The verdict in three sentences
A successful automation spec rests on three pillars: a quantified map of candidate processes, clear eligibility criteria (volume, rule stability, criticality) and a pilot lot capped at 8,000-15,000 EUR. Without objective scoring, you automate the wrong process and pay twice. This document must let three vendors return genuinely comparable quotes on the same scope.
Map and score candidate processes
Before any quote, list 5 to 10 processes and score them. A process deserves automation when volume is high, rules are stable and the interface is API-poor. Here is a 2026 scoring grid (rated out of 5 per criterion).
| Criterion | Weight | Key question | Favorable threshold |
|---|---|---|---|
| Monthly volume | 30 % | How many items/month? | > 200 |
| Rule stability | 25 % | Does it change often? | < 2 times/yr |
| Current human cost | 20 % | Hours/month spent? | > 15 h |
| Error rate | 15 % | Costly errors? | > 3 % |
| Technical feasibility | 10 % | API available? | Yes = bonus |
A process scoring above 3.5/5 is a priority. Below 2.5, postpone it: automation will cost more than the gain.
Choose the technology and frame the pilot
The spec must require a pilot lot before any wide rollout. It sets scope, SLA and target error rate. Compare the three technical paths against your context.
| Path | When to choose it | Pilot cost | Target SLA |
|---|---|---|---|
| API / native integration | Modern software with API | 8,000 - 12,000 EUR | error < 1 % |
| iPaaS (Make, Workato) | Multi-tool, scalable | 10,000 - 15,000 EUR | error < 2 % |
| RPA (UI robot) | Legacy software, no API | 12,000 - 18,000 EUR | error < 2 % |
Require in the spec: 6 to 12 week phasing, target error rate < 2 %, documentation, knowledge transfer and a reversibility clause. The vendor evaluation grid should weight sector experience (25 %), method (25 %), price (30 %) and proposed maintenance (20 %).
Mini case study
Marc, operations lead at a mid-market distribution firm in Nantes, hesitates between automating supplier data entry (600 invoices/month) or sales reporting (monthly). Scoring settles it: supplier entry scores 4.2/5 (high volume, stable rules, 22 h/month), reporting 2.1/5 (monthly, shifting rules). Marc launches an iPaaS pilot at 12,000 EUR on invoices only. Result in 9 weeks: error rate down from 4 % to 1.3 %, 22 h/month freed. The rigorous spec saved him from investing 20,000 EUR on the wrong process.
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
FAQ
Which processes should I automate first?
Those exceeding 200 items/month with stable rules and over 15 hours of monthly human work. The weighted scoring grid surfaces them objectively rather than by intuition.
Do I need a pilot lot before full rollout?
Yes, always. A pilot at 8,000-15,000 EUR over 6-12 weeks validates feasibility and real ROI before committing the full budget.
RPA, iPaaS or API: what goes in the spec?
Don't lock the tech: describe the process and let the vendor propose. Native API if the software exposes one, RPA for legacy tools, iPaaS for a scalable multi-tool environment.
How do I get comparable quotes?
Impose the same pilot scope, the same SLA (error rate < 2 %) and the same phasing on all vendors. A weighted evaluation grid (price 30 %, method 25 %, sector 25 %, maintenance 20 %) objectifies the choice.
What error rate should I target?
Under 2 % for an operational process, under 1 % for a critical financial flow. The spec must require error-recovery handling and alerts.
Let's scope your project. Send us your list of candidate processes and their volumes; we return a scoring and a priced pilot scope. Detailed quote within 48 h. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
