2026-04-24
Freelance Instructional Designer Toolkit 2026: Tools & Rates
A working freelance ID's toolkit for 2026 — the tools I actually use, real UK rate ranges, and how AI is changing freelance instructional design.
The Freelance Instructional Designer's Toolkit: Tools, Rates, and How AI Is Changing the Work
By Paul Thomas, L&D consultant and founder of The Human Co.
Freelance instructional design in 2026 is a different job than it was three years ago. The authoring tools you use matter less than they used to. The AI in your workflow matters more. The rates are under pressure. And the skill that increasingly separates the people winning client work from the people chasing it is the one the tools don't teach — knowing what actually good learning looks like and being able to produce it faster than anyone expected.
This is a working freelance ID's toolkit. The tools I actually use, the ranges clients actually pay, and what's changing in 2026.
What freelance ID work actually involves day-to-day
Before the toolkit — a clear frame of what the job is, because the answer shapes which tools matter.
A typical project runs roughly like this:
- Discovery. Client conversations, SME interviews, scope definition. 10–20% of project time.
- Content design. Learning outcomes, module structure, interaction design, scenario development. 20–30% of project time.
- Production. Building the course in an authoring tool, writing content, creating visuals, recording audio, configuring assessments. 40–60% of project time.
- Review and iteration. SME sign-off, client revisions, stakeholder feedback. 10–20% of project time.
- Export and handoff. SCORM testing, LMS upload, client handover. 5–10% of project time.
Production is typically the largest chunk. AI is compressing it fastest. The balance of the job is shifting from "building" to "deciding what to build and checking it's right" — which is good news for experienced IDs and less good news for people whose value proposition was being fast in Storyline.
Authoring tools — what to own and what to rent
Client projects determine which authoring tool you use more often than preference does. Most freelance IDs end up with two or three:
Articulate 360 (Rise and Storyline). The market default. If your clients are corporate L&D teams, they'll expect Articulate output more often than not. Subscription is roughly £1,149/year for the Personal plan — significant for a solo operator but sometimes unavoidable for client access.
iSpring Suite. Cheaper alternative at around £610/year. Useful if your clients are PowerPoint-heavy or if you primarily build quiz-driven compliance training.
Co.llab (launching June 2026). New category — AI-first desktop tool. Does the instructional design work itself, not just text generation. £199 founder edition (first 50 purchases) / £299 standard. One-time payment. Full comparison of authoring tools here.
Canva. For supporting visual work, quick course graphics, social media assets for course promotion. Not a primary authoring tool but genuinely useful in the toolkit.
The honest advice for freelancers in 2026: don't commit to Articulate unless client work requires it. The AI-first tools are mature enough to handle most projects, and the cost savings compound fast over a career.
AI tools — where the real shift is happening
This is the part of the toolkit that's changed most in 2026. Three tools worth having:
Claude (Cowork mode). General-purpose AI work — drafting content, restructuring briefs, generating scenario outlines, reviewing SME interview transcripts, writing assessment questions. The Cowork integration with existing tools (email, docs, notes) makes it substantially more useful than a standalone chat interface.
Claude Code (or equivalent). For IDs building custom tools, scripts, or experimental workflows. Most freelance IDs won't need this; those who do will find it transformative. I use it extensively for building Co.llab itself.
Perplexity. For research that needs to be sourced and verifiable. Not a replacement for Google Scholar for academic work, but excellent for the kind of quick industry research that used to eat hours of billable time.
What AI actually does for a freelance ID workflow:
- Drafts scenarios from case study notes (saves hours)
- Generates assessment question banks from source material (saves hours)
- Produces transcripts from SME interviews (saves hours)
- Writes initial learning objectives from a rough brief (saves time, always needs editing)
- Converts policy documents into learner-facing language (saves significant time)
What AI doesn't do reliably (yet):
- Produce final-quality assessment questions without editing
- Handle sensitive or regulated content without oversight
- Replace SME review for accuracy
- Know what your specific client wants without being told
The freelance IDs who are making the transition successfully are the ones treating AI as a production accelerator — not a replacement for the design work. Our walkthrough on how to actually build a course with AI covers the process.
