings and opening, put "client" back into the phrasing, and rescore the cluster in Step 4 with a stricter eye on agency fit.
Start with your top two clusters this week and get them live before the end of August. Then join the programme so the clicks they earn get credited to you from day one.
Earn 20% of the partner subscription every month each agency you refer keeps paying, with 90-day first-click attribution and monthly PayPal payouts.
Join the SendTo affiliate programme