Project management and client communication
This is the underrated part of the toolkit. Great instructional design delivered badly to a client usually loses the repeat work.
Obsidian. Personal knowledge management. Notes, project archives, SME interview transcripts, reusable frameworks. Plain-text markdown, works offline, doesn't hold your knowledge hostage to a subscription. This is where experienced IDs build the asset library that speeds up every subsequent project.
Loom. For review walkthroughs. When a client needs to review a course, a five-minute Loom walkthrough of the draft gets better feedback than a written brief asking for comments. It also demonstrates the work in a way static documents can't.
OBS. For screen recordings — when you need to record a demo, a walkthrough, or a training video yourself. Free, reliable, works for anything video-based.
Krisp. For audio cleanup on client calls. Cheap, effective, removes the need for a dedicated podcast mic for most calls. Makes you sound professional on recordings you didn't have time to set up properly for.
Asana or Notion. Shared project tracking with clients. Pick one based on what your clients already use — fighting a client's preferred tool costs more than adapting to it.
Video calls (Teams / Google Meet / Zoom). Client preference. Don't have a strong view here.
SME management — the make-or-break skill
Most freelance projects live or die on SME access. Tools help, but the skill matters more.
The pre-interview questionnaire. Before any SME call, send a short document that asks: what's your role, what does a typical day look like, what do learners get wrong most often, what do you wish they understood better. This prepares the SME and means the live interview starts where the useful conversation actually is.
The key question. "Walk me through what happens when someone gets this wrong." This draws out specifics, edge cases, and the tacit knowledge the SME has but hasn't written down. It outperforms "what should this course cover?" every time.
Recording and transcription. Record every SME interview (with permission). Transcribe it. The transcript becomes source material for AI-assisted course building and a reference you'll come back to during the build. Good tools: Otter.ai, Fireflies, or native recording in your call platform.
Keeping SMEs engaged. Short, frequent check-ins beat long review sessions. A 15-minute call every week beats a two-hour review at the end of the project. SMEs lose context fast; frequent contact keeps the content accurate.
Visual design and media
Most freelance ID work doesn't need a full design suite, but some does.
Canva. For 90% of visual needs. Course hero images, module graphics, supporting infographics, social media assets. Low learning curve, good templates.
Figma. If the client needs custom UI design, branded visual systems, or complex layouts. Much more powerful than Canva but requires actual design skill. Worth learning if you want to move upmarket in the types of projects you take.
Affinity (Designer, Photo, Publisher). The serious alternative to Adobe Creative Cloud, one-time purchase. Relevant if you produce print materials, custom illustrations, or complex editorial layouts alongside elearning. Most freelance IDs won't need it; those who do will find it excellent.
Stock imagery. Unsplash, Pexels for free work. Getty or iStock for client work where licensing matters. Don't cut corners on image licensing — clients who get caught with improperly licensed images remember who gave them those images.
Rates — what freelance IDs actually charge in 2026
Time for honest numbers. This is where most guides get vague. Here's what UK freelance IDs are actually charging in 2026, based on direct experience:
Hourly rates: £30–£50 per hour is the realistic range for UK freelance ID work delivered through platforms (Upwork, Fiverr) or direct-to-client projects at the mid-market level. Experienced IDs at the upper end; newer freelancers at the lower end. Specialist work in regulated industries can go higher — £80–£150/hour for compliance specialists with niche expertise.
Project pricing — the delivery ratio matters more than the hourly rate. Freelance IDs produce finished content at roughly 20–50 hours of work per 1 hour of finished seat time. A 30-minute module therefore represents 10–25 hours of delivery work. At £30–£50/hour, that's:
| Finished course length | Hours of work (mid-range 35 hrs/hr) | Raw project fee at £40/hr |
|---|---|---|
| 15-minute microlearning | ~9 hours | ~£360 |
| 30-minute module | ~17.5 hours | ~£700 |
| 60-minute module | ~35 hours | ~£1,400 |
| Half-day course (~3 hrs content) | ~105 hours | ~£4,200 |
What reduces the take-home:
- Platform fees. Upwork and Fiverr typically take 10–15% of each project. Direct-to-client work avoids this but requires client acquisition effort.
- Corporation tax. 19% in the UK (small business rate) on company profits if you operate through a limited company.
- Income tax and National Insurance. Further reduction once profits are drawn as salary or dividends.
The realistic take-home from a £1,400 raw project fee — after 12.5% platform fee, 19% corp tax, and basic-rate income tax on drawn income — is in the region of £800–£900. That's the number to budget around for business planning.
How AI changes these numbers. The production time per finished hour is compressing. A delivery ratio that was 40:1 two years ago might be 25:1 now for the same quality output, if you're using AI properly. The question most freelance IDs are wrestling with: does this mean clients pay less for the same course, or does it mean you can take on more clients? The answer is "both, depending on how you position yourself."
What's changing — and what to do about it
Four trends shaping freelance ID work in 2026:
1. AI is compressing timelines. Clients who previously expected 4–6 weeks for a 30-minute module are starting to expect 1–2 weeks. The IDs winning work are the ones delivering on the new timeline without sacrificing quality. The IDs losing work are pricing on old time estimates.
2. The conversation with clients is shifting from "how many hours?" to "what outcome?" Outcome-based pricing — charging for the completed module, the learning outcome achieved, or the business impact delivered — rewards IDs who are fast and good. Hourly pricing rewards IDs who are slow. This is going to accelerate.
3. Template-locked output is losing value. When an AI tool can produce a course, clients paying premium rates want courses that feel bespoke — reflecting their specific organisation, their specific learners, their specific situations. Rise courses that look like every other Rise course are commodity work. Distinctive output is premium work.
4. Specialisation is winning over generalism. The broad "I can build any kind of course" positioning is increasingly competed down to platform rates. The "I build accessibility-compliant training for healthcare clients" or "I specialise in onboarding for engineering teams" positioning holds rates better and wins repeat work.
What to do about it:
- Pick a niche and go deep on it
- Learn AI workflows properly — not as a novelty, as a core skill
- Move from hourly to project pricing where possible
- Invest in distinctive output (visual design, scenario quality, interaction depth) rather than template-speed
- Build a recognisable portfolio that clients can find and reference
The tools I actually use
For full transparency — the current stack I use for freelance ID and consultancy work in 2026:
- Claude (Cowork mode) — everyday AI work, content drafting, research, interview prep
- Claude Code — for building Co.llab and custom workflow tools
- Perplexity — quick-source research
- Obsidian — personal knowledge base, project archive, reusable frameworks
- Rise and Storyline — for client projects that require Articulate output
- Co.llab — for projects where I control the tooling decision (biased — I built it)
- Loom — client review walkthroughs
- OBS — screen recording for demos and training videos
- Krisp — call audio cleanup
- Canva — quick visual work and course graphics
- Figma — when client work needs custom design
- Affinity — for print and detailed design work
The list is intentionally curated. Plenty of tools do similar jobs. These are what I've settled on after trying alternatives. Yours will be different. The right stack is the one that works for the projects you actually do — not the one every LinkedIn post says you should be using.
The honest bottom line
Freelance instructional design in 2026 is still a good job for people who are good at it. The rates are under pressure at the bottom and the top is increasingly specialised. AI is making the production burden lighter, which means the design thinking matters more and the tool speed matters less.
The IDs doing well right now are the ones who've always been good — experienced, specific about what they build, clear about the outcomes they deliver. AI is an amplifier of existing skill, not a shortcut for people who don't have it.
If you're a freelance ID considering what tools to invest in, what rates to charge, what direction to move in — the answer isn't dramatically different from five years ago. Be good at what you do, charge what it's worth, don't commodity-price yourself. The tools matter less than the judgment.
Our deeper dive on what's replacing Articulate covers tool choice in more detail.
Try Co.llab when it launches
Co.llab is in closed beta, launching 18 June 2026. Built specifically for freelance IDs and small teams who want AI to handle the production burden. The first 50 purchases at launch get founder pricing — £199 for lifetime ownership. Standard pricing after that is £299, still one-time payment, no subscription.
Join the beta now and get 130 free AI prompts for instructional designers — a working toolkit you can use today, regardless of whether you end up buying Co.llab at launch.